EPE Association: EPE 2020 - DS3n: Data Analysis, Artificial Intelligence, and Communication Issues

EPE 2020 - DS3n: Data Analysis, Artificial Intelligence, and Communication Issues
You are here: EPE Documents > 01 - EPE & EPE ECCE Conference Proceedings > EPE 2020 ECCE Europe - Conference > EPE 2020 - Topic 10: Data Analysis, Artificual Intelligence and Communication Issues > EPE 2020 - DS3n: Data Analysis, Artificial Intelligence, and Communication Issues

   [return to parent folder]

   Energy Management for isolated renewable-powered microgrid using reinforcement learning and game theory
By Rui HU [View]
[Download]

Abstract: This paper presents a decentralized energy management system (EMS) solving for optimal load-response strategy applying reinforcement learning (RL) and game theory for islanded renewable-powered microgrids. The EMS enables the consumers in a microgrid to independently evaluate the tradeoff between satisfying load demand and maintaining sufficient stored energy to make load-response decisions correspondingly. The evaluation and decision-making process consists of two parts: an instant virtual two-player load-response game and a long-term linear-reward inaction (LR-I) learning process adjusting consumer power/load models. The virtual two-game solving process is an instantaneous decision-making system so that the consumers could make real-time decisions, while the LR-I process gradually improves the consumer payoff based on the system feedback during the operation. Simulation of a microgrid powered by PV cells and battery banks is conducted to evaluate the EMS performance. It is shown that the game-learning EMS has a better performance compared to both the direct virtual two-player game and the naive LR-I approach. Additionally, compared to the naive LR-I approach, the proposed game-learning algorithm has a faster converging-speed.

   Investigation of Bond Wire Lift-Off by Analyzing the Controller Output Voltage Harmonics for the Purpose of Condition Monitoring
By Firat YÃœCE [View]
[Download]

Abstract: This paper presents a new approach in the field of condition monitoring of power semiconductors in power converters. The new approach avoids additional sensors and uses only the data that is already available in a power electronic system. This results in advantages such as saving extra costs and eliminating potential sources of failure. In this paper, the aging mechanism of bond wire lift-off is investigated. First, it is shown by simulations that output voltage harmonics of the current controller contain information about the bond wire lift-off aging mechanism. For testing the new approach in practice, real data are recorded in a test bench for a power module with unharmed bond-wires in order to train an algorithm for normal behavior. Then the aging mechanism bond wire lift-off is induced by cutting off bond wires of a power module with the aim of testing the functionality of the algorithm. The results of this investigation show that the proposed approach is able to detect the anomaly in the data set.

EPE 2020 - DS3n: Data Analysis, Artificial Intelligence, and Communication Issues
You are here: EPE Documents > 01 - EPE & EPE ECCE Conference Proceedings > EPE 2020 ECCE Europe - Conference > EPE 2020 - Topic 10: Data Analysis, Artificual Intelligence and Communication Issues > EPE 2020 - DS3n: Data Analysis, Artificial Intelligence, and Communication Issues
	[return to parent folder]

	Energy Management for isolated renewable-powered microgrid using reinforcement learning and game theory By Rui HU	[View] [Download]
Abstract: This paper presents a decentralized energy management system (EMS) solving for optimal load-response strategy applying reinforcement learning (RL) and game theory for islanded renewable-powered microgrids. The EMS enables the consumers in a microgrid to independently evaluate the tradeoff between satisfying load demand and maintaining sufficient stored energy to make load-response decisions correspondingly. The evaluation and decision-making process consists of two parts: an instant virtual two-player load-response game and a long-term linear-reward inaction (LR-I) learning process adjusting consumer power/load models. The virtual two-game solving process is an instantaneous decision-making system so that the consumers could make real-time decisions, while the LR-I process gradually improves the consumer payoff based on the system feedback during the operation. Simulation of a microgrid powered by PV cells and battery banks is conducted to evaluate the EMS performance. It is shown that the game-learning EMS has a better performance compared to both the direct virtual two-player game and the naive LR-I approach. Additionally, compared to the naive LR-I approach, the proposed game-learning algorithm has a faster converging-speed.

	Investigation of Bond Wire Lift-Off by Analyzing the Controller Output Voltage Harmonics for the Purpose of Condition Monitoring By Firat YÃœCE	[View] [Download]
Abstract: This paper presents a new approach in the field of condition monitoring of power semiconductors in power converters. The new approach avoids additional sensors and uses only the data that is already available in a power electronic system. This results in advantages such as saving extra costs and eliminating potential sources of failure. In this paper, the aging mechanism of bond wire lift-off is investigated. First, it is shown by simulations that output voltage harmonics of the current controller contain information about the bond wire lift-off aging mechanism. For testing the new approach in practice, real data are recorded in a test bench for a power module with unharmed bond-wires in order to train an algorithm for normal behavior. Then the aging mechanism bond wire lift-off is induced by cutting off bond wires of a power module with the aim of testing the functionality of the algorithm. The results of this investigation show that the proposed approach is able to detect the anomaly in the data set.