供电系统微电网电源选址定容仿真研究

Simulation Research on Locating and Sizing of Power Sources in Micro Grid of Power Supply System

分布式风电和光伏具有随机性和间歇性的特点,接入微电网后会加剧电压偏移和负荷波动,把储能放置在单一节点不能有效抑制电压偏移和负荷波动等潮流问题,因此需要对微电网电源进行合理的选址定容。由于选址定容问题求解的复杂性,基于数学方法仅考虑极值的传统算法不仅耗时长而且速度慢,影响最优解选取精度,因此,提出自适应变异多目标粒子群算法,针对经典粒子群算法求解过程中,惯性权重缺乏指导以及局部收敛等问题,提出自适应变异改进,并引入拥挤距离算子排序求解获得最优解集,采用逼近理想解排序(TOPSIS)选出微电源的最优接入方案。建立以电压偏差、负荷缺电率以及储能容量为目标的微电源优化模型,进行仿真研究。结果表明,自适应变异多目标粒子群算法有效避免了取到次优的电源优化配置结果,能够快速收敛到最优解,验证了所提算法的有效性和准确性。

The distributed wind power and photovohaic system are of randomness and intermittence, and the volt- age offset and load fluctuations can be heightened after accessing to micro grid power. To put the stored energy on a single node cannot effectively suppress voltage offset and load fluctuation. Consequently, we need the appropriate lo- cating and sizing of microgrid power supply. Due to the complexity of solving the locating and sizing problem, the tra- ditional algorithm based on mathematical method considers only the extremum, which needs long time - consuming and affects the selection accuracy of optimal solution. Therefore, this article focused on the multi - objective particle swarm optimization based on adaptive mutation. Because the lack of guidance of inertia weight and local convergence in the solving process of classic particle swarm algorithm, this article put forward an adaptive mutation improvement and introduced operator ordering of crowding distance to get the optimized solution set. Then, the method of tech- nique for order preference by similarity to an ideal solution （TOPSIS） was used to choose optimal admission scheme for micro power sources. Moreover, the optimization model of micro power supply with the purpose of voltage devia- tion, loss of load probability and energy storage capacity were established for the simulation research. The adaptive multi -objective mutation particle swarm optimization algorithm can effectively avoid the suboptimal result of power optimization configuration and quickly converge to the optimal solution. The effectiveness and accuracy of proposed algorithm are verified.