Redox flow batteries have gained wide attention at home and abroad as a long-duration energy storage technology with the advantages of high safety,long lifespan,mutual independence of capacity and power,and easy recyc...Redox flow batteries have gained wide attention at home and abroad as a long-duration energy storage technology with the advantages of high safety,long lifespan,mutual independence of capacity and power,and easy recycling.However,the current battery management technology faces significant challenges,and there is room for development.Digital twin(DT),as a technology that collectively senses,evaluates,predicts,and optimizes characteristics,is promising to contribute to redox flow batteries’operation,maintenance,and management.This paper begins with a brief description of redox flow batteries,followed by a short explanation of the concept and application of DTs.DTs have already made some progress in the field of batteries,and can be applied to solve the problems of redox flow batteries in terms of thermal management and system optimization.Finally,the paper analyzes the combination of redox flow battery and DT architecture,which is expected to contribute to developing DT technology for redox flow batteries.展开更多
Discrete manufacturing workshops are confronted with problems of processing diverse products and strict real time requirements for data service calculation and manufacturing equipment,which makes it difficult to provi...Discrete manufacturing workshops are confronted with problems of processing diverse products and strict real time requirements for data service calculation and manufacturing equipment,which makes it difficult to provide real time feedback and compensation.In this study,a high-availability,high-performance,and high-concurrency digital twin reference model was constructed to satisfy a large number of manufacturing requirements.A multiterminal real-time interaction model and information aging classification rules for virtual and physical models were established.Moreover,a multiterminal virtual interaction model was proposed,and a generalized distributed computing service digital twinning system was developed.This digital twin system was considered a machine tool box processing line as an actual case.Consequently,a full closed-loop manufacturing process digital twin platform for physical request service,real-time response,and quality information feedback from iterations,which provides case guidance for subsequent factory digital twin systems,was realized.The proposed system can satisfy the requirements of multidevice big data computing services in modern manufacturing plants,as well as multiplatform,low-latency,and high-fidelity information visualization requirements for managers.Thus,this system is expected to play an important role in information factories.展开更多
Reinforcement learning has been applied to air combat problems in recent years,and the idea of curriculum learning is often used for reinforcement learning,but traditional curriculum learning suffers from the problem ...Reinforcement learning has been applied to air combat problems in recent years,and the idea of curriculum learning is often used for reinforcement learning,but traditional curriculum learning suffers from the problem of plasticity loss in neural networks.Plasticity loss is the difficulty of learning new knowledge after the network has converged.To this end,we propose a motivational curriculum learning distributed proximal policy optimization(MCLDPPO)algorithm,through which trained agents can significantly outperform the predictive game tree and mainstream reinforcement learning methods.The motivational curriculum learning is designed to help the agent gradually improve its combat ability by observing the agent's unsatisfactory performance and providing appropriate rewards as a guide.Furthermore,a complete tactical maneuver is encapsulated based on the existing air combat knowledge,and through the flexible use of these maneuvers,some tactics beyond human knowledge can be realized.In addition,we designed an interruption mechanism for the agent to increase the frequency of decisionmaking when the agent faces an emergency.When the number of threats received by the agent changes,the current action is interrupted in order to reacquire observations and make decisions again.Using the interruption mechanism can significantly improve the performance of the agent.To simulate actual air combat better,we use digital twin technology to simulate real air battles and propose a parallel battlefield mechanism that can run multiple simulation environments simultaneously,effectively improving data throughput.The experimental results demonstrate that the agent can fully utilize the situational information to make reasonable decisions and provide tactical adaptation in the air combat,verifying the effectiveness of the algorithmic framework proposed in this paper.展开更多
基金Supported by the Special Educating Project of the Talent for Carbon Peak and Carbon Neutrality of University of Chinese Academy of Sciences(E3E56501A2)。
文摘Redox flow batteries have gained wide attention at home and abroad as a long-duration energy storage technology with the advantages of high safety,long lifespan,mutual independence of capacity and power,and easy recycling.However,the current battery management technology faces significant challenges,and there is room for development.Digital twin(DT),as a technology that collectively senses,evaluates,predicts,and optimizes characteristics,is promising to contribute to redox flow batteries’operation,maintenance,and management.This paper begins with a brief description of redox flow batteries,followed by a short explanation of the concept and application of DTs.DTs have already made some progress in the field of batteries,and can be applied to solve the problems of redox flow batteries in terms of thermal management and system optimization.Finally,the paper analyzes the combination of redox flow battery and DT architecture,which is expected to contribute to developing DT technology for redox flow batteries.
基金Project(51975019)supported by the National Natural Science Foundation of ChinaProject(2019 ZX 04024001)supported by the National Science and Technology Major Project of ChinaProject(Z 201100006720008)supported by the Beijing Science and Technology Plan,China。
文摘Discrete manufacturing workshops are confronted with problems of processing diverse products and strict real time requirements for data service calculation and manufacturing equipment,which makes it difficult to provide real time feedback and compensation.In this study,a high-availability,high-performance,and high-concurrency digital twin reference model was constructed to satisfy a large number of manufacturing requirements.A multiterminal real-time interaction model and information aging classification rules for virtual and physical models were established.Moreover,a multiterminal virtual interaction model was proposed,and a generalized distributed computing service digital twinning system was developed.This digital twin system was considered a machine tool box processing line as an actual case.Consequently,a full closed-loop manufacturing process digital twin platform for physical request service,real-time response,and quality information feedback from iterations,which provides case guidance for subsequent factory digital twin systems,was realized.The proposed system can satisfy the requirements of multidevice big data computing services in modern manufacturing plants,as well as multiplatform,low-latency,and high-fidelity information visualization requirements for managers.Thus,this system is expected to play an important role in information factories.
文摘Reinforcement learning has been applied to air combat problems in recent years,and the idea of curriculum learning is often used for reinforcement learning,but traditional curriculum learning suffers from the problem of plasticity loss in neural networks.Plasticity loss is the difficulty of learning new knowledge after the network has converged.To this end,we propose a motivational curriculum learning distributed proximal policy optimization(MCLDPPO)algorithm,through which trained agents can significantly outperform the predictive game tree and mainstream reinforcement learning methods.The motivational curriculum learning is designed to help the agent gradually improve its combat ability by observing the agent's unsatisfactory performance and providing appropriate rewards as a guide.Furthermore,a complete tactical maneuver is encapsulated based on the existing air combat knowledge,and through the flexible use of these maneuvers,some tactics beyond human knowledge can be realized.In addition,we designed an interruption mechanism for the agent to increase the frequency of decisionmaking when the agent faces an emergency.When the number of threats received by the agent changes,the current action is interrupted in order to reacquire observations and make decisions again.Using the interruption mechanism can significantly improve the performance of the agent.To simulate actual air combat better,we use digital twin technology to simulate real air battles and propose a parallel battlefield mechanism that can run multiple simulation environments simultaneously,effectively improving data throughput.The experimental results demonstrate that the agent can fully utilize the situational information to make reasonable decisions and provide tactical adaptation in the air combat,verifying the effectiveness of the algorithmic framework proposed in this paper.
