We explore a Digital Twin-Based Approach for Smart Manufacturing to improve Sustainability, Efficiency, and Cost-Effectiveness for a steel production plant. Our system is based on a micro-service edge-compute platform that ingests real-time sensor data from the process into a digital twin over a converged network infrastructure. We implement agile machine learning-based control loops in the digital twin to optimize induction furnace heating, enhance operational quality, and reduce process waste. Key to our approach is a Deep Reinforcement learning-based agent used in our machine learning operation (MLOps) driven system to autonomously correlate the system state with its digital twin to identify correction actions that aim to optimize power settings for the plant. We present the theoretical basis, architectural details, and practical implications of our approach to reduce manufacturing waste and increase production quality. We design the system for flexibility so that our scalable event-driven architecture can be adapted to various industrial applications. With this research, we propose a pivotal step towards the transformation of traditional processes into intelligent systems, aligning with sustainability goals and emphasizing the role of MLOps in shaping the future of data-driven manufacturing.

This report focus on the intersection ofForeign Information Manipulation andInterference and Large Language Models.The aim is to give a non-technicalcomprehensive understanding of howweaknesses in the language models canbe used for creating malicious content tobe used in FIMI.

Digital twins (DT) of industrial processes have become increasingly important. They aim to digitally represent the physical world to help evaluate, optimize, and predict physical processes and behaviors. Therefore, DT is a vital tool to improve production automation through digitalization and becomes more sophisticated due to rapidly evolving simulation and modeling capabilities, integration of IoT sensors with DT, and high-capacity cloud/edge computing infrastructure. However, the fidelity and reliability of DT software are essential to represent the physical world. This paper shows an automated and systematic test architecture for DT that correlates DT states with real-time sensor data from a production line in the forging industry. Our evaluation shows that the architecture can significantly accelerate the automatic DT testing process and improve its reliability. A systematic online DT testing method can significantly detect the performance shift and continuously improve the DT's fidelity. The snapshot creation methodology and testing agent architecture can be an inspiration and can be generally applicable to other industrial processes that use DT to generalize their automated testing. 

Defects during production may lead to material waste, which is a significant challenge for many companies as it reduces revenue and negatively impacts sustainability and the environment. An essential reason for material waste is a low degree of automation, especially in industries that currently have a low degree of digitalization, such as steel forging. Those industries typically rely on heavy and old machinery such as large induction ovens that are mostly controlled manually or using well-known recipes created by experts. However, standard recipes may fail when unforeseen events happen, such as an unplanned stop in production, which may lead to overheating and thus material degradation during the forging process. In this paper, we develop a digital twin-based optimization strategy for the heating process for a forging line to automate the development of an optimal control policy that adjusts the power for the heating coils in an induction oven based on temperature data observed from pyrometers. We design a digital twin-based deep reinforcement learning (DTRL) framework and train two different deep reinforcement learning (DRL) models for the heating phase using a digital twin of the forging line. The twin is based on a simulator that contains a heating transfer and movement model, which is used as an environment for the DRL training. Our evaluation shows that both models significantly reduce the temperature unevenness and can help to automate the traditional heating process. © 2022 The authors

The purpose of the project was to develop solutions for automation of postprocessing of 3D-printed sand cores and the goal was to achieve the same production speed for 3D printed cores as for conventional production. Within three industrial demonstrators, communication solutions between printer, robot, and production systems were developed. Technical solutions were tested for adaption of manual work steps to robot. Preconditions for digitalization of the production flow via traceability was tested and the effect on the business models of the foundries were tested. In the making of these demonstrators, several technical issues have been solved, above all the fundamental aspects of data transfer and compatibility between different systems. Therefore, it is possible to motivate a robot investment already from increased efficiency of the first simple work steps.

The project was a cooperation between robot manufacturer, machine supplier, producers of castings, and research institute. The representation of the entire value chain turned out to be a crucial success factor. The results include decision data for investments in a printer and other automation solutions associated w with core manufacturing in the individual foundry. The effect of these results is that Sweden's foundries will strengthen their position with respect to the international competition both short and long term since the results will lead to increased productivity. By evaluating the use of vision systems. The project has also generated inspiration of the development in automation and AI solutions, and further cooperation between the participating companies.

Projektets syfte var att öka tillgängligheten i äldre pressgjutmaskiner och pressgjutceller med minst 10 % genom att identifiera och analysera data som redan idag loggas digitalt, och genom att identifiera vilket slags data som eventuellt ytterligare behöver loggas och hur detta i sådana fall skall ske. En analysmetod som gavs särskilt fokus baserades på maskininlärning, och syftade till att utveckla en modell för tillståndsbaserat, det vill säga prediktivt, underhåll. Projektet kunde slå fast att de deltagande gjuterierna loggar driftstoppdata som endast genom visualisering sannolikt skulle kunna hjälpa dem att fatta beslut som markant kan höja tillgängligheten. För att kunna utveckla mjukvara för tillståndsbaserat underhåll krävs däremot loggning av fler maskinrelaterade processparametrar än vad de i projektet eftermonterade sensorerna klarade av att logga, och dessutom behöver denna loggning ske under mycket längre tid, eftersom driftstoppen visade sig ske med så långa tidsmässiga mellanrum att mängden träningsdata inte hann bli stor nog. Tre olika maskininlärningsbaserade modeller utvecklades och testades, inklusive ett djupt faltningsnätverk, där den bästa av dem uppnådde ett resultat där 11,3 % av alla predikterade maskinstopp var falska positiva, vilket visserligen är bättre än slumpen men inte tillräckligt bra för praktisk implementering. Projektet har resulterat i kunskap och praktisk erfarenhet kring sensorer samt insamling och bearbetning av mätdata som kommer att ligga till grund för vidare arbete mot digitalisering av svenska pressgjuterier, ett arbete som är absolut nödvändigt för att kunna stå sig i den internationella konkurrensen.

The paper presents ongoing research on smart metal castings, meaning the technologicalinnovation of elevating cast metal components into metal components with integratedsensor functionality. Since the innovation targets aim straight at low cost industrial serialproduction, specific high cost and high-end solutions like inclusion of advancedelectronic equipment and after mounted sensors are not part of this innovationdevelopment. Integrating signal carriers inside metal castings to achieve metal castingswith sensor functionality requires robust solutions for connecting the sensor signal to thesensor interrogator and interpreter. The actual transmission of the signal may be donewirelessly or by wire. However, for several reasons there is a challenge with establishingan isolated and distinct connection between the sensor contact, and the contact at theexternal connection, regardless of whether it is to an antenna for wireless transmission orto a wire. This paper presents metallurgical challenges associated with choices ofmaterials, and combinations of metallurgical challenges and production process relatedchallenges, including the high melting temperatures. Aims are to find the rightcombinations of metal alloys, production simplicity, signal stability and robustness. Thepaper will present some of the tests made in the project so far. The project is run in aconsortium of the two Sweden-based industrial companies Husqvarna and SKF, and thetwo Swedish research institutes Swerea SWECAST and RISE Acreo.