The maximum gas temperature below the ceiling is an important parameter for tunnel safety. The present study analyzed the characteristics of the maximum excess ceiling gas temperature driven by double fire sources in a naturally ventilated tunnel. A series of small-scale tunnel fire experiments were carried out with different fire separation distances and heat release rates. Theoretical analysis based on the equivalent virtual origin was also performed. The results showed that there exists only one peak gas temperature when the two fire plumes are merged before reaching the ceiling, while two peak gas temperatures can be observed when the two fire plumes are completely separated. The maximum excess gas temperature below the tunnel ceiling gradually decreases with an increasing fire separation distance in the plume merging region (S < Scp). When the fire separation distance increases further (S > Scp), the effect of the fire separation distance on the maximum gas temperature below the ceiling is very limited. Furthermore, a model using an equivalent fire source was proposed to predict the maximum excess gas temperature below the ceiling, considering different plume merging states. The present study contributes to the understanding of the maximum excess gas temperature characteristics of the smoke flow driven by double fires with an equal heat release rate in naturally ventilated tunnels. 

Smoke control in a longitudinally ventilated tunnel with various blockage conditions was investigated experimentally. A total of 28 tests were conducted with a focus on single blockage with a short distance from the fire source, although continuous blockage and semicontinuous blockage were also discussed. Both gas and pool fires were used. The aim was to understand the influence of upstream blockage on critical velocity and babcklayering length. The results confirm that blockage ratio is a critical parameter when determining the critical velocity and backlayering length. The longitudinal location of the blockage in relation to the fire source also influences the values of critical velocity and backlayering length. The experiments presented are in scale 1 to 3.3, representing a medium sized tunnel. The focus was on free flow conditions and blockage ratios of regular sizes. For the various tested scenarios with single blockage, the reduction ratio of critical velocity appears to be slightly less than the blockage ratio. However, when the blockage is attached to the upstream side of the fire source, the reduction ratio of critical velocity approximately equals the blockage ratio. 

This report presents both small scale laboratory tests and tunnel fire tests carried out in a FORMAS project. Four series of small scale laboratory tests were conducted to obtain the material properties, burning properties, water spray distributions, and spray droplet size distributions. The main efforts were, by adopting the Froude scaling, seven series of tests conducted in a about 50 m long container tunnel with a scale of 1 to 3. This report presents results on the influence of low pressure, medium pressure and high pressure water-based fixed fire fighting systems (FFFSs) on fire development, fire spread to adjacent vehicles, structural protection, tenability, smoke control, spray deflection and spray resistances. The focus is to compare the performance of three default FFFSs and to evaluate the efficiency of each of the FFFS. The results show that the default low pressure FFFS performs well in term of suppressing the fire development, preventing the fire spread to nearby vehicles, providing tenable conditions for evacuation and rescue service, protecting tunnel structure and easing the problem with spray deflection due to tunnel ventilation. The default high pressure FFFS is usually on the opposite side while the default medium pressure FFFS usually lie in between.

This work aims to validate numerical modelling of water sprays against experiments carried out in this project. The focus has been on the influence of ventilation on water distributions on the tunnel floor, the influence of water sprays on control of smoke flow, and the modelling of pool fires and crib fires. Besides, full scale tunnel fires with FFFS in tunnels using longitudinal ventilation and point extraction ventilation systems are simulated and compared, as well recommendations for numerical modelling of such scenarios.

Low pressure, medium pressure and high pressure water-based fire suppression systems were tested in a medium scale tunnel (scale 1:3). The primary objective was to investigate which of these systems are most effective in the suppression or control of different types of tunnel fires. The default low, medium and high pressure systems refer to full scale water flow rates of 10 mm/min, 6.8 mm/min and 3.7 mm/min, respectively. Some other water densities were also tested to investigate the effects, as well as different ventilation velocities and activation criteria. Several series of fire tests were conducted for different fire scenarios. The fire scenarios considered included idle wood pallet fires, loosely packed wood crib fires, loosely packed wood and plastic crib fires, and pool fires, with or without a top cover on the fuel load. Comparisons of the three default systems based on the three parameters: heat release rate, energy released and possibility of fire spread, show that the performance of the default low pressure system is usually the most effective based on the parameters studied. The default high pressure system usually yields results less effective in comparison to the default low pressure system. The performance of the default medium pressure system usually lies in between them. The high pressure system behaves very differently in comparison to the others, in terms of tunnel ventilation velocity, water density, operating pressure, and the presence of the top cover. © 2024 The Authors

Sample tests were conducted to obtain thermal and kinematic parameters for wood and high-density polyethylene (HDPE) that were used in a series of intermediate scale tunnel fire tests with and without water-based fire suppressions systems. The thermal properties were measured using Transient Plane Source (TPS) and Transient Line Source (TLS). The pyrolysis kinetics parameters were tested based on Thermal Gravimetric Analysis (TGA) and Differential Scanning Calorimetry (DSC). Different methods were used to obtain pyrolysis kinetics parameters. Different oxygen concentrations exposed to samples were tested and the results showed its significant influence in the charring process.

Society is changing ever faster, and tunnels are complex systems where performance is affected by many different stakeholders. These conditions suggest that safety management needs to be proactive and based on a systems perspective that acknowledges socio-technical theories. Although systems thinking principles are foundational in overarching European regulations and goals, system principles generally don’t affect tunnel fire safety design principles or engineering practice. In the countries investigated in this study, tunnel fire safety management (TFSM) builds on experience-based and risk management-based principles that are optimized independently system by system. This is usually done with limited consideration of how these systems are interconnected and affect the overall tunnel system. The purpose of this paper is to investigate how systems thinking could support existing engineering practice. The work presented in this article is the outcome of a collaboration between fire safety researchers and practitioners from five countries and three continents. Through three workshops, current TFSM principles have been compiled and discussed. It is suggested that tunnel safety regulations be redesigned to strengthen the ability of engineers to work in design teams using systems thinking principles.

Vehicles that are powered by gaseous fuel, e.g., compressed natural gas (CNG) or hydrogen (H2), may, in the event of fire, result in a jet flame from a thermally activated pressure relief device (TPRD), or a pressure vessel explosion. There have been a few incidents for CNG vehicles where the TPRD was unsuccessful to prevent a pressure vessel explosion in the event of fire, both nationally in Sweden and internationally. If the pressure vessel explosion would occur inside an enclosure such as a road tunnel, the resulting consequences are even more problematic. In 2019 the authors investigated the fire safety of CNG cylinders exposed to localized fires. One purpose of the tests conducted in 2021 reported in this paper is to investigate whether extinguishment with water, e.g., from a tunnel deluge system, may compromise the safety of vehicle gas cylinders in the event of fire. Steel cylinders handles the situation with localizde fire and/or cooling with water well. Composite tanks can rupture if the fire exposure degrades the composite material strength, and the TPRD is not sufficiently heated to activate, e.g., if the fire is localized or if the TPRD is being cooled by water, which prevents its activation.

This report describes the fire dynamic conditions in natural ventilated tunnels. A summary of the today’s knowledge about the dynamic fire conditions in nonventilated road or railway tunnel is given. The concept of fire dynamic conditions includes the variation of different parameters in length and height at different ventilation conditions. This primarily applies to parameters such as fire development, heat and smoke gas dispersion, gas temperatures, heat radiation towards objects and surrounding wall construction, flame lengths and sight length and toxic conditions in the smoke gases.

This paper systematically studied the coupling effect of ambient pressure and tunnel slope on temperature distribution and smoke propagation in full-scale tunnel fires under natural ventilation by FDS. The downstream length (longitudinal length from fire source center to tunnel downstream exit) was also considered. The concept of “height difference of stack effect” was put forward when analyzing the mutual effect of tunnel slope and downstream length on smoke movement. The results show that the maximum smoke temperature beneath the ceiling decreases with the increasing ambient pressure or tunnel slope. The longitudinal smoke temperature decays faster with the decreasing ambient pressure or slope in inclined tunnel. The induced inlet airflow velocity increases with the increasing height difference of stack effect, while decreases with the increasing ambient pressure. And the smoke backlayering length decreases with the increasing height difference of stack effect. Taking heat release rate (HRR), ambient pressure, tunnel slope and downstream length into account, the prediction models of dimensionless induced inlet airflow velocity and smoke backlayering length in inclined tunnel fires at high altitude were developed, which agree well with our and others’ results. The outcomes of current study are great meaningful to fire detection and smoke control in inclined tunnel fires at high altitude. © 2023, The Author(s)

This study was performed by RISE Research Institutes of Sweden on behalf of Volvo Trucks. RISE Research Institutes of Sweden was requested to conduct a study regarding the differences between fires in conventional internal combustion engine (ICE) trucks and electric trucks. A set of guiding questions (see section Aim) were given by Volvo Trucks and in this report these questions have been answered. The questions have been answered by performing literature searches and through previous knowledge of RISE. However, for some questions, due to scarcity of data on electric truck fires, knowledge regarding electric passenger cars has been used. In addition, contact has been made with fire and rescue services around the world (Australia, UK, USA, Sweden and Finland) to collect their views on management of fires in electric vehicles (EVs). The main conclusions are: • Data on electric truck fires are scarce due to the low number of vehicles as well as the low number of fire incidents. Available data show that battery electric passenger vehicle fires are less common than ICE vehicle fires, but that the risks are different. The main differences are that battery fires tends to be harder to extinguish than fires in ICE vehicles and that there is a risk of accumulation of flammable gases, especially in enclosed spaces, upon thermal runaway. • Lithium iron phosphate (LFP) type cells, in comparison with nickel-based type cells (such as lithium nickel manganese cobalt oxide (NMC) and lithium nickel cobalt aluminium oxide (NCA)), have a higher thermal runaway onset temperature, a slower temperature increase rate, a lower maximum temperature as well as a lower gas production in total amount. However, the specific total gas production (L Ah-1) can sometimes be higher for LFP-type cells and depends on the state of charge and on the amount of electrolyte in the cell. However, the safety of a battery pack in a vehicle is determined by several factors such as preventive measures aimed at reducing the occurrence of fires (safe design). For example, by early detection and pro-active mitigation using the battery management system and thermal management system and by limiting the thermal propagation in the battery pack, reducing the extent of damage. • Fires in enclosed spaces, such as in underground parking garages and tunnels, generally imply a higher risk for firefighters due to the trapped smoke, decreased visibility and longer access routes than in open structures. Risk reduction measures for battery fires should focus on early detection of harmful events, reducing thermal propagation in the battery pack and on limiting the extent of fire spread. The severity of the consequences of vehicle fires (no matter if is an EV or an ICEV) in enclosed spaces could be reduced using suppression systems, such as a water sprinkler system, to hinder fire spread between vehicles.

Lately, concerns regarding fires in electric vehicles in enclosed spaces such as in road tunnels and parking garages have been raised and there are indications that parking of electric vehicles may be prohibited in some spaces. For the success of electromobility and the transition from fossil to renewable fuels, it is important to understand the risks and consequences of fires in electric vehicles and to provide technical solutions if necessary, so as not to hinder the widespread adoption of electric vehicles.

In this work, a literature review on fires in vehicles has been conducted. The focus was on fires in enclosed spaces involving electric vehicles. A comprehensive risk assessment of electric vehicle fires was performed using systematic hazard identification. In addition, a workshop with representatives from three Swedish fire and rescue services was carried out to evaluate the emergency rescue sheets/response guides.

The main conclusions are; That statistics regarding vehicle fires need to be improved, as of today the root causes of fires are missing in the data, which could potentially result in non-fact based regulations; The data studied in this work does not imply that fires in electric vehicles are more common than fires in internal combustion engine vehicles; Fires in electric vehicles and internal combustion engine vehicles are similar in regards to the fire intensity and peak heat release rates. 

The most effective risk reductions measures on vehicle level, to decrease the number of fires in EVs, could not be defined based on that relevant data on the root causes of fires in EVs are currently not publicly accessible. The most effective risk reduction measures, to limit fire spread, on infrastructure level were the use of fire sprinkler systems, fire detection systems (early detection) and increased distance between parked vehicles.

The characteristics of fire spread among multiple vehicles in tunnels using longitudinal ventilation were investigated by analyzing the experimental data from a series of fire tests in a 1:15 scale tunnel. Further, a simple theoretical model for gas temperature in a tunnel with multiple fire sources was proposed and used in analysis of the experimental data. The results showed that, for objects (wood piles) placed at a same separating distance downstream of the fire, the fire spread occurred faster and faster along the tunnel. Validation of the simplified temperature model for multiple fire sources was made against both model and full-scale tunnel fire tests. The model was further used to predict the critical conditions for fire spread to the second and third objects. Comparisons with the test data showed that average excess temperature of 465 K (or an equivalent incident heat flux of 18.7 kW/m2) could be used as the criterion for fire spread, and this was verified further by other model-scale tests and full-scale tests. The results showed that the critical fire spread distance monotonously increases with the heat release rate, and decreases with the tunnel perimeter. For multiple fire sources with equivalent heat release rates, as the separation distance between the first two fire sources increases, the critical fire spread distance from the second fire source to the third fire source decreases, but the total fire spread distance from the first fire source to the third one increases. If the total heat release rate at the site of a downstream fire source is greater than that at the former fire source, the critical fire spread distance becomes longer.

The paper focuses on the flow structures and mass flow rates of thermally driven smoke flows induced by fires in long transportation tunnels under natural ventilation. The important influencing factors including heat release rate (HRR), tunnel width and height, are taken into consideration. The mechanism of the smoke flow movement is explored. The results show that for a fire in a long naturally ventilated transportation tunnel, there exists a critical point which is dependent on HRR and tunnel geometry. This critical point is defined as the location where the smoke layer thickness and the outgoing mass flow rate increase towards it and decrease after it. Further, it is found that the critical point moves farther away from the fire source in a wider or higher tunnel, while it lies closer to the fire source for a higher HRR. A correlation is proposed to estimate the location of the critical point. The outgoing mass flow rates along the tunnel are calculated using the two-layer flow model and well-mixed flow model of thermally driven flows, and the results indicate that these models produce satisfactory predictions of the mass flow rates if the vertical temperature profile is known. © 2023 The Authors

This publication includes the Proceedings of the 10th International Symposium on Tunnel Safety and Security (ISTSS) held in Stavanger, Norway, April 26-28, 2023. The Proceedings include 45 papers and 16 posters. The papers were presented in 16 different sessions, i.e., Keynote sessions, Alternative Fuel Vehicle Safety, Risk Management & Explosion, Digitization, Explosion, Poster Corner, Ventilation 1&2, Fixed Fire Fighting Systems, Tenability and Evacuation, Emergency Management, Evacuation, Safety Management, Fire Dynamics and Resistance. Each day was opened by invited Keynote Speakers (in total five) addressing broad topics of pressing interest. The Keynote Speakers, selected as leaders in their field, consisted of Ove Njå (University of Stavanger, Norway), Vladimir Molkov (Ulster University, UK), Ulf Lundström (Swedish Transport Administration, Sweden), Mirjam Nelisse (TNO, The Netherlands), and Gunnar Jenssen (SINTEF, Norway). We are grateful that the keynote speakers were able to share their knowledge and expertise with the participants of the symposium.

RISE has previously studied alternative fuels, such as batteries and gaseous fuels including liquid and compressed hydrogen (GH2). Each fuel has its unique risks. Liquid hydrogen (LH2) is a cryogenic fluid and is thus stored in cooled liquid form, which entails specific risks. The purpose of this report is to, based on the current state of research, map the risks of hydrogen in underground facilities in relation to conventional fuels and investigate which technical measures can be taken to reduce the risks. Unlike diesel, hydrogen (and for instance methane or gasoline) has such a low flash point that an emission can be ignited at normal temperature by a small ignition source. Hydrogen is also very buoyant, with strong diffusion and dispersion characteristics, accordingly it accumulates at high points in a subsurface environment. Hydrogen requires very low energy to ignite at or near stoichiometric mixing with air at around 30%. The lower flammability limit is, compared to other flammable fuel/air mixtures high at around 4%, which means that many smaller releases in ventilated spaces will be too lean. Explosions would require a higher hydrogen concentration, above 8% or more. In subsurface environments, containment contributes to a higher increase in pressure, as well as an increased risk of explosion for both GH2 and LH2. The handling of hydrogen underground can therefore be seen as problematic. When it comes to hydrogen as a vehicle fuel, however, there are safety measures to achieve equivalent safety with conventional vehicles. For example, the shut-off valve (mandatory in regulation) on each tank that reduces the risk of leakage, and through the development of explosion-free composite tanks (not mandatory in regulation) in the event of fire that provide a less dangerous fire scenario than a diesel or gasoline tank in case of fire. When it finally comes to transporting hydrogen, pipelines are the long-term sustainable (and safe) alternative. Transport of compressed hydrogen gives a low amount of gas per trailer and entails relatively higher risks than CNG underground, for example in tunnels. The usage of liquid hydrogen, so far, has an impressive safety record, events like BLEVE or fireballs appear to be rare. The transport of liquid hydrogen provides a larger amount of hydrogen per trailer (than for compressed hydrogen) with a relatively lower risk than, for example, LNG in the open, but a slightly higher risk for explosion of accumulated gas compared to GH2 in enclosed spaces. The safety requirements for transport of compressed hydrogen are less stringent than for road vehicles, e.g., with regard to shut-off valves and melt-fuses and could be improved. Several risk mitigation measures for tunnels and other underground facilities have been identified.

Räddningsinsatser i undermarksanläggningar kan vara komplexa och kräva såväl kunskap som stora resurser. Spårtunnlar är dock generellt sett lättare att orientera sig i då tågtrafiken kräver relativt raka tunnlar utan stora lutningar. Om tunneln inte försetts med brännbar kondensisolering gör bergets eller betongens ytskikt och egenskaper att risken för övertändning är försumbar utom exempelvis inuti ett fordon eller i slutna teknikutrymmen. Detta gör att behovet av släckvatten för självskydd längre bort från branden är liten och att behovet av snabb och effektiv förflyttning vid rekognosering och avsökning motiverar förflyttning utan samtidig slanguppbyggnad.

Inom Ostlänken-projektet har ett insatskoncept tagits fram, där huvudsaklig förflyttning sker i en brandtekniskt avskild servicetunnel och förflyttning från närliggande tvärtunnlar till en plats på säkert avstånd från branden sker i det drabbade tunnelröret. Förflyttningen från tvärtunneln mot branden föreslås ske utan samtidig slanguppbyggnad i de fall detta kan ske utan risk för den enskilda rökdykaren. Insatskonceptet för Ostlänken bygger också på att räddningstjänsten ska kunna dra nytta av de tekniska installationer som projekterats som stöd för utrymning, exempelvis belysning i handledare ovan gångbana.

Projektet har genomförts i syfte att studera förflyttnings- och avsökningshastigheter i spårtunnelmiljö utan samtidig slangdragning, samt för assisterad förflyttning med bår eller med spårgående räddningstralla.  Målsättningen var att påbörja insamlandet av ett vetenskapligt underlag som kan användas i verifieringen av insatskoncept liknande Ostlänken och vid andra tillfällen när bedömning av räddningstjänstens förmåga eller metoder behöver genomföras.

Rökdykarnas hastighet vid avsökning var snabbare än vad som visats i andra försök med samtidig slanguppbyggnad. Avsökningshastigheten var ca 0,5–0,6 m/s gående på gångbana och ca 0,6–2,0 m/s när avsökningen skedde från räddningstrallan. Ljusförhållandena var dock betydligt bättre än under tidigare försök i och med den kontinuerliga belysningen på handledaren. När avsökningen skedde från trallan var det endast det rökdykarpar som hade den lägsta förflyttningshastigheten som lyckades hitta den nödställde på gångbanan bakom tåget.

Tillgången till värmekamera angavs i all utvärdering som en förutsättning för säker förflyttning i den rökfyllda miljön. Användandet av dubbla värmekameror skapade en trygghet i att båda rökdykarna enskilt kunde avsöka och orientera sig i tunneln. En skarp rekommendation som stöds av detta, samt tidigare genomförda, försök är att samtliga rökdykare som rör sig i den rökfyllda tunnelmiljön förses med egen värmekamera.

Även om samtliga medverkande rökdykare i enkäter och vid de gemensamma genomgångarna uppgav att de kände sig trygga att genomföra förflyttning och insats utan samtidig slanguppbyggnad, fanns ett mindre antal rökdykare som angav att de vid en skarp insats gärna sett tillgång till säkert vatten, trots att de inte kunde redogöra för vilken nytta de skulle ha av brandslangen i liknande miljöer längre bort från branden. Detta visar på ett behov av ytterligare utbildning, övning och utveckling för att räddningstjänstens personal ska känna sig trygga med att genomföra insatser längre bort från brandplatsen utan tillgång till säkert vatten.

Microencapsulated phase-change-materials (PCMs) incorporated in cementitious grout can be used as a source of energy in an underground thermal energy storage system. Differential scanning calorimetry (DSC) is a widely used technique to measure the latent heat or specific heat of PCM-embedded cementitious materials. However, using milligram sample sizes (as required by DSC) of a cementitious material fails to represent the actual scale of cementitious components. This is the reason why, in the present paper, non-isothermal heat conduction microcalorimetry (MC) was evaluated as a tool for determining the thermal properties of PCM-embedded grout as well as pure PCM (three PCMs were used). An MC experimental protocol (using both single and 5–6 temperature cycles) was developed and used to measure latent heat and melting and crystallization temperatures, which were in good agreement with those reported for pure PCMs by the producers. In addition, the specific heats of the PCM-containing grout also agreed with measurements using the hot disk technique. Overall, the results show that the MC technique can be used as a potential standard method in determining thermal processes in complex systems, such as in PCM-embedded cementitious systems, where a large sample size is needed to represent the material.

Supportive report with suggestions at revision of GRAMKO mining guidelines regarding fire suppression systems in mining vehicles

A need has been identified to revise swedish minings guidelines BRANDSKYDD I GRUV- OCH BERGANLÄGGNINGAR Bilaga 1-FORDON BRANDSKYDDSKONTROLL dated 2016 as SBF 127 has been revised in 2021. SBF 127:17 has been revised on crucial points regarding the installation of automatic extinguishing systems in engine compartments. The transition rules for extinguishing systems according to previous editions of SBF 127 expire on 31-12-2023. The work has also included a review of how other needed revisions in the same appendix can be handled. A proposal for how extinguishing systems according to SBF 127:17 can be applied for fire protection of vehicles with Li-ion batteries that are used in underground facilities has been elaborated. Recommendations of additional measures that may need to be implemented to achieve a reasonable fire protection of these vehicles and machines is included. This project has been financed by TUSC.

The smoke stratification and the smoke descent along a tunnel are of the utmost importance for personnel evacuation. The paper investigates the smoke descent along a tunnel during a naturally ventilated tunnel fire. A theoretical model is developed to predict the smoke depth below the ceiling along the tunnel. A series of numerical simulations of full-scale tunnel fires are conducted to compare with the developed model, and some coefficients such as the entrainment coefficient are determined from the simulation results. The concepts of critical moment and critical distance are proposed to characterize the smoke descent along the tunnel. The results show that as the smoke spreads longitudinally, the smoke depth below the tunnel ceiling continuously increases. The temperature decay along the tunnel due to heat losses and air entrainment at the smoke layer interface is considered as the main parameter for the smoke descent. After the vitiated air returns back to the fire source, the smoke stratification in the entire tunnel will be significantly reduced. The smoke layer depth along the tunnel based on the temperature distribution is relatively stable in the process of smoke development, which is not sensitive to the HRR, but influenced by the tunnel width, and this method could only be used before the critical moment. The outcomes of this study could provide references for a better understanding of smoke movement in naturally ventilated tunnels and provide technical guidelines for fire safety designers.

The paper analyses the breathing air consumption among participating firefighters during full-scale tests performed in the Tistbrottet mine in Sweden 2013. The availability of breathing air during firefighting has in earlier work been identified as a critical tactical factor in underground firefighting. Results from the tests show that there are differences in the breathing air consumption and that this depends on the methods used, equipment and the workload. The use of BA-teams, i.e. firefighters equipped with breathing apparatuses, is a complex group activity where the largest breathing air consumer will set the limits for the whole team. Light equipment and a structured command and control during the activities will enhance the endurance and the firefighting performance. Equipment and methods affect both firefighting performance and the durability of the firefighting activities. Examples of tested methods and equipment during the test series are: different variations of conventional hose lay-out; CAFS; cutting extinguisher; and trolley for equipment and complementary air. The aid of additional air supply and the use of trolleys will support the activities but is dependent on a large degree of preparation and training to function properly. Based on the tests, it is concluded that the larger model of air bottles should be considered for distances longer than 75 m. © 2022 The Authors

An evacuation experiment was carried out at the Äspö Hard Rock Laboratory in March 2018 to investigate human behaviour during evacuation in underground facilities via escape routes with long ascending tunnels. The objective of the experiment was to collect data that could be used as a basis for evacuation risk and safety assessments in underground tunnels and other large infrastructure projects related to e.g., mining. In total, 32 participants individually ascended the 907 m long tunnel with an inclination of 14%. During the evacuation, each participant's walking speed, vertical walking speed, heart rate and estimated physical exertion was documented. The measured walking speeds were found to be higher than the walking speeds obtained in previous experiments, but the vertical walking speeds were lower. The strategy of 44% of the participants was to adjust their walking speed to a pace they thought could be maintained over a longer distance. The results of this experiment show that the walking speed decreased as the level of perceived exertion increased. Moreover, the results indicate that the walking speed and the level of perceived exertion stabilized during the movement along the tunnel. © 2022 The Author(s)

A series of 1:3 intermediate scale tunnel fire tests was performed to investigate the performance of a fully automatic sprinkler system in a road tunnel. The experiments were carried in a container tunnel with scaled geometry, using wood pallets as the fire source to simulate HGV fires. The activation of the sprinklers was simulated by using thermocouples that corresponded to a given Thermal Response Index (RTI) of a sprinkler bulb or a link. A total of 12 tests were carried out with varying longitudinal velocities (0.8–1.7 m/s), sprinkler activation temperatures (68–141 °C), water densities (2.9–8.7 mm/min) and types of arrangement of the fuel. The activation times, number of activated sprinklers, maximum heat release rates and other key parameters are presented and analyzed. The results show that the water density plays a key role in the performance of the automatic sprinkler systems tested. A high tunnel ventilation velocity, low water density and low sprinkler activation temperature are not recommended. © 2022 The Authors

ReliS – Reliable Sprinkler, är en förstudie som undersökt sprinklersystem (gruppaktiveringssystem) i rorolastutrymmen på rorofartyg, även kallade drenchersystem. Det övergripande målet med projektet var att ta fram förslag på tekniska och operativa förbättringar, för att undvika felfunktion och göra systemen mer tillförlitliga. Den huvudsakliga funktionen för ett drenchersystem är densamma som för ett delugesystem: när en deluge sektion är aktiverad ska vatten tömmas ut från alla öppna munstycken i deluge ("drencher") sektionen. Dessa fyra tillvägagångssätt för att samla in kunskap och erfarenhet om drenchersystemen användes i studien: - Litteraturstudie för att granska föreskrifter och andra vägledande dokument för drenchersystemet. - Intervjustudie för att få förståelse för goda exempel, frågeställningar, skeppsspecifika applikationer m.m. - Fältstudie för att undersöka systeminstallationen tillsammans med besättningsrepresentanter ombord och för att delta i flödestester av drenchersystem. - Systemgruppsworkshops efter en strukturerad metod för att hitta innovativa lösningar. De framtagna förslagen till förbättringar av drenchersystemens tekniska och operativa design utvecklades under Workshop #3 av Systemgruppen. Idéerna Rörmaterial och Undvik havsvatten vid testning valdes för att vidareutveckla. Användningen av havsvatten ökar korrosionshastigheten i stålrörssystem. Vattnets salthalt har också en effekt på korrosionshastigheten. För de fartyg som ingick i denna förstudie var det skillnad mellan besättningarnas upplevelse av igensatta munstycken och korrosion om fartyget trafikerade den svenska västkusten eller östkusten. Besättningsmedlemmar på fartyg som trafikerar östkusten, som har lägre salthalt i vattnet, verkade inte ha erfarenhet av igensatta munstycken och korrosion i samma utsträckning som de på fartygen som trafikerar västkusten. Besättningsmedlemmar på fartyg som trafikerade östkusten var mer oroade över att uppfylla kraven och att faktiskt dämpa en brand, vilket också var en uttryckt oro från de intervjuade inspektörerna. Problem med intern korrosion i rör är inte unikt för drenchersystem på fartyg, det är även problematiskt för landbaserade applikationer, som i vägtunnlar och i byggnader. Resultatet av projektet kommuniceras direkt till sjöfartsnäringen via de referens- och systemgrupper som upprättats för projektet samt till Transportstyrelsen för vidare anmälan till IMO för utveckling av säkerhetsregler för fartyg.

Rapporten avses att utgöra ett stöd för planering och riskbedömning inför och vid räddningsinsatser i spårtunnlar. Rapporten sammanfattar tidigare forskning och kunskap avseende branddynamik i tunnlar, räddningstjänstens förflyttning vid insats, brandsläckning i fordon samt insatshöjande åtgärder i form av såväl bärbar teknik och utrustning som permanenta installationer i tunneln.

Räddningsinsatser i tunnlar har traditionellt innefattat samtidig slanguppbyggnad, baserat på tolkningen av arbetsmiljöföreskrifterna för rökdykning (AFS 2007:7) samt förfarandet vid rökdykarinsatser vid brand i byggnad. Förhållandena vid bränder i byggnader och vid bränder i tunnlar skiljer väsentligt. I en byggnad är tillgången till säker vattenförsörjning en förutsättning för rökdykarens säkerhet. I en tunnel, där ytskikten är obrännbara, brandgaserna inte ansamlas i ett begränsat utrymme och berget kyler brandgaserna längre bort från branden, är inte alltid tillgången till säker vattenförsörjning det som bäst tillgodoser säkerheten för räddningstjänstens personal när arbetsmoment som inte innefattar släckning genomförs. Detta innebär inte att kraven på säkerhet ska sänkas, utan att andra åtgärder som bibehåller eller höjer säkerhetsnivån behöver vidtas. Sådana åtgärder kan vara utökad användning av värmekamera, utökad övervakning av luftförsörjningen, tekniska installationer som stödjer orientering och vägledning samt utrustning, exempelvis spårbundna räddningstrallor, som medför snabbare förflyttning och minskad luftförbrukning.

I rapporten föreslås en indelning i fyra zoner, som direkt kopplar till de branddynamiska förhållanden som råder i respektive zon. Förenklat innebär zonindelningen följande:

• Kall zon – Ingen risk för rökfyllnad. Ingen risk för värmepåverkan från branden.
• Sval zon – Viss risk för rökfyllnad. Ingen risk för värmepåverkan från branden.
• Varm zon – Delvis eller helt rökfylld tunnel, men liten risk för värmepåverkan från branden.
• Het zon – Direkt påverkan från branden. Brand eller risk för brand.

Till respektive zon kopplas sedan relevanta skyddsnivåer och det arbete som bedöms kunna genomföras beskrivs. I varm zon bedöms arbete kunna genomföras utan tillgång till säkert vatten, utom i de fall syftet med arbetet är att bygga upp vattenförsörjning för senare arbete i het zon. I het zon ska alltid säker tillgång till vattenförsörjning finnas.

En viktig del av en säker insats i tunnlar, med eller utan säkert vatten, är den kunskap som både den enskilde rökdykaren och de befäl som beslutar om operativa eller strategiska åtgärder behöver besitta. Rapporten ger därför både en vetenskaplig grund för beslut och pekar på indikatorer som kräver övervakning i syfte att säkerställa rökdykarnas arbetsmiljö.

The study focuses on the behaviors of spilled liquid from a continuously leaked tank in sloped tunnels. Spillage width and area, which impact the potential heat release rates in case of fire, are investigated under different tunnel slopes and leakage flow rates by numerical simulations using interFoam based on the VOF method in the OpenFOAM toolbox following the validation. The simulation results show that the spillage width initially decreases rapidly and then slowly as the tunnel slope increases. Other parameters, including road surface roughness, physical properties of liquid and leakage source height, are also considered. Empirical models for predicting the spillage width and area are established considering both tunnel slope and leakage flow rate. The results may provide guidance for tunnel safety design and drainage system design affiliated with a tank leakage inside a tunnel. © 2021 The Author(s)

There can be situations, for example if gas containers have been damaged in a vehicle crash, when the fire and rescue service would like to depressurize the gas containers through shooting with a civilian rifle. Modern high-pressure hydrogen containers are designed for a working pressure of 700 bars. This means that they have a very thick and strong shell made of composite material. At the same time the fire and rescue service only have access to civilian rifles and ammunition that can be bought for hunting purposes. Thus, tactical and safe depressurization of hydrogen containers is a big challenge. RISE have, together with the Södra Älvsborgs Fire and Rescue Services (SÄRF), Swedish Civil Contingency Agency, and Lund University conducted shooting tests of gas tanks mounted on a hydrogen gas vehicle and three stand-alone hydrogen gas tanks. The shooting tests were conducted at Remmene shooting field in Sweden. Thirteen shooting tests with hydrogen tanks placed in favouarable positions were performed. Out of these, only four tests were succesful in puncturing the individual gas tank in a single shot. Furthermore, two unwanted events occurred; one rupture (after 7 shots) and two powerful jets (after 20 and one shot respectively). This shows that further development and research is required in order to develop a method to safely depressurize high pressure hydrogen tanks.

Ingen ska skadas eller drabbas av sjukdom på grund av sitt arbete. Detta ställer krav på genomtänkt utformning av brandskyddet och utrymningssäkerhet under byggtiden för tunnel eller andra undermarksanläggningar. Ett syfte med denna handbok är att tydliggöra vilka krav som ställs på brandskydd under byggtiden, samt ge exempel på vanliga brandrisker och förslag på hur dessa kan hanteras.

Handboken vänder sig till aktörer, såsom byggherrar, projektörer, entreprenörer och byggarbetsmiljösamordnare, involverade i planering, projektering och byggnation av undermarksanläggningar. Med undermarksanläggning menas här väg-, järnväg-, vatten-, avlopps- kabel- och distributionstunnel, bergrum och slutförvar.  Handboken omfattar inte gruva, byggnader ovan mark, eller brandskydd i färdigställd anläggning.

Handboken är avsedd att användas som vägledning och inspiration vid planering, projektering, genomförande, samt utbildning och övning med fokus på brandsäkerhet. 

Klicka på länken ”TUSC Handbok” högst upp till höger för att öppna handboken. Strax under finns det en länk till en MS Power Point introduktion till handboken, ett övningsprotokoll och ett kommunikations material.

Fire spalling of concrete is a complex phenomenon with many influencing factors. The concrete mix, geometry and cross section, stress state, fire exposure, age, pre-conditioning, moisture state and the amount of spalling reducing polypropylene fibres added all influences in different ways if a fire exposed concrete structure is prone to spalling or not. During fire testing of concrete structures for tunnels it is in practice difficult to do fire test on real designs with real boundary conditions. Therefore, more or less representative test methods are used. If small test specimens are used, around 600 x 500 x 200 mm3, these tests are in general only indicative and shall mainly be used as “screening tests”, i.e. to optimise the mix before scaling up to more realistic sizes. Further on, the small test specimen size is not suitable for evaluating the fire resistance including heat penetration for longer duration of fire exposure as the geometrical and boundary effects are not representative for larger specimens.

To manage thermal-driven smoke from fires in a subway station, a mechanical ventilation system is usually installed. Such a system normally consists of an air supply system and a smoke exhaust system. In case of a platform fire, the ventilation systems will be activated to control the smoke and to provide better environmental conditions for personnel evacuation. This paper conducted a theoretical analysis and a series of CFD simulations to study the critical velocity (minimum air velocity) at the stairways for preventing smoke propagating from the platform to the upper floor through the stairways in case of a platform fire. Correlations for critical velocities are established for two typical types of stairways (i.e., with and without side slabs). It is found that the critical velocity is well correlated with the heat release rate by a 1/3 power law function for both types of stairways, but it varies with the height of the smoke curtain at the stairway by a 1.375 and a 2.55 power law function for stairways with and without side slabs, respectively. The results may serve as a useful reference for smoke control in subway platform fires.

A series of fire tests was carried out to investigate the diffusion flame characteristics of double fires generated from separated burners in a naturally ventilated tunnel, considering different heat release rates and fire separation distances. The results show that the flame tilt angle, as well as the horizontal projected flame length, first increases with fire separation distance and then remains constant, but the vertical flame length first decreases and then remains constant, which is different from two fires in free spaces where flames do not tilt when the separation distance is relatively long. This difference is caused by the non-dimensional fire induced air flow velocity in the tunnel, which is mainly related to the tunnel cross-section dimensions and burner radius. Three regions can be identified, i.e. flame vertical merging, plume vertical merging and non-merging with flame tilted. The critical flame merging separation distance, estimated by the flame merging probability, is greater than that of two fires in an open environment due to the fire-induced air flows. The merging flame height is lower than that of a single fire with a same heat release rate. A correlation was proposed to estimate flame height of two fires in a tunnel by the modified non-dimensional heat release rate using an air entrainment perimeter as the characteristic length. This work enhances the understanding of diffusion flame behaviors of double fires in naturally ventilated tunnels.

In this work, a review of current literature on artificial intelligence (AI) and more specifically machine learning (ML) is presented. ML is illustrated by two case studies where artificial neural networks are used for regression analysis of 110 spalling experiments and 81 Fire Dynamics Simulator (FDS) models of tunnel fires. Tunnel fires are often assessed by fire safety engineers using time-consuming simulation tools where a trained model has the potential to significantly reduce time and cost of these assessments.

A regression model based on a neural net is used to study small scale spalling experiments and similar accuracy compared to least-square fits are obtained. The result is a function based on 14 determining experimental parameters of spalling and result in, spalling times and depths. It is a relatively small effort to get started and set up models, comparably to regular curve fitting. In this first case study the training times are short, it is thus possible to establish how the model performs on average.

The 81 tunnel fire simulations are trained using a similar neural net however it takes considerable time to organize data, creating input, target data of the desired format and training. Here, it is also crucial to normalize the data in order to have it in a suitable format when training. 

It should be noted that ML is often an iterative process in such a way that it may be difficult to know what settings will work before starting the process. It is equally important to illustrate and get to know the data, e.g., if there are large differences or orders of magnitude differences in the data. A normalization procedure is most often practical and will give better predictions.

A series of tests was carried out to investigate the influence of different parameters on design fires for tunnels with water-based fire suppression systems in a 1:4 scale tunnel. The key parameters tested included fuel load covers, activation time, water density, nozzle type, ventilation velocity, sprinkler section length and tunnel width. Analyses of test data are presented with a focus on the influence of these parameters on the design fire. The main findings are that the fuel load covers, activation time and ventilation velocity all play significant roles in the fire development. Further, suggestions on design fires for tunnels with water-based fire suppression systems are proposed. © 2020 The Authors

The study focuses on the mass flow rate of the buoyancy-driven gases exhausted from the shaft in naturally ventilated urban road tunnel fires. Theoretical analyses and numerical simulations are performed. The model to predict the mass flow rate of the incoming smoke exhausted by the nearest shaft is developed by considering that the smoke is exhausted along the four sides of the shaft separately. Based on the heat balance between the incoming smoke exhausted and the total gas flow exhausted, the model to estimate the total mass flow rate exhausted from the shaft (both smoke and entrained air) is also established. Meanwhile, a series of numerical simulation in a naturally ventilated tunnel considering the heat release rate (HRR), the shaft height, shaft length and width, shaft location was carried out. The simulation results show that the shaft height has a limited contribution to the mass flow rate of the incoming smoke exhausted while a larger shaft cross-sectional area shows a favorable performance in exhausting the smoke. Further, the air entrainment into the shaft increases with both the shaft height and shaft cross-sectional area. Comparisons of the mass flow rates of the incoming smoke and the total mass flow rates exhausted between simple calculations and simulations are made, showing that the simple models perform well. Further, it is found that there exist two regimes for the total mass flow rate corresponding to different smoke modes in the shaft (complete plug-holing, plug-holing and without plug-holing), which is caused by the different driven forces in the shaft. The outcomes of this work could provide some guidance for the design of vertical shaft and smoke control in naturally ventilated tunnel in urban area.

I denna rapport beskrivs genomförande av försök med en ventilationstub i tunnel i reducerad skala (1:3). Försöken är genomförda i en tunnel som byggts av fyra 40 ”fot” stålcontainrar i rad med en total längd på 49 m. Tunneln var öppen i ena ändan och stängd i den andra (stuff). Ventilationstuben var upphängd i taket och kopplad till ett plåtrör som i sin tur var kopplad till en axialfläkt placerad 2 m utanför tunneln. Den totala längden på tuben var 32 m därav 30 m inne i tunneln. Syftet med försöken var att undersöka effektiviteten hos ventilationstuber gjorda av en flamskyddad polyesterväv med PVC-beläggning som utsatts för en brand med flammor underifrån. Tunneln som användes var 2,7 m hög och 2,6 m bred. Diametern på ventilationstuben var 0,6 m. Genom att multiplicera alla längder med 3 så uppnås motsvarande dimensioner i fullskala. Till exempel motsvarar försökstunneln en 147 m lång tunnel med stuff och en ventilationstub på 1,8 m i diameter. Tilluftsflödet i ventilationskanalen var 2,3 m3/s vilket motsvarar 36 m3/s i fullskala. Brandens dimensionerade storlek var 0,8 MW respektive 1,9 MW vilket motsvarar 12 respektive 30 MW i fullskala. Det i sin tur motsvarar brandeffekter från en stor hjullastare och en borrigg [1]. Försöken kombinerade med beräkningar visar att för en 30 MW brand kommer ventilationstuben inom 7 m (på varje sida) att antändas och brinna bort. Tuben kommer att påverkas av brandgaserna genom att den delvis krymper eller smälter och därmed ramlar ner inom 90 m (på varje sida). Alla siffror är uppräknade till fullskala. Beräkningar visar att för en 20 MW brand kommer tuben med 3 m (på varje sida) att antändas och brinna bort. Tuben kommer att påverkas av brandgaserna genom att den delvis krymper eller smälter och därmed ramlar ner inom 60 m (på varje sida). För en 15 MW brand kommer förmodligen enbart tuben ovanför branden att antändas av flamman och brinna bort och tuben kommer att påverkas av brandgaserna inom 40 m (på varje sida). För en 10 MW brand kommer förmodligen enbart tuben ovanför branden att antändas, medan den kommer att påverkas av branden inom 20 m på varje sida. Det är också möjligt att ventilationstuben ovanför flamman smälter utan att den antänds. Alla de beräknade siffrorna är beroende på tunnelhöjden då den påverkar temperaturen inne i tunneln. Ett fortsatt luftflöde genom tuben kommer att inträffa om fläkten är i funktion hela tiden. Om fläkten stängs av under perioden finns det risk för att luftflödet igenom inte kan återupptas. Orsaken till detta är att änden på tuben som brinner av smälter ihop (stängs). En observation från försöken är att om gastemperaturen i taket ligger mellan smälttemperaturen (180 oC) och självantändningstemperaturen (500 oC), kommer tuben i första hand krympa ihop för att sedan även smälta. Det inkluderar även väven i tuben. Tuben kan smälta om temperaturen är hög (över 260 oC för polyesterväven). För temperaturer över 500 oC kommer tuben att brinna bort helt och hållet. Ovan angivna värden är konservativa för de scenarier som beaktas. Flamman som sprids genom tuben har inte inkluderats och därmed kan skadeintervallet vara större än de observerade. Försöken indikerar dock ingen flamspridning i ventilationstuben förutom i det område där flamman finns. Även om försöken är gjorda i skala 1:3 uppfyller de kraven om rimlighet när det gäller geometri, ventilation, brandstorlek och inverkan av branden på ventilationstuben. I skalmodeller kan man inte alltid uppfylla kraven på alla skalparametrar, till exempel strålning, men eftersom den reducerade skalan är så pass stor (1:3) kan vi anta att resultaten representerar väl vad som kan inträffa i en verklig skala. Ett fullskaletest är alltid att föredra men det kräver mycket resurser, speciellt när det gäller brandeffekten som kan bli hög. Istället för att göra försök i skala 1:3 med 1–3 MW, som presenteras i denna rapport, kräver det 10–30 MW i fullskala. Temperaturerna bör dock vara ungefär de samma som i fullskalan och därför bedömer vi att resultaten ger oss ett underlag för att kunna gå vidare för bedömning av resultaten i en fullskalig anläggning. I följande delar av rapporten presenteras bakgrunden till projektet, metoderna, resultaten och de slutsatser som dras.

Denne studien belyser ulike aspekter ved personsikkerheten ved brann i tunnel og svarer ut konkrete spørsmål omkring temaet.

Oppdragsgiver er Bane NOR. Prosjektet har fått innspill fra en arbeidsgruppe som er koordinert og ledet av hhv. KS Bedrift og Bane NOR – med fagressurser fra Vestfold Interkommunale Brannvesen IKS (VIB), Bergen brannvesen (BB), Oslo brann- og redningsetat (OBRE), Bane NOR, operatørselskaper (Vy og Flytoget), Direktoratet for samfunnssikkerhet og beredskap (DSB) og Statens havarikommisjon for transport (SHT).

Rapporten er delt inn i to hoveddeler. Del 1 omhandler kartlegging av relevante forskningsprosjekt, dimensjonerende brannscenarier og røykkontroll, se sammendrag og forslag til veien videre i underkapittel 3.5. Del 2 omhandler kartlegging av kunnskap om menneskelig atferd i forbindelse med tunnelbrann, se sammendrag og forslag til veien videre i underkapittel 4.5. Denne delen er utarbeidet av Lunds Tekniska Högskola og WSP Sverige, og er følgelig på svensk.

This report includes the Proceedings of the 9th International Symposium on Tunnel Safety and Security (ISTSS) held in Münich, Germany, 11-13th of March, 2020. The Proceedings include 42 papers given by session speakers and 13 extended abstracts presenting posters exhibited at the Symposium. The papers were presented in 12 different sessions. Among them are Keynote sessions, Tunnel Safety Concepts, Fire Dynamics, Risk Analysis 1&2, Tunnel Safety Design Concepts, Poster Corner, Explosion Hazards, Active Protection 1&2, Emergency Management, Ventilation, Passive Protection and Evacuation.  

Each day was opened by invited Keynote Speakers (in total six) addressing broad topics of pressing interest. The Keynote Speakers, selected as leaders in their field, consisted of Anne Lehan, German Highway Research Institute, Germany, Marc Tesson, Centre for Tunnel Studies (CETU), France, Trond H. Hansen, Oslo Fire and Rescue Service, Norway, Mia Kumm, RISE, Sweden, Roland Leucker, Research Association for Tunnels and Transportation Facilities (STUVA), Germany and Rune Brandt, HI Haerter, Switzerland.   We are grateful that the keynote speakers were able to share their knowledge and expertise with the participants of the symposium.   

Fortifikationsverket (FORTV) has expressed a desire to investigate the design, reliability, performance and cost of a sprinkler system for a typical underground fortification facility. Based on the cost and the benefit associated with a sprinkler system, a cost-benefit analysis was performed. In addition, water mist fire protection systems were studied. The installation cost analysis was based on two fictious facilities; a small facility with a net area of 1 000 m² and a large facility with a 5 000 m² net area.

The estimated installation cost for a traditional sprinkler system in the smaller type facility is about SEK 1,3 million and about SEK 3,3 million for the larger type facility. The installation cost for a high-pressure water mist system is higher than that of a traditional sprinkler system for the smaller type facility but comparable for the larger type facility. A low‑pressure water mist system seems to be the least expensive option for both types of facilities. This is probably because the system, unlike a traditional sprinkler system, requires smaller pipe sizes, smaller water pumps and a smaller water tank and unlike a high-pressure system uses normal steel pipes and less expensive centrifugal pumps.

The cost-benefit analysis for the fictitious type facilities shows that a sprinkler system is cost-effective, especially for the larger type facility. But it should be noted that the uncertainty in the data base is quite large, which means that the trends in the result can be used for further analysis, but that the actual values ​​of the benefit ratio should be viewed with some caution. The sprinkler system mainly has an effect to reduce the property loss. The expected benefit for personal injury is around one percent of the total benefit of the sprinkler system. This is because the risk of fatality and injuries in the event of a fire is small, as people can usually put themselves in safety. The reduction in property loss was assumed to be 75%, and an assumed lowered benefit of sprinklers (50% and 25% property loss reduction, respectively) leads to a lower benefit ratio but for the large type facility the benefit ratio is still above 1,0. The benefit of sprinklers also decreases if the assumed fire frequency is reduced. However, for the larger type plant, the calculation shows that there is still a benefit, even if the assumed fire frequency is halved. The same applies if the cost of replacement of expensive equipment is assumed to be half as high.

The paper presents a theoretical study on the ceiling jets induced by small fires in tunnels. The ceiling jet thickness, temperature rise and velocity are analyzed theoretically with Non-Boussinesq approximation. The study focuses on the radial and one-dimensional ceiling jets. Numerical solutions in the radial region and one-dimensional shooting region are obtained and new analytical solutions in the critical flow region are achieved. Analytical solutions indicate that the ceiling jet thickness increases with distance away from the fire source, which largely differs from the existing models implying that the ceiling jet thickness in the one-dimensional critical flow remains constant. Additionally, impacts of the air entrainment, friction and heat transfer on the ceiling jet are analyzed. It is found that in the radial and one-dimensional shooting flow regions, the air entrainment has a much more significant effect than the friction and heat transfer. However, in the one-dimensional critical flow region, the impact of air entrainment seems to be negligible and the flow is dominated by the friction and heat transfer. Further, validation of the present theory is made by comparing with previous theories, semi-empirical models, and experiments. The results show that the present theory provides a good prediction of the ceiling jet properties with natural ventilation for a small fire. © 2020 The Authors

Smoke control is a key issue in tunnel fire accidents. While jet fans have been widely used in road tunnels, mobile fans provide a good complement, due to its flexibility to operate, especially if there are no fixed jet fans present. To confirm the feasibility of a mobile fan system, full scale fire tests were conducted in Kalldal tunnel in Sweden, with fire size in the range of 1–2.6 MW and a mobile fan placed at the tunnel portal. In the tests, it takes about 4–5 min to establish full flow in the opposite direction and the final flow velocity can reach 1.5–1.9 m/s. To describe the transient behavior of flow development inside tunnel, a one-dimensional lumped theoretical model has been developed. The model takes into account the pressure losses of external wind, the stack effect of fire and friction losses over tunnel walls and two portals. The model is validated using the data from Kalldal tunnel fire tests. Results show that the model can well predict the initial and final steady state velocity, but underestimates the flow development. The model gives prediction on the safe side.

In this study, a series of small-scale experiments was carried out in a model scale tunnel with dimensions of 20 m (Length) × 2 m (Width) × 1 m (Height) to investigate the characteristics of buoyancy-driven smoke flow in a tunnel with vertical shafts. Different shaft settings and four different longitudinal ventilation velocities were tested in the experiments. A theoretical model for the mass flow rate of buoyancy-driven smoke flow in the shaft was developed and validated. The gas temperature along the tunnel ceiling and smoke stratification were subsequently analyzed and discussed. The results showed that more shafts, greater shaft heights and greater shaft cross sectional areas can significantly increase the smoke extraction rate, and the total smoke mass flow rate in the shafts increases with the increasing ventilation velocity. The local pressure loss coefficient at the shaft inlet may not be a fixed value. An average value of 1.0 for this coefficient was recommended for engineering estimation and design of rectangular-shaped natural shafts. The presence of vertical shaft is beneficial to the smoke stratification and could increase the height of the smoke layer interface, especially for the downstream of the shaft.

Due to the increasing demand to minimize the global pollution, the research has made significant advances in electrically powered vehicles. This interest has also spiked in the mining industry since the current diesel-powered vehicles emits exhaust and particles, which results in the need for a costly ventilation system. But switching from a known, and widely used, propulsion system results in a step towards the unknown. This literature survey has mainly studied battery electric vehicles (BEV).

This literature survey focusses on the issues related to thermal runaway of a battery and what it might cause. This includes the toxicity of gasses and combustion products, the impact on rescue operations, the effect on extinguishment and differences in fire behavior between BEVs and internal combustion engine (ICE) vehicles.

Full scale experiments have shown that the fire behavior might not depend solely on the energy storage as other component also contribute to the overall fire behavior as whole. These experiments have been conducted on single passenger vehicles and not mining vehicles, and it is reasonable to draw the conclusion that batteries as an energy storage do not change the overall fire behavior considerably. Changes do, however, occur when it comes to the extinguishment of a fire, since the battery design interferes with the possibility to cool the battery cells effectively. This difficulty in cooling the battery cells makes it very challenging prevent thermal runaway to stop a thermal runaway and propagation event inside the battery using conventional fire suppression methods.

Fuels with a high energy density have contributed to the development of modern communities. On the other hand, fuels contain energy that, during some conditions, can result in incidents, not least within transportation. CNG vehicles are designed according to safety standards of UNECE, including events such as fire. In case of a fire a thermally activated Pressure Relief Device (TPRD) should empty the container before a pressure vessel explosion potentially can occur. CNG tanks are according to UNECE regulation 110 tested against a 1.65 m long pan fire. However, local fires are not included in these tests. This report presents fire tests of CNG containers performed both with a UNECE compatible fire source and with a local fire source. Any pressure vessel explosion and jet flames were characterized for two different types of CNG containers, namely steel and composite. In five out of six tests the safety of the CNG containers prevailed also in the event of a local (0.24 m by 0.24 m) pan fire, meaning that no pressure vessel explosion occurred. In real vehicle fires, where the fire extends from its local characteristics to a more developed fire that expose the CNG containers to a larger extent, these tests support that TPRDs most likely will activate. The experience from running these test series call for that the fire source should be more accurately defined with regards to fuel and dimensions and a local fire should be included in the UNECE Regulation 110.

With the purpose to increase the knowledge on human behaviour when evacuating along elevated platforms, an experiment was carried out at the subway station of Skarpnäck, Stockholm, in October 2016. The overall project objective was to develop basic data for guidelines regarding fire safety design concerning evacuation along elevated platforms. The experiment was designed as a group experiment divided into five different scenarios. In total, 111 persons of mixed gender and age participated. The results from the experiment show that the flow rate along the elevated platform decreased as the walkway was getting narrower. It could also be seen that along the first half of the walkway, where a train was located on the rail track next to the elevated platform, the flow rate was higher and the width of the walkway was used to a lager extent compared to the second half of the walkway where the platform was open to the track area. One of three wheelchair users who participated in the experiment expressed discomfort caused by the height and the width of the walkway and nearly half of the participants experienced problems with passing others walking slower than themselves.

An increasing number of tunnels have been built around the world. They play an important role to relieve traffic congestions and facilitate goods transportation. However, in the event of a fire in tunnels, the consequences can be serious due to its narrow-long structure. The previous studies about tunnel fire dynamics and mitigation measures are mostly based on good ventilation conditions in tunnels, such as longitudinal ventilation and natural ventilation with the premise that a tunnel has two open portals. However, the studies about the characteristics of tunnel fires under confined portal boundaries with complete or incomplete sealing at both portals are rare. Typical fire scenarios can appear in a subway train, a building corridor, an underground utility tunnel, a mining tunnel, a tunnel during construction and the application of sealing tunnel portals for fighting large tunnel fires and so on. The knowledge of tunnel fire dynamics for tunnels under good ventilation conditions is probably not applicable to the scenarios of tunnel fires under confined portal boundaries. Conducting the studies of tunnel fires under confined portal boundaries is of great significance for better understanding the characteristics of this type of tunnel fires and developing tunnel fire mitigation measures. Therefore, by combining model-scale tunnel experiments and theoretical analyses, this thesis studies the fire behaviors and smoke transportation law of tunnel fires under confined portal boundaries. The main research contents include:

1．Scaling effects of mass loss rate per unit area (MLRPUA) for well-ventilated pool fires are studied by summarizing large amounts of experimental data from the literature together with theoretical analyses. As a further extension of tunnel fire similarity theory, it provides the basis and reference for later model/medium-scale tunnel experiments. Results show that when a small-scale pool fire (D<1 m) occurs in the open, increasing wind velocity tends to increase the MLRPUA, especially for pools with D<0.2 m. This is because the ventilation significantly increases the conductive and convective heat feedbacks (leading role). But when small-scale pool fires occurs in tunnels with a short distance between the pool surface and ceiling (H_ef/D<3), the radiative heat feedback from the tunnel ceiling is probably dominating, leading to a much higher MLRPUA than that in the free burn. When subjected to longitudinal flows, the MLRPUA decreases due to the reduced radiation effect from the ceiling. With the increase of pool diameter, the influence of wind on the MLRPUA decreases gradually, no matter whether the pool occurs in the open or in a tunnel. Finally, when the pool diameter exceeds 1 m, the radiation from flame itself is probably predominant. The MLRPUA is not significantly affected by increasing wind velocity and most likely fluctuates within 30% for a wide range of wind velocities based on the test data collected.

2．The flame behaviors and the maximum gas temperature rise beneath the ceiling in an enclosed tunnel are studied using a model-scale tunnel. Results show that when a fire (small fire) is not located at the tunnel center, the flame inclines towards the closer tunnel end due to the asymmetric flow field on both sides of the flame. The flame inclination angle keeps increasing when the fire is moving away from the tunnel center. Furthermore, when a fire is in Region I (0< ≤0.64), the maximum gas temperature rise decreases with the increasing dimensionless fire distance due to the increasing flame inclination angle. When a fire is in Region II (0.64< <1), the maximum gas temperature rise increases with the increasing dimensionless fire distance due to the heat feedback of returned hot smoke bounced from the end wall. By introducing a concept of equivalent ventilation velocity based on the flame inclination mechanism, a prediction model of maximum gas temperature rise beneath the ceiling in Region I is developed. Beyond that, an extra correction factor is proposed to the improved model in Region II with a consideration of heat feedback of returned hot smoke bounced from the end wall. Besides, further dimensional analysis indicates that the normalized maximum gas temperature rise follows an exponential attenuation law with the dimensionless fire distance.

3．The coupling control effects of sealing ratio and initial sealing time on the fire development (large fire) are studied using a model-scale tunnel. Results show that sealing tunnel portals can decrease the mass loss rate of fuel and gas temperature inside the tunnel, no matter whether the sealing is complete or incomplete. The earlier the initial sealing time is, the better the fire can be controlled. For the incomplete sealing, when the sealing is implemented during the violent burning stage, the sealing not only does not limit the fire growth but also exacerbates the tunnel fire, producing an extremely high CO concentration at tunnel portals and a longer ceiling flame jet. This will result in a huge threat to the rescue service at tunnel portals. Besides, if the tunnel portals are sealed incompletely, it will leave a small area for the exchange of smoke and air. The smoke will not continue to spread horizontally after leaving the tunnel portals under the action of inertial forces. In order to maintain the combustion of fuel, the fresh air from external environment flows into the tunnel vigorously and quickly from the gap and then uplifts the smoke out of the tunnel portals, which is also an important phenomenon for firefighters and needs to draw their attentions.

4．The critical conditions for the occurrence of under-ventilated tunnel fires and the combustion mechanisms under confined portal boundaries are studied by using both model-scale and medium-scale tunnels. Results show that the critical equivalence ratio for the occurrence of under-ventilated tunnel fires is within 0.53 - 0.6, which is less than the theoretical value of 1. This is related to the occurrence of vitiation, consequently reducing the level of oxygen around the flame by diluting the O₂ concentration. The low ventilation rate and vitiation result in a low O₂ volume fraction around the flame, and then the MLRPUA starts to decrease and at the same time the air mass flow into the tunnel becomes almost constant. Also, an oscillating MLRPUA and lifted flame are observed in the model-scale tests. Consequently, the ventilation rate approaches and even reaches the amount required for complete combustion of vaporized fuel. This means that the insufficient combustion in early under-ventilated tunnel fires has converted to sufficient combustion (from the perspective of the change of equivalence ratio, the fire has converted from under-ventilated to well-ventilated). As a result, no significant increase in CO production in under-ventilated fires is observed in both test series.

5．The critical conditions for the occurrence of self-extinguishment and influencing factors in under-ventilated tunnel fires are studied in a model-scale tunnel during construction. The tunnel consists of an inclined access tunnel and a horizontal main tunnel. Results show that when a fire is in the horizontal main tunnel, the critical equivalence ratio for self-extinguishment is within 0.28 - 1.38 for the propane gas burner and 1.11 - 3.6 for the fibre board soaked with heptane. The difference is related to the burning behavior of the different fuels used. Moreover, the critical O₂ volume fraction is about within 12 - 15% when the fires self-extinguish. When a fire is at the closed end of the horizontal main tunnel, the stratification of smoke is destroyed after hitting the closed end, and then the smoke seems to spread over the entire cross section of the tunnel. The smoke spread velocity is proportional to the ventilation rate. However, when a fire occurs at the closed end of the inclined access tunnel, the fire does not self-extinguish, even when the ventilation rate is 0 m³/s. The corresponding smoke spread velocity is higher than that in the horizontal main tunnel. This is probably related to the increasing component of buoyancy in the longitudinal direction in the inclined access tunnel. Besides, no insignificant vitiation behind the fire is found. These two characteristics in the inclined access tunnel increase the temperature of smoke flowing out of the tunnel portal and in turn promote the natural ventilation and increase the O₂ volume fraction.

The paper presented the main results of large-scale fire tests with different types of fixed firefighting systems (FFFS) conducted in the Runehamar tunnel in June 2016. The background to the tests, the performance of the different systems, and conclusions regarding the efficiency of the systems were presented. The fire load consisted of 420 standardised wooden pallets and a target of 21 wooden pallets. Five of the tests were carried out with a 30 m long deluge zone delivering varying water densities using three different types of side-wall nozzles with an interval distance of 5 m. One test with 93 °C glass-bulb automatic sprinklers in the same zone was also conducted. In the five deluge tests, the detection system was simulated using thermocouples in the tunnel ceiling. The alarm was registered when the ceiling gas temperature reached 141 °C, and the system was activated manually after a delay of 4 min. The protection goal of the system was to prevent fire spread to a target positioned 5 m from the rear of the main fuel area, and to ensure that the fire did not exceed 30 MW in size. The system setups tested were found to meet these goals.

Thepurpose of this study is to highlight the possibility of a better workingenvironment through an increased use of Self-Compacting Concrete (SCC). Suchconcrete is denser and more durable, has higher compressive strength and abetter filling capacity. The main advantage of SCC is that it doesn’t requirevibration for proper compaction and therefore is better to work with. Despiteinitial higher cost to order, it becomes cheaper in a life cycle analysis. Ofall industries, the construction industry has the highest number of workersaffected by vibrations. According to statistics 2016, vibration damage totaled 36% of all approved occupational diseases. The resulting damage can rarely becured and often leads to reduced working capacity and to severe life-longproblems. A study showed that exposure increase of vibrations of just 1 m/s²increases the risk of white fingers (9%), Raynaud’s phenomenon (6.9%),neurosensory injury (7.4%) and carpal tunnel syndrome (2.9%). Most of thevarious activities with elements of vibration on the construction site, such assawing, grinding, screwing with machine etc., are difficult to replace withvibration-free methods. But for concrete casting, it is now possible to almostcompletely exclude vibrations by using SCC. The increased knowledge of theimportance of a healthy workplace for the economic performance of constructioncompanies has played a key role for work environment work. One study comparedthe benefits of accident prevention initiatives with the costs of the same. Itwas revealed that the benefit surpasses the costs with the relationship 3:1. Improvementsfor a healthier workplace can be achieved by: New innovations, both mechanicaland material innovations, making work easier. Studies using modern portablesensors from which information can play an important role in the possibility ofreducing work-related musculoskeletal disorders.

Encouragingworkers to use wearable sensors that can alert when ergonomically dangerousmovements are carried out. Changing of an often reluctant culture at theworkplace. Better risk assessment at the design stage. Better planning inproject design in the early stages using new digital tools such as BIM. Betterinformation about risks and safety in order to affect the safety behaviour ofworkers. We recommend a larger working environment study that will show how apositive workplace development with SCC is possible and how much can be savedin this way. Here, medical expertise must be involved to increase reliability.In addition, there is a desire in medical research, on the working environmentand on vibration-related injuries, to fill in the gaps in previous researchthat remain, such as how cold weather and individual risk factors, such assmoking, affect the risk of vibration-related work injuries.  Improved working environment is the mainreason for increased use of SCC. Increased use should provide a moresustainable society.

This paper studies the overall smoke control of natural ventilation systems with vertical shafts during fires in a common road tunnel by numerical modelling. The variables studied include the heat release rate, longitudinal fire location along the tunnel, length of shafts and the interval between two shafts. Simulation results indicate that the total smoke spread length on both sides of fire source is closely independent of the heat release rate and longitudinal fire locations. For a given dimensionless shaft interval (the ratio of the shaft interval to shaft length), with the increase of shaft length, the smoke spread length firstly increases, reaching a maximum at 12 m, and then decreases significantly until 18 m. For a fire less than 30 MW, the first shaft pair on both sides of fire source prevents the critical-temperature smoke (270 °C) from spreading beyond this shaft. For a 100 MW fire, in the cases with shorter shaft lengths (L shaft ≤9 m), the critical-temperature smoke can't be controlled between the first shaft pair. The gas temperature at human height (1.8 m) is less than 60 °C in all cases with shafts. Downdraught occurs when the smoke front stabilizes at the bottom of a shaft and the buoyancy force could be too low to overcome the kinetic pressure of the air flow flowing into this shaft, consequently destroying the structure of smoke layer. In most scenarios, the total exhaust area of shafts that is required to exhaust all the smoke is about 100 m 2 . The first shaft pair plays a critical role to exhaust the smoke, and its exhaust efficiency is also affected significantly by the shaft length. This study investigates how to control the smoke by using vertical shafts in a road tunnel fire and the conclusions are useful to tunnel fire protection engineering.

The re-direction of smoke flow in inclined tunnel fires refers to the phenomenon that the smoke flow direction suddenly changes due to the changes of thermal buoyancy or outside pressure or the activation of fans. This poses special risk for fire rescue services fighting fires in tunnels. Both small-scale tunnel fire tests (28 scenarios) and numerical simulations of full-scale tunnel fires (31 scenarios) were conducted to study this special phenomenon. A one-dimensional model was used to predict the flow velocity in the inclined tunnels, based on two different methods for calculating the mean smoke temperature (Method I and Method II, respectively). Results show that the smoke flow direction could be well predicted by the model with Method II. When the ventilation velocity is relatively large and the flow tends to be one dimensional, both methods produce similar results. Further, the influences of important factors on the re-direction of smoke flows were systematically analyzed. These factors include heat release rate, tunnel slope, tunnel length, friction factor, tunnel cross sectional area and fire source location.

This paper investigates the influence of wind on mass loss rate per unit area (MLRPUA) of fuel-controlled pool fires both in an open environment and inside tunnels and the scale effect of pool fires is also investigated. Large pool fires with a diameter D greater than 1 m (D > 1 m) are of key concern but small pool fires (D < 1 m) are also considered for comparison. This is done by analyzing large amounts of experimental data from the literature. Results show that for small pool fires (D < 1 m) in an open environment, increasing wind speed tends to increase the MLRPUA, especially for pools with D < 0.2 m, where the MLRPUA could increase significantly with the increase of wind speed. But when small pool fires occur in tunnels, the results are more complex. When the ratio of effective tunnel height to pool diameter is less than 3, increasing wind speed tends to decrease the MLRPUA. When this ratio is greater than 3, the influence of wind on MLRPUA of pool fires in tunnels is similar to that in an open environment. The influence of wind on the MLRPUA decreases for larger pool diameters, no matter whether the pool fire occurs in an open environment or in a tunnel. For large pools with D > 1 m, the MLRPUA is not affected significantly by increasing wind speed and most likely varies within 30% for a wide range of wind speeds based on the test data collected. This influence is far less than the values concluded by previous studies based on small pool fire experiments. The outcome of this study contributes to improving the understanding of burning characteristics of pool fires under windy conditions, especially large pool fires, which are much more meaningful than small pool fires from the perspectives of fire protection engineering and fire hazard assessment.

This paper presents results concerning fire and explosion hazards of various alternative fuel vehicles in tunnels. The alternative fuels are divided into four types: liquid fuels, liquefied fuels, compressed gases, and electricity, and detailed parameters are obtained. Three important types of fire hazards: pool fires, jet fires and fireballs are identified and investigated. From the perspective of pool fire size, the liquid fuels pose equivalent or even lower fire hazards compared to traditionally used fuels, but the hazards of liquefied fuels may be higher. For pressurized tanks, fires are generally much larger in size but shorter in duration. The gas releases from pressure relief devices and the resulting jet flames are highly transient. For hydrogen vehicles, the fire sizes are significantly higher compared to CNG tanks, while flame lengths are only slighter longer. Analysis of the peak overpressure in case of an explosion in a tunnel shows that the consequences of tank rupture and BLEVE are relatively tolerable for a position further than 100 m away, but the situations in case of a cloud explosion are highly severe and intolerable for tunnel users in most cases. These hazards need to be carefully considered in both vehicle safety design and tunnel fire safety design. 

The paper presents a study on the characteristics of tunnel fires during construction. A model-scale tunnel was built and fire tests were conducted. The tunnel consists of an inclined access tunnel and a horizontal main tunnel. The main tunnel has two dead ends (excavation faces) and the only opening is from one side of the access tunnel. Propane gas burner and the fibre board soaked with the heptane were used as fuels. The flame characteristics, O 2 and CO volume fraction and gas temperature were measured and recorded. Two typical characteristics of self-extinguishment and smoke spread were found in the tunnel fires during construction. Results indicate that when a fire occurs in the horizontal main tunnel, the critical equivalence ratio for the occurrence of self-extinguishment is within 0.28–1.38 for the propane gas burner and 1.11–3.6 for the fibre board soaked with heptane. The difference is related to the burning behavior of the different fuels used. The fire location in the horizontal tunnel also has a significant influence on the fire development. A well-ventilated fire at the center of the horizontal tunnel becomes under-ventilated due to vitiation when it is located at the closed end of the horizontal tunnel. Besides, when a fire occurs at the closed end of the horizontal main tunnel, the stratification of smoke is destroyed after hitting the closed end, and then the smoke seems to spread over the entire cross section of the tunnel. The smoke spread velocity is found to be proportional to the ventilation rate. However, when a fire occurs at the closed end of the inclined access tunnel (lower end), the fire does not self-extinguish, even when the ventilation rate is 0 m 3 /s. The corresponding smoke spread velocity is higher than that in the horizontal main tunnel. The outcomes of this study provide new experimental information that contributes to improve the understanding of characteristics of tunnel fires during construction and can help firefighters to make better decisions during the rescue processes.

This paper investigates the characteristics of under-ventilated fires in tunnels. This was done by using both model and medium-scale tunnels. The fuels used were heptane and xylene. The mass loss rates per unit area, ventilation rates from tunnel inlet, flame characteristics, O 2 , CO and CO 2 concentrations, optical densities and heat release rates were measured and recorded. Results show that the fire behaviors in under-ventilated tunnel fires are different from that in well-ventilated fires. In under-ventilated fires, the mass loss rate per unit area is found to decrease during identical periods due to the low oxygen concentration resulting from the low ventilation rate and vitiation, meanwhile the flame size dramatically reduces with a lifted and fluttering flame. This was clearly observed in model-scale tests, but due to limited optical view there was no possibility to observe this in the medium-scale tests. As a result, the ventilation rate approaches the amount required for complete combustion of vaporized fuel. This indicates that the combustion has converted from ventilation-controlled to fuel-controlled. No significant increase in CO production is observed in under-ventilated fires. Besides, the equivalence ratio and combustion efficiency were analyzed in order to judge whether the combustion is fuel-controlled or ventilation-controlled. This study provides new experimental information that contributes to improving the understanding of characteristics of under-ventilated fires in tunnel and can help firefighters to make right judgements and take related protective measures during the rescue processes.

Determination of critical velocity is a key issue for smoke control in any tunnel with longitudinal ventilation. The critical Froude model using single Froude number of 4.5 has for decades been widely used in engineering applications. This value was originally used by Danziger and Kennedy and they argued that the critical Froude number obtained by Lee et al. was in a range of 4.5 and 6.7 and therefore a conservative value of 4.5 was obtained. This paper explores the validity of using single critical Froude number of 4.5 by investigating the original sources and comparing it to recent research results. It was found that the value of 4.5 obtained in the original source corresponds to a large tunnel fire and it correlates well with data from other literature within a narrow range of large fire sizes. Using this value produces a significantly lower critical velocity for a wide range of fire sizes and therefore it is not conservative. The Froude number of 6.7 obtained by Lee et al. corresponds to another Froude number with a different definition and it is therefore not comparable with the value of 4.5. It is found that the use of a single value of 4.5 for the critical Froude number is not reasonable in calculation of the critical velocity for smoke control in tunnels with longitudinal ventilation.

A series of model scale tunnel fire tests was carried out to investigate effects of the fire suppression system on production of key combustion products including CO and soot. The key parameters accounted for in the tests include fuel type, ventilation velocity and activation time. The results show that fire suppression indeed has influence on production of combustion products especially for cellulose fuels. In case that the fire is not effectively suppressed, e.g. when the water density is too low or activation is too late, the CO concentration and visibility could be worse than in the free-burn test. From the point of view of production of combustion products, only fire suppression systems with sufficient capability and early activation are recommended to be used in tunnels.

Tunnel and rock lining systems are used for drainage and icing protection. These systems can consist of any combination of concrete, metal, plastic or textile. The report summaries the available methods, both for testing and for installation. The large variation in both systems and test methods often make it difficult for constructors or designer to understand the importance of different methods. The report gives indication of what type of linings exists and how to ensure the fire safety of such systems. Fire safety properties can be verified in three different ways: #1 Full systems can be tested in full scale fire tests, #2 a section of the system can be tested in standardized furnace tests, or #3 plastic and/or textile membrane can be tested with regards to requirements on fire spread. It is suggested to require that a fire should not be able to propagate in the system. This can be verified with #3 above requiring class B, C or D according to EN 13501-1. If the lining system offers structural fire protection, it can be verified suing #2 above.

Model scale tests with varying materials as tunnel structure were carried out to further study the theoretical model of maximum gas temperature for structural protection. New correlation for calculation of air mass flow rate is introduced. Test results showed that the maximum ceiling gas temperatures increases with the increasing heat release rate and decreases with the increasing tunnel width and thermal inertia of the tunnel linings. Higher ventilation velocity may also result in slightly higher temperatures for large fires.

Comparisons of model scale tests and theoretical models showed that the theoretical models predict the maximum ceiling gas temperature very well. A fire with a fixed heat release rate or a time-varying heat release rate, the effects of tunnel structure, tunnel ventilation, tunnel width and fire size have been well considered by the model. Comparisons of other model and full scale tests with theoretical models further verified this.

In the past two decades, the interest in fire safety science of tunnels has significantly increased, mainly due to the rapidly increasing number of tunnels built and the catastrophic tunnel fire incidents occurred. This paper presents an overview of research on fire safety in underground road and railway tunnels from the perspectives of fire safety design. The main focuses are on design fires, structural protection, smoke control and use of water-based fire suppression systems. Besides, some key fire characteristics, including flame length, fire spread, heat flux and smoke stratification, are discussed.

The report describes different underground systems including mines and tunnels during construction (tunneling). The key factors that affect fire development in underground systems are described. Proposal and recommendations for ventilation strategies in case of fire are given. The report covers both fuel- and ventilation-controlled fires. In general, a minimal ventilation limits the fire growth and may even inert the fire through ascended smoke. A minimal ventilation also contributes to improved conditions for a first fire extinguishing attempt and evacuation.

These proceedings include papers and extended abstracts from the 5th International Conference on Fires in Vehicles – FIVE 2018, held in Borås, Sweden October 3-4, 2018. The proceedings include an overview of research and regulatory actions coupled to state-of-the-art knowledge on fire related issues in vehicles, such as passenger cars, buses, coaches, trucks and trains.

Fires in transport systems are a challenge for fire experts. New fuels that are efficient and environmentally friendly are rapidly being introduced together with sophisticated new technology such as e.g. fuel cells and high energy density batteries. This rapid development, however, introduces new fire risks not considered previously and we risk getting a situation where we do not have enough knowledge to tackle them. In this context FIVE represents an important forum for discussion of the fire problem and for exchange of ideas.

Fire protection in road, rail, air, and sea transport is based on international regulations since vehicles cross borders and the safety requirements must be the same between countries. Therefore, understanding of safety and regulations must be developed internationally and the FIVE-conference has a significant role to play as a place to exchange knowledge.

FIVE attracts researchers, operators, manufacturers, regulators and other key stakeholders. Of particular value is the mix of expertise and the international participation in the conference. The conference is unique as it includes fires in different vehicles. It is not confined to bus fires or train fires but includes them both, naturally since fire problems are often similar regardless of type of vehicle. This means that for example solutions for trains are useful for fire problems in buses and vice versa.

In the proceedings you will find papers on the fire problem, alternative fuel and electric vehicles, fire investigations and case studies and finally fire mitigation. We are grateful to the renowned researchers and engineers presenting their work and to the keynote speakers setting the scene. We sincerely thank the scientific committee for their expert work in selecting papers for the conference.

This report includes the Proceedings of the 8th International Symposium on Tunnel Safety and Security (ISTSS) held in Borås, Sweden, 14-16th of March, 2018. The Proceedings include 41 papers given by session speakers and 16 extended abstracts presenting posters exhibited at the Symposium. The papers were presented in 12 different sessions. Among them are Fire Safety Engineering: Cases & Incidents, Fire Safety Engineering: The Aims, Fire Detection, Explosions, Risk Analysis, Fire Safety Engineering: Case studies, Ventilation, Fire Safety Engineering: State of the Art, Fire Dynamics, Fixed Fire Fighting Systems (FFFS) and Evacuation and Human Behavior.

Each day was opened by invited Keynote Speakers (in total six) addressing broad topics of pressing interest. The Keynote Speakers, selected as leaders in their field, consisted of Hans Brun, the Swedish Defence University, Dr Iain Bowman, Mott MacDonald, Canada, Dr Ying Zhen Li, RISE Research Institutes of Sweden, Dr Johan Lundin, WSP, Sweden, Allan Skovlund, Greater Copenhagen Fire Department, Denmark and Prof David Purser, Hartford Environmental Research, UK. We are grateful that the keynote speakers were able to share their knowledge and expertise with the participants of the symposium.

An investigation of fire and explosion hazards of different types of alternativefuel vehicles in tunnels is presented. The different fuels are divided into four types:liquid fuels, liquefied fuels, compressed gases, and electricity, and detailed parameters are obtained. Three types offire hazards for the alternativefuel vehicles: pool fires, jet fires and fireballs are identified andinvestigated in detail. Fromthe perspective of pool fire size, the liquid fuels pose equivalent or evenmuch lower fire hazards compared to the traditionally used fuels, but theliquefied fuels may pose higher hazards. For pressurized tanks, the fires are generally much larger in size butshorter in duration. The gas releases from pressure relief devices and the resulting jet firesare highly transient. Forhydrogen vehicles, the fire sizes are significantly higher compared to CNGtanks, while flame lengths only slighter longer. Investigation of the peakoverpressure in case of an explosion in a tunnel was also carried out. Theresults showed that, for the vehicles investigated, the peak overpressure of tankrupture and BLEVE are mostly in a range of 0.1 to 0.36 bar at 50 m away. Thesituations in case of cloud explosion are mostly much more severe andintolerable. These hazards need to be carefully considered in both vehiclesafety design and tunnel fire safety design. Further researches on thesehazards are in urgent need.

It’s getting more and more common to

build infrastructures underground, which results in more people using underground facilities in their everyday life. The evacuation routes from this environment often involve long, ascending tunnels. In order to evaluate the evacuation time for these facilities, knowledge about people’s movement in this kind of environment is required. Today the knowledge within this area is limited, why new research within this field is needed.

The current study included two essential parts; 1) initial literature review where the state of the art within the current research field was mapped and 2) an experiment. The purpose of the experiment was to study people

’s walking speed and behaviour during ascending evacuation in inclined tunnels. The aim of the study was to develop data that can be used as basis for guidelines regarding fire safety design in major infrastructure projects and risk and safety assessment of underground facilities.

The experiment was carried out at the Äspö Hard Rock Laboratory in March 2018. In total 32 persons participated and they were asked to individually walk upwards 907 m in a tunnel with an inclination of 14 %. When walking in the tunnel, the participants

’ walking speed, vertical walking speed, pulse and rating of perceived exertion were documented. In addition to the individual tests in the tunnel, the experiment included reference tests and a survey.

When comparing the walking speeds collected in the current experiment with the walking speeds from previous experiments where people were asked to walk upwards long stairs, it can be seen that the walking speeds in the current experiment are higher. When comparing the vertical walking speeds, the result is reversed. A possible explanation for this is that climbing the stairs requires a larger vertical movement compared to moving in the tunnel where the movement is more horizontal.

In the current experiment, 59 % of the participants used an identifiable strategy when moving upwards in the tunnel. The strategies have been categorized as follows:

1) Adjust the walking speed to a pace the participant believe he/she can keep for a longer distance.

2) Focus on breathing.

3) Keep a lower walking speed in the beginning.

4) Concentrate on the surroundings to avoid thinking about how tired he/she is.

5) Focus on the goal.

In general, when walking up the tunnel the walking speed was more or less the same during the whole climb, but the participants experienced a considerable increased perceived exertion. This can be the result of Strategy 1 presented above.

The results of the current experiment show a tendency for walking speed to decrease with increased fatigue. A comparison between the rating of perceived exertion and the normalized walking speed indicates that the walking speed and fatigue stabilized during the movement in the tunnel.

The focus of the present study is analyzing the best position of a sprinkler nozzle in a tunnel cross-section in the Muskö tunnel, south of Stockholm, Sweden. Activation of the sprinklers installed along the centerline and along the sidewall is investigated through analysis of full scale experiments and by three dimensional numerical modelling. Then the tunnel velocity is analyzed by one dimensional numerical modelling for various fire locations in the Muskö tunnel. For both activating the automatic sprinklers nearby the fire and avoiding activation of the sprinklers further downstream, the automatic sprinklers are recommended to be installed along the centerline of the tunnel. It has also been found that the tunnel velocity varies significantly with the fire location. When the fire is on the left side of the tunnel, the flow velocity mostly remains in a range of 1 m/s (positive or negative) within the first 10 minutes, which helps early activation of the automatic sprinklers. When the fire is on the right side of the tunnel, the flow velocity mostly remains within a range of -1 m/s and 1 m/s within the first 5 minutes, and the velocity mostly increases to 2 m/s at around 10 min. Therefore, the scenario for fire located on the left side is better than that for fire on the right side, especially when it is located between the middle of the right section and the right portal. As one typical case with fire on the right side, the tunnel velocity maintains at 1 m/s for the first 5 min and gradually increases to 2 m/s at 10 min. Under such conditions, the automatic sprinkler system is expected to perform well. 

Numerical and theoretical work was conducted to investigate the effect of tunnel cross section on critical velocityfor smoke control in longitudinally ventilated tunnel fires. The results show that for small fires, the criticalvelocity decreases with both the increasing tunnel height and tunnel width. For large fires, the critical velocitysignificantly increases with the increasing tunnel height but is independent of tunnel width. Differentcalculation models are compared with a focus on effect of tunnel cross section. A new correlation is proposedto account for the effect of tunnel width based on the previous model.