Synergetic Scheduling Optimization Method of Grid-Connected Home Wind-Solar-Storage System Considering Uncertainty Factors

Abstract

Considering uncertainty factors such as output of wind power and photovoltaics and load, and effect of real-time electricity price on optimal scheduling of home wind-solar-storage system, an optimal scheduling model of the system is put forward, taking into account the users' electricity sale revenue, electricity purchase cost, government subsidy, operation and maintenance cost and distributed power investment cost. An optimization method is proposed based on Latin hypercube sampling, scene reduction and self-learning difference algorithm to solve the proposed model. Firstly, a large number of samples are generated with Latin hypercube sampling method according to probability distribution of wind, solar light and load prediction, and the generated samples are reduced with scene reduction method. Secondly, a new method is proposed to solve the problem that "convergence" phenomenon possibly occurring in late stage of standard difference algorithm is prone to falling into local optimal situation. A self-learning difference algorithm combining self-learning selection operation and vertical crossover operation is proposed to obtain the best decision of each scene. Finally, simulation results for typical examples show that the proposed model and method are feasible and effective.