| With the large-scale and clustered construction and development of microgrids,the regional microgrid clusters composed of multiple independent microgrids in a certain area can realize the spatial complementarity and friendly interaction of energy through mutual assistance of electric energy,and improve the power supply reliability of the system.However,the problem of multi-microgrid optimization is more complex than that of single microgrid.How to ensure the stability,economy and environmental protection of the optimal operation in microgrid clusters system is an urgent problem to be solved in the research field of multi-microgrid.So,taking a regional microgrid clusters as the object,this paper discusses and researches the optimal scheduling strategy and cooperative operation.The main contents are as follows:(1)Aiming at the composition and structure of microgrid clusters system,the mathematical models of micro-sources such as photovoltaic power generation unit,wind power generation unit,diesel generator unit,energy storage battery unit and load under dynamic electricity price are established.Meanwhile,research the basic working principle and operating characteristics.(2)In order to provide accurate data support for the collaborative optimal scheduling of regional microgrid clusters and improve the utilization rate of renewable energy,a method of bidirectional long-short term memory neural network based on bayesian optimization and attention mechanism(BO-AM-Bi-LSTM)is proposed to accurately predict the data of wind power,photovoltaic and load power respectively.Through the simulation analysis of an example,the accuracy of the prediction method is proved.(3)The double-layer optimized framework of regional microgrid clusters is constructed.The upper layer takes the minimum fluctuating power of tie line and the comprehensive cost of the system as the goals,and takes the power transmission restriction between microgrid clusters and distribution network and the power balance of the system as the constraints;In the lower layer,aiming at the lowest average operation cost of the microgrid and the highest proportion of scenery consumption,and taking micro-source output,climbing limit and power balance of single microgrid system as constraints,the upper and lower multi-objective optimization models are established respectively,and a chaotic multi-objective particle swarm optimization algorithm based on niche technology(NC-MOPSO)is proposed,which optimizes and solves the multi-objective optimization problems respectively,to obtain the best scheme for the stable,economic and environmental protection operation of the regional microgrid clusters system.Through the simulation analysis of an example,the superiority of the double-layer and multi-objective optimization strategy of microgrid clusters is proved.(4)Aiming at the shortage or surplus state of microgrid power in the system,three optimal operation scheduling strategies are proposed.The generation of controllable units,the charging and discharging of the batteries,the power interaction between microgrids and the energy mutual assistance between single microgrid and distribution network under different dispatching strategies are analyzed.The feasibility of the proposed scheduling strategy is proved by example simulation. |
