A multi-agent optimal bidding strategy in microgrids based on artificial immune system. (15th December 2019)

• Record Type: Journal Article
• Authors: Kong, Xiangyu; Liu, Dehong; Xiao, Jie; Wang, Chengshan
• Is Part Of: Energy. Volume 189(2019)
• Publication Date: 2019-12-15
• Subjects: Microgrid -- Distributed energy -- AIS -- Bidding -- Power market; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2019.116154
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: The utilization of distributed energy resources (DERs) is growing worldwide, and the commercial prospects of microgrids (MGs) are clear. To optimally coordinate the power outputs of DERs owned by different owners while considering uncertainties in the commercial MG, a multi-agent optimal bidding strategy based on the artificial immune system (AIS) is proposed. The method takes the multi-agent system (MAS) control structure of the MG, distributing the profits of different owners through the market mechanism to conduct the optimization. A novel AIS is established and integrated into the MAS to help DERs participate in the optimal bidding operation of MG. The antigen is transformed by the environmental information, the price of the main grid, other DERs' bidding strategies, and the predicted deviation coefficient while accounting for the uncertainties of DER facilities, which is solved by AIS to find the optimal bidding strategy. A mixed-integer programming model is solved by the bidding manager agent to get bidding results, which are fed back to the DERs to help them form the next round of strategies until the result reaches the equilibrium. Results show that the proposed method is efficient in coordinating the power generation with uncertainty and maximizing the interests of each investor. Highlights: Multi-agent bidding and trading mechanism of DERs in microgrid was established. The method realizes the efficient and fair profit distribution by the market mechanism. The method provides an effective interaction mode between the MG and the main grid. An artificial immune algorithm is improved to search for the optimal strategy. Gene database is used for memory and evolution of good solutions to quickly respond.