基于改进双种群蚁群算法的无功优化研究

Reactive power optimization based on the improved dual population ant colony algorithm

针对电力系统无功优化多变量、多约束、非线性的特点,提出一种新的改进双种群蚁群算法。基本蚁群算法在众多优点之外也存在着搜索时间长,容易出现停滞等缺点。因此在基本蚁群算法的基础之上,引入双种群独立搜索,进行信息交流,较大概率的打破了单一蚁群搜索的停滞状态,保证了算法中解的多样性,提高了全局收敛能力。并在蚁群算法的信息素更新策略和参数上做出进一步的改进应用于无功优化。通过对IEEE30节点算例进行仿真计算以及与现有算法进行比较,验证算法的有效性。

According to the characteristics of multivariable and multiple constraints and also nonlinearity which reactive power optimization possesses, the improved dual population ant colony algorithm is presented. It is found that the ant colony algorithm has the shortcomings of a long time search and easy to stagnation and so on except of many advantages. Therefore, the dual-population algorithm is introduced on the basis of the basic ant colony algorithm to conduct independent search, information exchange, greater probability of breaking the stag- nation of a single ant colony search to ensure the diversity of the solutions in the algorithm and enhance the capability of global convergence. The improvements in the pheromone update strategy and parameters of the ant colony algorithm are applied to reactive power optimization. It verifies that the algorithm is effective by means of the simulation of IEEE30 node and the comparison of existing algorithms.