Research On MAS Based Distributed Cooperative Power Optimal Control For Islanded Microgrid

Islanded microgrid,as an important part of smart distribution system,has already been successfully used in the areas far away from the centers of power supply,such as rural areas,islands,boards,ships and so on.The islanded microgrid can offer power supply to local demands using local renewable sources.In this way,the problem of lacking power supply in remote areas can be solved effectively.Different from the power system,the islanded microgrid consists of massive power electronic devices with small equivalent inertia,and large amounts of intermittent renewable energies.How to keep the power balance and stability of voltage has become the focus of current researches.In addition,the optimal power allocation of islanded microgrid should be considered as well,reasonable power flow can improve renewable energy utilization,reduce operation costs and improve overall efficiency.Therefore,an effective,reliable and scalable strategy for power optimization of islanded microgrid is imperative.However,existing researches are mainly focus on centralized,decentralized or hierarchical distributed control method of islanded microgrid power optimization problems,which have some disadvantages.Based on the three-layer control architecture in microgrid,the distributed control framework for islanded microgrid is designed,and the MAS-based distributed power optimal control method is proposed.Furthermore,the distributed information interaction method and cooperative optimal control strategy based on MAS are studied by the way of modeling,simulation and implementation.Research achievements of this study preliminarily establish the framework of distributed control theory and method for islanded microgrid.The main research contents are as follows:1)Distributed control architecture and distributed power optimal control method for islanded microgrid:Referring to the centralized/decentralizcd control architecture,the distributed control architecture for islanded microgrid is proposed.The management and communication mechanism complying with the international organization FIPA's standards are introduced for MAS,then,the distributed power optimal control method is proposed.Furthermore,the self-organization,distributed collaborative optimization and data interaction between two levels are realized by using communication and search services provided by MAS.2)Discrete consensus algorithm based distributed power proportionate control:In order to realize the adaptive power proportionate control,a distributed power proportionate control strategy based on discrete consensus algorithm is proposed according to the distributed power optimization control method.Designing system state transition matrix according to network structure,the global average voltage can be acquired based on local information interactions.The distributed power consensus control based on the average voltage is built to realize the adaptive power proportionate control and secondary voltage recovery.3)Incremental cost consensus based distributed optimal economic control:For achieving the optimal economic dispatch for islanded microgrid,an incremental cost consensus based distributed optimal control strategy is proposed,which contains incremental cost consensus control module and secondary voltage recovery module respectively.The optimal economic dispatch can be achieved by consensus control equations based on incremental cost,and the global voltage recovery can be realized based on distributed information interactions at the same moment.4)Distributed secondary optimal control for islanded AC microgrid:In order to solve the coupling problem of active and reactive power,a distributed secondary control strategy based on MAS is proposed according to the distributed control method.The global voltage information can be discovered through average consensus algorithm which is used to recover overall average voltage.Moreover,the proposed power consensus control strategy based on MAS is applied to realize the adaptive power proportionate control.5)The performance of the proposed distributed control strategy and the optimization of communication topology:The impacts of communication delay on the proposed strategies are analyzed at length.Taking into account the economic fact and the impact of time delay,a multi-objective optimization model for evaluating the performance of communication network is formulated.The Pareto points of the model are searched by no-domination sort method.Then,the optimal network topology that satisfies the customer's demand is obtained through a suitable evaluation function.Besides,an agent self-organization strategy is proposed to realize "plug and play"capability for the distributed strategy.