摘要
提出了一种基于双种群粒子群算法的分时段无功优化算法,将复杂的动态无功优化问题转化为静态无功优化进行处理。文中把设备的调节代价和网损费用之和作为目标函数,考虑了包括设备相邻时段调节次数限制等约束条件。在该模型基础上,根据负荷的变化趋势,提出了一种考虑设备一天内调节次数限制的分时段算法。该方法简便,不需设置划分时段的门槛值,保证了分段的有效性和可操作性,且采用双种群粒子群算法对每个时段进行静态无功优化。该算法对离散变量进行了特殊编码,较好地解决了连续和离散变量的共同寻优,降低了网损,并减少了设备动作次数。IEEE30节点算例系统结果验证了所提算法的正确性和有效性。
Based on double-population particle swarm optimization （PSO）, a time-interval based reactive power optimization algorithm, by which complicated dynamic reactive power optimization is transformed into static reactive power optimization, is proposed. In this paper the sum of control cost and network loss cost is taken as objective function while the restriction of regulating times of control devices in adjacent time-intervals is brought into constraint conditions. On the basis of the proposed mathematical model and according to the variation trend of load, a time-interval based algorithm that considers the restriction of regulating times in one day is put forward. The time-interval based method is simple and convenient and has no need of setting the threshold for time-interval division, thus the effectiveness and operability of time-intervals can be ensured. The proposed algorithm performs static reactive power optimization in each time-interval by double-population PSO and discrete variables are specially coded, therefore the cooperative searching of discrete variables and continuous variables, namely daily regulating times and network loss, can be implemented, and the network loss and regulating times of devices can be reduced. Calculation results of IEEE-30bus system show that the proposed algorithm is correct and effective.
出处
《电网技术》
EI
CSCD
北大核心
2009年第8期72-77,共6页
Power System Technology
关键词
聚类
分时段无功优化
调节代价
双种群粒子群算法
cluster
time-interval based reactive poweroptimization
control cost
double-population particle swarmoptimization algorithm
