A multi-agent system approach for optimal microgrid expansion planning under uncertainty

• Published in: International journal of electrical power & energy systems 2019-07, Vol.109, p.696-709
• Main Authors: Mishra, Sambeet, Bordin, Chiara, Tomasgard, Asgeir, Palu, Ivo
• Format: Article
• Language: English
• Subjects: Coordinated decision making; Generation and transmission expansion planning; Integrated uncertainty; Math-heuristic technique; Mathematical optimization; Microgrid; Multi agent systems
• ISSN: 0142-0615; 1879-3517
• DOI: 10.1016/j.ijepes.2019.01.044

•A novel peer to peer decision-making mechanism for capacity expansion is introduced.•A two-stage stochastic math-heuristic formulation for capacity expansion is proposed.•A multi-agent system with agent configuration for coordination is developed.•An Evolutionary Vertical Sequencing protocol produces smart permutations of grids.•The advantage of coordinated decision making is validated over traditional approaches. Power system planning for modern distribution networks is undergoing a change because of distributed power generation and grid ancillary services. The vast electricity retail utility industry with many distribution network operators plans generation and transmission expansion planning for determining optimal investment decisions. This paper addresses this planning problem in a decentralized and distributed context. This paper introduces a coordinated decision making approach for optimal investments in generation and transmission expansion planning problems for distribution networks. The distribution networks are further classified into microgrids. Considering an agent as the functional information exchanging entity just like an energy meter with the objective to coordinate the expansion decisions of participants; a novel math-heuristic optimization model Coordinated Microgrid is presented. To simulate the coordination of information a multi-agent-system based coordinated decision making method is adopted and the value of coordination is investigated. The evolutionary vertical sequencing protocol, a heuristic method, is developed and implemented to simulate the coordination process among agents on the top level. The proposed protocol produces smart permutations of microgrids for coordination. On the bottom level, a two-stage chance-constrained stochastic MILP formulation for investment decisions with operational uncertainties is modeled. For market clearing a nodal-pricing scheme is adopted that maintains the Nash equilibrium among and across the microgrids for energy transactions. The proposed model is tested with consumption, network configuration data from three islands in west-coast of Norway. The models are solved to optimality and results lead to the observations that the value of coordination lies in profit increment of individual microgrid. The novel protocol proposed demonstrates an advantage of retrieving smart permutations from combinations of microgrids. In summary, CoMG is a novel expansion planning model for optimal investments in m