基于协同进化法的电力系统无功优化

针对无功优化问题非线性、非连续性等特点以及大范围内无功优化控制变量较多的问题，提出基于协同进化的无功优化算法以及相应的求解步骤。协同进化算法借鉴分解协调的思想，将无功优化问题分解为一系列相互联系的子优化问题，每个子优化问题对应于进化算法的一个种群，各种群通过共同的系统模型相互作用，共同进化，从而使整个系统不断演进，最终达到问题求解的目的。与常规的遗传算法相比，协同进化算法不但能得到更好的优化结果，收敛性好，而且克服了普通遗传算法计算时间过长的缺点，算例结果表明，该算法更适合于求解大系统的无功优化问题。

REACTIVE POWER OPTIMIZATION BASED ON COOPERATIVE COEVOLUTIONARY APPROACH

Cooperative Coevolutionary Approach (CCA) is a new architecture of evolutionary computation. Based on CCA, the paper proposes a new method for power system reactive power optimization problem, which is non-convex, non-linear, discrete, and usually with a large number of control variables. According to the decomposition-coordination principle, the reactive power optimization problem is decomposed into a number of sub-problems, which is optimized by a single evolutionary algorithm population. The populations interact with each other through a common system model and coevolve. These means promote the continuous evolution of the whole system. The reactive power optimization problem is solved when the Coevolutionary process ends. Simulation results show that compared with conventional Genetic Algorithm, CCA not only can obtain better optimal results, but also have better convergence property. Furthermore, CCA obviously reduces the over-long computational time, and is more suitable for solving large-scale reactive power optimization problems.