基于遗传算法的配电网无功补偿优化论文.doc

毕业设计说明书 基于遗传算法的配电网无功补偿优化 学生姓名： 班级学号： 院、系、部： 专 业： 指导教师： 合作指导教师： Undergraduate Thesis Reactive Power Optimization of Distribution Network Based on genetic algorithm  摘要 配电网进行合理的无功补偿,能够有效地维持系统的电压水平,降低有功网损, 提高网络输送容量,节约发电成本。而且由于我国配电网长期以来无功匮乏,尤其造成的网损较大,因此无功功率补偿是投资少回报高的方案。 目前无功补偿算法非常多，有传统的数学规划法，如线性规划法、非线性规划法、混合整法、动态规划法等及围绕这些算法的一些改进，也有近几年刚兴起的人工智能方法，如遗传算法，模拟退火法，模糊集合理论、功率矩法、粒子群优化算法等，其中遗传算法可以通过多路径进行全局搜索，以避免收敛于局部最优点。本文用遗传算法解决配电网无功补偿优化问题，将年运行费用最小为目标函数，无功补偿源节点注入无功为控制变量，以节点电压为状态变量，建立无功优化模型，在各种负荷下找出电容的补偿位置和容量。 关键词 配电网系统，遗传算法，无功优化，无功补偿  Abstract Reasonable reactive power compensation methods in distribution networks can keep the systems voltage profile,reduce the active power losses of networks,increase the feeding capabilities of networks,and cut down the generating cost.the scarcity reactive of distribution network for a long time in our country,especially caused by network loss greatly.So reactive power compensation is the scheme which investment less and return higher. At present,the algorithms of reactive power compensation have so many,such as traditional mathematical programming method,linear programming method,nonlinear programming method,mixed whole method,dynamic programming method and some methods based on the improvement of these algorithms.Recent years,the artificial intelligence methods recently sprong up,for example,genetic algorithm,simulated annealing,fuzzy set theory,power moment,particle swarm optimization and so on.GA can search the global by multipath and avoid local optimal result.In this paper,a method of GA to solve optimal reactive power compensation of distribution network.The mathematic models for reactive power optimization,in which annual operation expenses is taken as the objective function,the node reactiv