一种求解最优潮流的组合算法

提出了一种基于现代内点（MIP）理论与退火选择遗传算法（AGA）的组合算法：将原总是去掉整数变量约束，形成一个非线性规划问题；通过赋予整数变量矢量不同的初值，形成一个非线性规划问题集合，将其看作是AGA的进化种群，以MIP求出每一个非线性规划问题的最优值作为它的适应值，通过AGA试探，找出最优个体，该个体整数变量和连续变量的取值即为原问题最优解中各变量的值，AGA与MIP二者取长补短既能精确处理整数变量，改善计算结果的质量，又保证了算法的计算速度；对AGA的改进提高了算法的收敛性能，增强了逃脱局部极值的能力。通过对IEEE 14－118节点系统的仿真计算验证了所提算法的有效性。

A COMBINED ALGORITHM FOR OPTIMAL POWER FLOW

In this paper, we propose a new combined algorithm based on modern interior point (MIP)theory and genetic algorithm with annealing selection(AGA). At first, we get rid of integer constrains to transform the primal OPF problem to a normal nonlinear programming. Secondly, by setting integer variable vectors with different values, we can get a set of nonlinear programming sub-problems, this set is the evolvement population of AGA. We regard the optimum of each sub-problem calculated by MIP as its fitness. Finally, AGA iteratively improves all the values of integer vectors until the optimum is found, the setting of this optimum is the answer of primal OPF. Using MIP to deal with continuous variables can guarantee the calculation speed of new algorithm, at the same time using AGA to treat discrete variables can improve the global optimal ability. Numerical simulations on IEEE standard test systems ranging from 14 to 118 buses have shown that the proposed method is efficient in solving OPF problems for large-scale power systems.