Change search
Link to record
Permanent link

Direct link
Alternative names
Publications (10 of 22) Show all publications
Efkarpidis, N., Imoscopi, S., Bratukhin, A., Brännvall, R., Franzl, G., Leopold, T., . . . Sauter, T. (2024). Proactive Scheduling of Mixed Energy Resources at Different Grid Levels. IEEE Transactions on Sustainable Energy, 15, 952
Open this publication in new window or tab >>Proactive Scheduling of Mixed Energy Resources at Different Grid Levels
Show others...
2024 (English)In: IEEE Transactions on Sustainable Energy, ISSN 1949-3029, E-ISSN 1949-3037, Vol. 15, p. 952-Article in journal (Refereed) Published
Abstract [en]

The optimal utilisation of distribution grids requires the proactive management of volatilities caused by mixed energy resources installed into different grid levels, such as buildings, energy communities (ECs) and substations. In this context, proactive control based on predictions for energy demand and generation is applied. The mitigation of conflicts between the stakeholders' objectives is the main challenge for the control of centralized and distributed energy resources. In this paper, a bi-level approach is proposed for the control of stationary battery energy storage systems (SBES) supporting the local distribution system operator (DSO) at the transformer level, as well as distributed energy resources (DERs) operated by end customers, i.e., EC-members. Model predictive control (MPC)- based and hybrid approaches merging rule- and MPC-based control schemes are evaluated. Simulation studies based on a typical European low voltage (LV) feeder topology yield the performance assessment in terms of technical and economic criteria. The results show an advantage of hybrid approaches with respect to the DSO's cost savings from peak shaving. From the EC's perspective, both hybrid and MPC-based schemes can achieve effective cost savings from proactive energy management.

Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers Inc., 2024
Keywords
Controllers; Costs; Electric power distribution; Electric substations; Energy management systems; Energy resources; Level control; Model predictive control; Predictive control systems; Bi-level energy management framework; Cost saving; Deterministics; Distributed Energy Resources; Level controllers; Low-level controllers; Management frameworks; Model-predictive control; Predictive models; RBC; Robust; Stakeholder; Transformer; Uncertainty; Upper level controller; Energy management
National Category
Mechanical Engineering
Identifiers
urn:nbn:se:ri:diva-67990 (URN)10.1109/TSTE.2023.3320055 (DOI)2-s2.0-85173370167 (Scopus ID)
Note

The transnational project SONDER has received funding inthe framework of the joint programming initiative ERA-NetSmart Energy Systems’ focus initiative Integrated, RegionalEnergy Systems, with support from the European Union’sHorizon 2020 research and innovation programme under grantagreement No 775970. 

Available from: 2023-11-24 Created: 2023-11-24 Last updated: 2024-06-10Bibliographically approved
Brännvall, R. (2024). The Inhibitor: ReLU and Addition-Based Attention for Efficient Transformers. In: Proceedings of the AAAI Conference on Artificial Intelligence: . Paper presented at 38th AAAI Conference on Artificial Intelligence, AAAI 2024. Vancouver, Canada. 20 February 2024 through 27 February 2024 (pp. 23445-23446). Association for the Advancement of Artificial Intelligence, 38(21)
Open this publication in new window or tab >>The Inhibitor: ReLU and Addition-Based Attention for Efficient Transformers
2024 (English)In: Proceedings of the AAAI Conference on Artificial Intelligence, Association for the Advancement of Artificial Intelligence , 2024, Vol. 38, no 21, p. 23445-23446Conference paper, Published paper (Refereed)
Abstract [en]

To enhance the computational efficiency of quantized Transformers, we replace the dot-product and Softmax-based attention with an alternative mechanism involving addition and ReLU activation only. This side-steps the expansion to double precision often required by matrix multiplication and avoids costly Softmax evaluations but maintains much of the core functionality of conventional dot-product attention. It can enable more efficient execution and support larger quantized Transformer models on resource-constrained hardware or alternative arithmetic systems like homomorphic encryption. Training experiments on four common benchmark tasks show test set prediction scores comparable to those of conventional Transformers with dot-product attention. Our scaling experiments also suggest significant computational savings, both in plaintext and under encryption. The ReLU and addition-based attention mechanism introduced in this paper may enable privacy-preserving AI applications operating under homomorphic encryption by avoiding the costly multiplication of encrypted variables.

Place, publisher, year, edition, pages
Association for the Advancement of Artificial Intelligence, 2024
Keywords
Artificial intelligence; Computational efficiency; Computational savings; Conventional transformer; Core functionality; Double precision; Ho-momorphic encryptions; Homomorphic-encryptions; MAtrix multiplication; Scaling experiments; Test sets; Transformer modeling; Cryptography
National Category
Computer and Information Sciences
Identifiers
urn:nbn:se:ri:diva-72840 (URN)10.1609/aaai.v38i21.30422 (DOI)2-s2.0-85189627116 (Scopus ID)
Conference
38th AAAI Conference on Artificial Intelligence, AAAI 2024. Vancouver, Canada. 20 February 2024 through 27 February 2024
Available from: 2024-04-29 Created: 2024-04-29 Last updated: 2024-04-29Bibliographically approved
Fredriksson, S., Eleftheriadis, L., Brännvall, R., Bäckman, N. & Gustafsson, J. (2023). ANIARA: Experimental Investigation of Micro Edge Data Centers with Battery Support on Power-Constrained Grids. In: Companion Proceedings of the 14th ACM International Conference on Future Energy Systems: . Paper presented at e-Energy '23 Companion: Companion Proceedings of the 14th ACM International Conference on Future Energy Systems. June 2023 (pp. 72-78). Association for Computing Machinery
Open this publication in new window or tab >>ANIARA: Experimental Investigation of Micro Edge Data Centers with Battery Support on Power-Constrained Grids
Show others...
2023 (English)In: Companion Proceedings of the 14th ACM International Conference on Future Energy Systems, Association for Computing Machinery , 2023, p. 72-78Conference paper, Published paper (Refereed)
Abstract [en]

As the demand for data privacy and low latency grows, edge computation carried out at edge data center nodes is believed to become increasingly important for future telecom applications. Providers must consider various factors, including power consumption, thermal dynamics, and the ability to maintain high-quality service, in addition to deployment and service orchestration. This paper presents a detailed description of two different prototype edge data centers designed to investigate the power performance and thermal dynamics of edge nodes under various applied services. The prototypes were developed and tested at the RISE ICE Datacenter research facility. We present the results of power flow experiments in which input current from the grid was limited while the computational load was maintained using the energy stored in batteries. We further discuss implications for placing edge data center nodes in locations with temporal power constraints and opportunities for participation in support services at the grid level.

Place, publisher, year, edition, pages
Association for Computing Machinery, 2023
Keywords
Thermodynamics, Edge DC, power infrastructure, Power flow dynamics
National Category
Communication Systems
Identifiers
urn:nbn:se:ri:diva-65655 (URN)10.1145/3599733.3600252 (DOI)
Conference
e-Energy '23 Companion: Companion Proceedings of the 14th ACM International Conference on Future Energy Systems. June 2023
Available from: 2023-07-05 Created: 2023-07-05 Last updated: 2023-07-05Bibliographically approved
Brännvall, R., Stark, T., Gustafsson, J., Eriksson, M. & Summers, J. (2023). Cost Optimization for the Edge-Cloud Continuum by Energy-Aware Workload Placement. In: Companion Proceedings of the 14th ACM International Conference on Future Energy Systems: . Paper presented at e-Energy '23 Companion: Companion Proceedings of the 14th ACM International Conference on Future Energy Systems. June 2023 (pp. 79-84). Association for Computing Machinery
Open this publication in new window or tab >>Cost Optimization for the Edge-Cloud Continuum by Energy-Aware Workload Placement
Show others...
2023 (English)In: Companion Proceedings of the 14th ACM International Conference on Future Energy Systems, Association for Computing Machinery , 2023, p. 79-84Conference paper, Published paper (Refereed)
Abstract [en]

This article investigates the problem of where to place the computation workload in an edge-cloud network topology considering the trade-off between the location-specific cost of computation and data communication. For this purpose, a Monte Carlo simulation model is defined that accounts for different workload types, their distribution across time and location, as well as correlation structure. Results confirm and quantify the intuition that optimization can be achieved by distributing a part of cloud computation to make efficient use of resources in an edge data center network, with operational energy savings of 4–6% and up to 50% reduction in its claim for cloud capacity.

Place, publisher, year, edition, pages
Association for Computing Machinery, 2023
Keywords
cost optimization, sustainability, data center, edge, energy efficiency
National Category
Computer Systems
Identifiers
urn:nbn:se:ri:diva-65654 (URN)10.1145/3599733.3600253 (DOI)
Conference
e-Energy '23 Companion: Companion Proceedings of the 14th ACM International Conference on Future Energy Systems. June 2023
Available from: 2023-07-05 Created: 2023-07-05 Last updated: 2023-07-05Bibliographically approved
Brännvall, R., Forsgren, H. & Linge, H. (2023). HEIDA: Software Examples for Rapid Introduction of Homomorphic Encryption for Privacy Preservation of Health Data. Studies in health technology and informatics, 302, 267-271
Open this publication in new window or tab >>HEIDA: Software Examples for Rapid Introduction of Homomorphic Encryption for Privacy Preservation of Health Data
2023 (English)In: Studies in health technology and informatics, Vol. 302, p. 267-271Article in journal (Refereed) Published
Abstract [en]

Adequate privacy protection is crucial for implementing modern AI algorithms in medicine. With Fully Homomorphic Encryption (FHE), a party without access to the secret key can perform calculations and advanced analytics on encrypted data without taking part of either the input data or the results. FHE can therefore work as an enabler for situations where computations are carried out by parties that are denied plain text access to sensitive data. It is a scenario often found with digital services that process personal health-related data or medical data originating from a healthcare provider, for example, when the service is delivered by a third-party service provider located in the cloud. There are practical challenges to be aware of when working with FHE. The current work aims to improve accessibility and reduce barriers to entry by providing code examples and recommendations to aid developers working with health data in developing FHE-based applications. HEIDA is available on the GitHub repository: https://github.com/rickardbrannvall/HEIDA.

Place, publisher, year, edition, pages
IOS Press, 2023
Keywords
Artificial Intelligence, GDPR, Privacy Preservation, Sensitive Data, algorithm, computer security, privacy, software, Algorithms
National Category
Computer Sciences
Identifiers
urn:nbn:se:ri:diva-64939 (URN)10.3233/SHTI230116 (DOI)2-s2.0-85159768596 (Scopus ID)
Note

 Corresponding Author: Rickard Brännvall,RISE, Sweden. E-mail: rickard.brannvall@ri.se

Available from: 2023-06-12 Created: 2023-06-12 Last updated: 2024-04-09Bibliographically approved
Forsgren, H., Brännvall, R., Vesterlund, M. & Minde, T. B. (2023). Homomorphic Encryption Enables Data and Algorithm Confidentiality for Remote Monitoring and Control: An Application to Data Center Systems. In: Companion Proceedings of the 14th ACM International Conference on Future Energy Systems: . Paper presented at e-Energy '23 Companion: Companion Proceedings of the 14th ACM International Conference on Future Energy Systems. June 2023 (pp. 85-90). Association for Computing Machinery
Open this publication in new window or tab >>Homomorphic Encryption Enables Data and Algorithm Confidentiality for Remote Monitoring and Control: An Application to Data Center Systems
2023 (English)In: Companion Proceedings of the 14th ACM International Conference on Future Energy Systems, Association for Computing Machinery , 2023, p. 85-90Conference paper, Published paper (Refereed)
Abstract [en]

The design of intelligent algorithms used for device monitoring and control can be costly and is an investment that must be protected against reverse engineering by competitors. An algorithm can be safeguarded by running remotely from the cloud instead of locally on the equipment hardware. However, such a setup requires that sensitive data is sent from the device to the cloud. Fully Homomorphic Encryption (FHE) is an emerging technology that offers a solution to this problem since it enables computation on encrypted data. A cloud service using FHE can protect its proprietary algorithms while simultaneously offering customer data confidentiality. The computational overhead for the technology is, however, still very high. This work reports on a practical investigation of using FHE for data center remote control problems: What applications are feasible today? And at what cost?

Place, publisher, year, edition, pages
Association for Computing Machinery, 2023
Keywords
remote control, homomorphic encryption, confidential computing, remote monitoring, neural networks
National Category
Computer Sciences
Identifiers
urn:nbn:se:ri:diva-65653 (URN)10.1145/3599733.3600254 (DOI)
Conference
e-Energy '23 Companion: Companion Proceedings of the 14th ACM International Conference on Future Energy Systems. June 2023
Available from: 2023-07-05 Created: 2023-07-05 Last updated: 2023-10-31Bibliographically approved
Brännvall, R., Gustafsson, J. & Sandin, F. (2023). Modular and Transferable Machine Learning for Heat Management and Reuse in Edge Data Centers. Energies, 16(5), Article ID 2255.
Open this publication in new window or tab >>Modular and Transferable Machine Learning for Heat Management and Reuse in Edge Data Centers
2023 (English)In: Energies, E-ISSN 1996-1073, Vol. 16, no 5, article id 2255Article in journal (Refereed) Published
Abstract [en]

This study investigates the use of transfer learning and modular design for adapting a pretrained model to optimize energy efficiency and heat reuse in edge data centers while meeting local conditions, such as alternative heat management and hardware configurations. A Physics-Informed Data-Driven Recurrent Neural Network (PIDD RNN) is trained on a small scale-model experiment of a six-server data center to control cooling fans and maintain the exhaust chamber temperature within safe limits. The model features a hierarchical regularizing structure that reduces the degrees of freedom by connecting parameters for related modules in the system. With a RMSE value of 1.69, the PIDD RNN outperforms both a conventional RNN (RMSE: 3.18), and a State Space Model (RMSE: 2.66). We investigate how this design facilitates transfer learning when the model is fine-tuned over a few epochs to small dataset from a second set-up with a server located in a wind tunnel. The transferred model outperforms a model trained from scratch over hundreds of epochs.

Keywords
edge data center, heat management, heat reuse, modular machine learning, transferable machine learning, recurrent neural network, transfer learning, meta-learning
National Category
Electrical Engineering, Electronic Engineering, Information Engineering
Identifiers
urn:nbn:se:ri:diva-64222 (URN)10.3390/en16052255 (DOI)
Note

Funding: Vinnova through the Celtic Next project AI-NET Aniara with project-ID C2019/3-2

Available from: 2023-03-10 Created: 2023-03-10 Last updated: 2023-08-28Bibliographically approved
Linge, H. & Brännvall, R. (2023). Secure Sharing of Health-Related Data: Research Description of the VINTER, DELFIN, and HEIDA Projects. Studies in Health Technology and Informatics, 302, 143-144
Open this publication in new window or tab >>Secure Sharing of Health-Related Data: Research Description of the VINTER, DELFIN, and HEIDA Projects
2023 (English)In: Studies in Health Technology and Informatics, ISSN 0926-9630, E-ISSN 1879-8365, Vol. 302, p. 143-144Article in journal (Refereed) Published
Abstract [en]

The need for secure and integrity-preserved data sharing has become increasingly important in the emerging era of changed demands on healthcare and increased awareness of the potential of data. In this research plan, we describe our path to explore the optimal use of integrity preservation in health-related data contexts. Data sharing in these settings is poised to increase health, improve healthcare delivery, improve the offering of services and products from commercial entities, and strengthen healthcare governance, all with a maintained societal trust. The HIE challenges relate to legal boundaries and to the importance of maintaining accuracy and utility in the secure sharing of health-related data.

Place, publisher, year, edition, pages
NLM (Medline), 2023
Keywords
Artificial Intelligence, GDPR, Privacy Preservation, Sensitive Data, health care delivery, information dissemination, trust, Delivery of Health Care
National Category
Computer and Information Sciences
Identifiers
urn:nbn:se:ri:diva-64940 (URN)10.3233/SHTI230087 (DOI)2-s2.0-85159764063 (Scopus ID)
Available from: 2023-06-12 Created: 2023-06-12 Last updated: 2023-10-31Bibliographically approved
Heimerson, A., Sjölund, J., Brännvall, R., Gustafsson, J. & Eker, J. (2022). Adaptive Control of Data Center Cooling using Deep Reinforcement Learning. In: Proceedings - 2022 IEEE International Conference on Autonomic Computing and Self-Organizing Systems Companion, ACSOS-C 2022: . Paper presented at 3rd IEEE International Conference on Autonomic Computing and Self-Organizing Systems Companion, ACSOS-C 2022, 19 September 2022 through 23 September 2022. Institute of Electrical and Electronics Engineers Inc.
Open this publication in new window or tab >>Adaptive Control of Data Center Cooling using Deep Reinforcement Learning
Show others...
2022 (English)In: Proceedings - 2022 IEEE International Conference on Autonomic Computing and Self-Organizing Systems Companion, ACSOS-C 2022, Institute of Electrical and Electronics Engineers Inc. , 2022Conference paper, Published paper (Refereed)
Abstract [en]

In this paper, we explore the use of Reinforcement Learning (RL) to improve the control of cooling equipment in Data Centers (DCs). DCs are inherently complex systems, and thus challenging to model from first principles. Machine learning offers a way to address this by instead training a model to capture the thermal dynamics of a DC. In RL, an agent learns to control a system through trial-and-error. However, for systems such as DCs, an interactive trial-and-error approach is not possible, and instead, a high-fidelity model is needed. In this paper, we develop a DC model using Computational Fluid Dynamics (CFD) based on the Lattice Boltzmann Method (LBM) Bhatnagar-Gross-Krook (BGK) algorithm. The model features transient boundary conditions for simulating the DC room, heat-generating servers, and Computer Room Air Handlers (CRAHs) as well as rejection components outside the server room such as heat exchangers, compressors, and dry coolers. This model is used to train an RL agent to control the cooling equipment. Evaluations show that the RL agent can outperform traditional controllers and also can adapt to changes in the environment, such as equipment breaking down. 

Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers Inc., 2022
Keywords
Adaptive Control, CFD Modeling, Data-center Cooling, Reinforcement Learning, Adaptive control systems, Computational fluid dynamics, Controllers, Cooling, Deep learning, Learning systems, Computational fluid dynamics modeling, Cooling equipment, Data center cooling, Datacenter, First principles, Machine-learning, Reinforcement learning agent, Reinforcement learnings, Thermal dynamics
National Category
Control Engineering
Identifiers
urn:nbn:se:ri:diva-61598 (URN)10.1109/ACSOSC56246.2022.00018 (DOI)2-s2.0-85143075382 (Scopus ID)9781665471374 (ISBN)
Conference
3rd IEEE International Conference on Autonomic Computing and Self-Organizing Systems Companion, ACSOS-C 2022, 19 September 2022 through 23 September 2022
Note

Funding details: VINNOVA, ITEA3-17002; Funding text 1: This work was supported by Vinnova grant ITEA3-17002 (AutoDC).

Available from: 2022-12-21 Created: 2022-12-21 Last updated: 2023-06-07Bibliographically approved
Brännvall, R., Stark, T., Gustafsson, J., Eriksson, M. & Summers, J. (2022). Cost Optimization by Energy Aware Workload Placement for the Edge Cloud Continuum.
Open this publication in new window or tab >>Cost Optimization by Energy Aware Workload Placement for the Edge Cloud Continuum
Show others...
2022 (English)Report (Other academic)
Abstract [en]

This report investigates the problem of where to place computation workload in an edge-cloud network topology considering the trade-off between the location specific cost of computation and data communication.

National Category
Computer Systems
Identifiers
urn:nbn:se:ri:diva-64293 (URN)
Available from: 2023-04-17 Created: 2023-04-17 Last updated: 2023-06-07Bibliographically approved
Organisations
Identifiers
ORCID iD: ORCID iD iconorcid.org/0000-0003-4293-6408

Search in DiVA

Show all publications