论文部分内容阅读
针对传统进化规划方法寻优效率低的问题,提出了配电网络电容器优化投切的制约进化策略。利用配电网络物理规则求得投切点无功补偿容量调节范围,用于约束解的产生和进化,缩小寻优空间。通过在解空间中心点附近合理配置初始点,避免初始解不可行。采用小种群多代进化方式,避免求解过程产生大量不可行解。最后选择适合电容器优化投切问题的随机动态步进方式,在寻优的不同阶段,采用不同的步长,使优化效果和效率得到明显改善。使用制约进化策略对不同的配电系统进行分析,结果表明:制约进化策略在保证优化精度的同时,具有更高的优化效率,可以适应在线应用。
Aiming at the problem of low efficiency in traditional evolutionary programming, this paper proposes a constrained evolution strategy for optimal switching of distribution network capacitors. Using the physical rules of distribution network, the adjustment range of reactive power compensation capacity for switching point is obtained, which is used to generate and evolve the constraint solution and reduce the space for optimization. By reasonably configuring the initial point near the center of the solution space, it is not feasible to avoid the initial solution. Multi-generational evolution using small populations avoids a large number of infeasible solutions to the solution process. In the end, we choose the stochastic dynamic stepping method which is suitable for the optimization of capacitor switching. At different stages of optimization, different steps are used to improve the optimization effect and efficiency remarkably. Using constrained evolutionary strategy to analyze different power distribution systems, the results show that the constrained evolution strategy has higher optimization efficiency and can adapt to online applications while ensuring the optimization accuracy.