Decentralized Optimal Scheduling For Multi-micro Networks Considering Multiple Uncertainties

At present,with the rapid consumption of fossil fuels and increasingly serious air pollution,clean energy has gradually become the focus of power related research.Microgrid can integrate a variety of distributed power generation,with high initiative,strong flexibility,safety and reliability,and other advantages,but also enhance the power quality of the power system.Therefore,more and more scholars pay attention to the optimization of multi-micro grid scheduling.In this paper,the decentralized economic dispatching of multi-micro grid considering multiple uncertainties is studied.The specific contents are as follows:Firstly,the concept of multi-micro grid structure diagram and multi-micro grid partition in decentralized scheduling is introduced,and the decentralized scheduling model of multi-micro grid is established,the objective function is its total operating cost.The power generation principle of the common renewable energy of multi-micro grid is discussed,and the mathematical models of the power generation system such as wind-solar generator set,diesel generator set and battery energy storage system are built.The uncertain factors affecting the power and load prediction of renewable energy generation are analyzed respectively.The principle and common forms of the standard alternative direction multiplier method(ADMM)are introduced,which lays a foundation for the following research work.Then,the standard alternating direction multiplier method is improved to solve the decentralized scheduling model of multi-micro grid.The simulation proves that the decentralized scheduling model has high stability and can reliably converge to the global optimal solution consistent with the centralized optimization.Compared with the standard ADMM method,the improved alternating direction multiplier method can shorten the iteration time,and the information exchanged between each microgrid is relatively less,which realizes the decentralized autonomy of multi-microgrid scheduling.The feasibility and superiority of the improved ADMM algorithm to solve the decentralized economic scheduling model are verified.Finally,on the basis of the above research,the stochastic uncertainty of renewable energy power generation and user-side load is fully considered,and a multi-micro grid decentralized scheduling model is established in a robust optimization environment.The objective function adds a penalty charge for the randomness of wind-generating power.The Latin hypercube sampling method is used to transform the objective function into an effective objective function,and the improved algorithm is used to solve the problem.It is proved that although the robust optimization method considering multiple uncertainties increases the economic cost to a certain extent,it can better deal with various random uncertainties in the actual life of multi-micro grid scheduling.