分布式电源和微电网靠近用户侧，能够实现新能源发电的分散利用，同时提升用户供电可靠性，是智能配电网的重要组成部分。如何对它们进行合理有效的运行管理，是实现智能配电网的关键。配电网中的分布式电源及微电网有不同的运行目标，存在冲突与合作，其运行管理方式比较适合采用博弈论的理论和方法进行研究。本文基于博弈论研究分布式电源能否自发形成微电网，以及微电网之间能否主动进行交易或者形成合作，以期为分布式电源及微电网接入配电网的规划设计与运行管理提供依据。本文的主要工作如下：（1）建立了研究分布式电源运行模式的非合作博弈模型，通过研究连续多时段非合作博弈问题纳什均衡点的移动情况，判断分布式电源能否自发形成微电网。通过连续多时段纳什均衡点的整体分布情况，定量计算分布式电源利用率。研究成果可为政府制定微电网激励政策提供理论依据，并可用于预测微电网的商业化发展前景。（2）研究了微电网以各自经济收益最大化为目标的交易行为。建立了考虑配电网参与的多微电网交易的非合作博弈模型，证明了多微电网交易中纯策略纳什均衡点的存在性。仿真分析了微电网之间进行交易的边界条件，以及多种调控手段对交易的影响。研究成果可为多微电网交易系统的构建提供前期理论支持。（3）提出了结合可靠性指标分析微电网合作可能性的方法。建立了以减小停电损失为目标的微电网联盟型博弈模型，在此基础上提出了结合可靠性指标分析微电网合作概率的方法，并通过仿真进行了验证。研究成果可用于指导不同地区多微电网接入配电网的运行管理方式设计。（4）研究了多个微电网之间以降低停电风险为目标形成合作的条件。提出了微电网储能—负荷均匀的概念及数学判别式，证明了微电网储能—负荷不均匀是微电网以抵御风险为目标形成合作的充要条件，并进行了仿真验证。研究成果可为微电网及配电网考虑风险的规划和运行管理提供建议。本文基于博弈论提出了研究分布式电源及微电网运行模式的模型和方法，分析了分布式电源自发形成微电网的前提条件，并从经济收益、停电损失、停电风险三个维度对多微电网之间主动进行交易或者形成合作的条件进行了理论证明和仿真验证，相关成果为分布式电源、微电网接入配电网的规划设计以及政府政策制定等提供了理论依据，拓展了博弈论在电力系统中的应用。
Distributed generations (DGs) and microgrids (MGs) are located near customers, which can achieve decentralized applications of new energy generations and improve reliability of power supply. They are important components of smart distribution systems. Therefore, how to operate and manage them efficiently may affect the development of smart distribution systems. In a distribution system, different DGs and MGs have various operation goals, which may bring conflicts and cooperation in DG and MG operations. Game theory is a suitable tool to study the issues in this situation. Whether DG owners would integrate their DGs with local loads therefore form MGs and whether MGs would actively trade or cooperate with each other, are studied in this paper based on game-theoretic models and methods, in order to guide the planning, operation and management of DGs and MGs in distributed systems. The main contributions of this dissertation are listed as follows.(1)A non-cooperative game model is set up to estimate DG owners’ motivations of building MGs. An approach to study the preconditions of MG’s formation through analysis of Nash Equilibrium movements in continuous multi-period non-cooperative games is proposed. A method to quantitatively assess DG application extent through Nash Equilibrium distribution of all time slots is also put forward. The work could provide theoretical support for government policy making to stimulate MG formation, and predict developing tendency of commercial MGs.(2)Researches on energy trading behaviors of MGs to pursue economic benefits are conducted. A non-cooperative game model for multi-microgrids power trading considering distribution system participation is built, and the existence of pure-strategy Nash Equilibrium is proved mathematically. Simulations on boundary conditions which stimulate trading between MGs, and how generation cost and government incentive measures affect energy trading behaviors, are carried out. The work could guide the construction of multi-microgrids trading system.(3)A method to evaluate the possibility of cooperation among MGs according to a reliability index is put forward. A coalitional game model for multi-microgrids to reduce outage loss is proposed. An approach to analyze the MG cooperation possibility according to a reliability index is set up, and a criterion is proposed and proved mathematically. Simulation results verify the correctness and effectiveness of the proposed model and criterion. This work would help operation and management of multi-microgrids in different districts.(4)Necessary and sufficient conditions for MGs to cooperate in order to diminish outage risk are studied. The concept and algebraic expressions of storage-load-uniformity for MGs are put forward, and the relationship between this concept and MGs’ willingness of cooperation to withstand outage risk is revealed and proved mathematically; simulation results validate the conclusion. The work could give advice to planning, operation and management of MGs to enhance the ability of electric power system to withstand disasters.To summarize, game theoretic models and methods for studying DG and MG operation modes are proposed. Researches on the necessity of DGs’ forming MGs are conducted, and the conditions of MGs’ power trading or cooperation for improving economic benefits, reducing outage loss or diminishing outage risk are proposed and theoretically proved. The correctness of the conditions is verified by simulations. The work in this dissertation may support planning and design of distribution systems with DGs and MGs, instruct related policy making for the government, and extend game theory applications in electric power systems.
