含有离散变量的无功优化问题的混合优化方法

A Hybrid Optimization Method for Reactive Power Optimization with Discrete Variables

提出一种混合优化方法求解含有离散变量的无功优化问题。该方法将遗传算法和传统优化方法分别运用于离散变量和连续变量的优化搜索,遗传算法对种群进行全局广度搜索,而对连续变量的传统优化则作为遗传算法的局部深度搜索,使遗传算法擅长处理离散变量和传统优化方法速度快、数值稳定性好的优势得到发扬。这种混合优化算法模型简单规范,其实用性和有效性通过算例验证。

A hybrid optimization method is proposed to resolve the issue of reactive power optimization with discrete variables. The method adopts both the genetic and traditional optimization algorithms, and the genetic algorithm is for global search to the population while the traditional algorithm is for local deep search in favour of the genegic method, so that the advantages of both the algorithms are carried forward, adept at dealing with discrete variables by the genetic as well as in high speed and good value-stability by the traditional. The model of hybrid optimization method is simple and normal, and its practicability and effectiveness are validated by case studies.