【摘要】 无功优化是电力系统运行中的一个典型难题,遗传算法具有线性时间复杂度和全局收敛的特点,正好适合于该问题的求解。文章首先对遗传算法进行了探讨,对其中的杂交算子作了较为深入的研究,提出了一种新型的启发式杂交算子,克服了传统算术杂交算子中经常发生的“种群早熟”问题。为了提高求解速度,结合无功优化问题和遗传算法的特点对其中潮流计算的运算精度进行了动态控制。通过对IEEE30节点测试算例的求解,证明了本文提出的改进遗传算法具有很强的全局寻优能力,求解速度比传统遗传算法快了近1倍。更多还原

【Abstract】 Reactive-power optimization is a hard problem in power system operation.GA has good characteristics in linear complexity and global optimization,which is suitable for Reactive-power optimization.This paper discusses GA firstly;makes deep consideration in cross over operator.Then the paper presents a heuristic cross over operator to overcome the ’premature problem’ which often happens in traditional GA.To accelerate GA used in Reactive-power optimization,we used dynamic error control strategy in power flow computation.Test case of IEEE 30 node systems shows that the operator has special ability in global optimization.The computation time of improved GA algorithm is about two times faster than the traditional one.更多还原