At present,the direct integration of renewable energy into the power grid still faces stability and economic problems,and the impact on the power grid can be mitigated by virtual power plant integration.This paper takes the minimum total operating cost after system integration as the objective for predicting new energy output and load in the next 24 hours.It also considers the changes in electricity prices on the grid side at different times and uses the chaotic mapping adaptive particle swarm optimization algorithm based on opposition-based learning to discuss five scheduling schemes with different combinations of wind,photovoltaic,energy storage and demand response.Compared with the original particle swarm algorithm,this algorithm can jump out of local optimal solution and find out the global optimal solution.The calculation results show that the operating cost can be reduced by 4.47％when both wind,photovoltaic and energy storage and the demand response are involved in power supply,as well as improve user comfort by 3.51％compared with only wind photovoltaic and energy storage participating in power supply.
关键词
虚拟电厂/风-光-储/需求响应/经济调度/反向学习的混沌映射自适应粒子群算法
Key words
virtual power plant/wind-photovoltaic-energy storage/demand response/economic scheduling/chaotic mapping adaptive particle swarm optimization algorithm based on opposition-based learning
维普
万方数据