The increasing use of renewable energy in the power system results in strong stochastic disturbances and degrades the control performance of the distributed power grids.In this paper,a novel multi-agent collaborative ...The increasing use of renewable energy in the power system results in strong stochastic disturbances and degrades the control performance of the distributed power grids.In this paper,a novel multi-agent collaborative reinforcement learning algorithm is proposed with automatic optimization,namely,Dyna-DQL,to quickly achieve an optimal coordination solution for the multi-area distributed power grids.The proposed Dyna framework is combined with double Q-learning to collect and store the environmental samples.This can iteratively update the agents through buffer replay and real-time data.Thus the environmental data can be fully used to enhance the learning speed of the agents.This mitigates the negative impact of heavy stochastic disturbances caused by the integration of renewable energy on the control performance.Simulations are conducted on two different models to validate the effectiveness of the proposed algorithm.The results demonstrate that the proposed Dyna-DQL algorithm exhibits superior stability and robustness compared to other reinforcement learning algorithms.展开更多
Anti-ram bollards used in perimeter protection are tested to meet performance requirements of established standards such as the US Department of State Specification SD-STD-02.01. Under these standards, tests are condu...Anti-ram bollards used in perimeter protection are tested to meet performance requirements of established standards such as the US Department of State Specification SD-STD-02.01. Under these standards, tests are conducted in prescribed conditions that should be representative of the service installation. In actual project, conditions encountered on site may vary from the test environment and it would be expensive and time consuming to validate each deviation with a physical test. High-fidelity physics-based (HFPB) finite element modeling can provide precise simulations of the behavior of anti-ram bollards. This paper presents the use of HFPB finite element modeling, using LS-DYNA, in an actual project to evaluate the performance of an anti-ram bollard design subjected to various boundary conditions representing the physical conditions encountered on site. The study shows that boundary conditions can have a significant influence on the performance of the anti-ram bollards. This suggests that anti-ram bollards must be designed and engineered according to actual conditions that are found on site. It also shows that HFPB modeling can be an effective tool that supplements physical testing of anti-ram bollards.展开更多
基金supported by the National Natural Sci-ence Foundation of China(No.52277108)Guangdong Provincial Department of Science and Technology(No.2022A0505020015).
文摘The increasing use of renewable energy in the power system results in strong stochastic disturbances and degrades the control performance of the distributed power grids.In this paper,a novel multi-agent collaborative reinforcement learning algorithm is proposed with automatic optimization,namely,Dyna-DQL,to quickly achieve an optimal coordination solution for the multi-area distributed power grids.The proposed Dyna framework is combined with double Q-learning to collect and store the environmental samples.This can iteratively update the agents through buffer replay and real-time data.Thus the environmental data can be fully used to enhance the learning speed of the agents.This mitigates the negative impact of heavy stochastic disturbances caused by the integration of renewable energy on the control performance.Simulations are conducted on two different models to validate the effectiveness of the proposed algorithm.The results demonstrate that the proposed Dyna-DQL algorithm exhibits superior stability and robustness compared to other reinforcement learning algorithms.
文摘Anti-ram bollards used in perimeter protection are tested to meet performance requirements of established standards such as the US Department of State Specification SD-STD-02.01. Under these standards, tests are conducted in prescribed conditions that should be representative of the service installation. In actual project, conditions encountered on site may vary from the test environment and it would be expensive and time consuming to validate each deviation with a physical test. High-fidelity physics-based (HFPB) finite element modeling can provide precise simulations of the behavior of anti-ram bollards. This paper presents the use of HFPB finite element modeling, using LS-DYNA, in an actual project to evaluate the performance of an anti-ram bollard design subjected to various boundary conditions representing the physical conditions encountered on site. The study shows that boundary conditions can have a significant influence on the performance of the anti-ram bollards. This suggests that anti-ram bollards must be designed and engineered according to actual conditions that are found on site. It also shows that HFPB modeling can be an effective tool that supplements physical testing of anti-ram bollards.
