마이크로 그리드의 에너지 거래 애플리케이션을 위한 하이브리드 진화 알고리즘 기반 조합 경매

Abstract

With rapid rise of smart grids and microgrids, electricity trading has been extensively studied. Different kind of auctions have emerged as key frameworks for such trades. Therefore, efficient auction design for different electricity market applications has become an important research issue. Like the traditional auction mechanism, the aim of energy auctions is to maximize the overall auction efficiency. But one of the main problems of already implemented auctions in electricity markets is the exposure problem. Combinatorial auctions have shown to overcome the exposure problem in the auction mechanisms using the concept of package bidding. But their implementation has been quite difficult, and have not been studied in case of electricity trading. In this dissertation, we present the combinatorial auctions that can be used in volatile electricity markets in microgrids. While explaining the different bid types based on logical bidding languages, winner determination process and winner price determination processes are also explained. Winner determination process of combinatorial auctions is considered to be NP-Hard and similar to 0-1 Knapsack problem. In this study, we have used evolutionary algorithms to solve the winner determination problem. Evolutionary algorithms have often been used as a tool to solve knapsack problem. An algorithm combining genetic algorithm and particle swarm optimization (PSO) is presented in this study. In the in the proposed algorithm, AUCGENPSO, the genetic operators (crossover and mutation) and PSO are used in the same iteration allowing the algorithm to converge at an optimal value. To show the implementation of the combinatorial auctions and proposed winner determination problem, we implement it on two scenarios. In the first scenario, multiple microgrids trade energy with each other. We consider a region consisting of multiple interconnected microgrids wherein given time t, some microgrids have excessive energy that they wish to sell, whereas other microgrids desire to buy additional energy to meet their local demands. In the second scenario, the proposed auction scheme is implemented in a microgrid market where multiple households share a common energy storage system. The users bid to get storage capacity in the common energy storage system. Using Matlab, performance evaluation and stability tests of the proposed auction technique are performed and compared with the more commonly used trading techniques in microgrid markets. The results show that the proposed auction mechanism works better with better economic efficiency and is stable.