摘要
以系统节点电压波动、负荷波动以及储能系统总容量为目标建立了储能选址定容优化模型。求解过程中提出了一种改进多目标粒子群算法(improved multi-objective particle swarm optimizer,IMOPSO)。该算法根据粒子与种群最优粒子的距离来指导惯性权重的取值,使得各粒子的惯性权重可以自适应调整,并在二者距离较小时引入交叉变异操作,避免陷入局部最优解,同时采用动态密集距离排序来更新非劣解集并指导种群全局最优解的选取,在保持解集规模的同时使解的分布更均匀。为避免决策者偏好对最终结果的影响,采用基于信息熵的序数偏好法从最优Pareto解集中选取储能的最优接入方案。以IEEE-33节点配电系统为例进行仿真验证,结果表明该方法在储能选址定容问题求解中具有很好的收敛性以及全局搜索能力。
An optimization model for energy storage locating and sizing was established. It was based on a fully consideration of the voltage fluctuations of system node, load fluctuation, and the total capacity of energy storage system. And an improved multi-objective particle swarm optimizer （IMOPSO） was proposed. In the algorithm, the value of inertia weight was directed by the distance between the particle and the global optimal particle, crossover and mutation operations were introduced to avoid falling into local optimal solution. Also, dynamic crowding distance was used to update the Pareto solution set and guide the selection of global optimal solution, making solution more evenly distributed while maintaining the size of the solution set. To avoid the influence on the final result from decision makers, the TOPSIS （technique for order performance by similarity to ideal solution） based on the entropy was adopted to choose the optimal solution from Pareto solution set. The proposed algorithm has been applied to IEEE-33 nodes distribution system. Simulation results show that the method has rapid convergence speed and superb global search ability.
出处
《电网技术》
EI
CSCD
北大核心
2014年第12期3405-3411,共7页
Power System Technology
基金
国家863高技术基金项目(2014AA052004)~~
关键词
配电网
储能
多目标粒子群算法
自适应惯性权
重
PARETO解
distribution system
energy storage
multi-objective particle swarm optimizer
adaptive inertia weight
Pareto solution
