摘要
无功优化是一个混合整数优化问题,为了提高算法性能,提出使用遗传算法中的锦标赛选择机制。在已有的蚁群算法基础上,通过引入基于编码的蚁群算法,对不同类型的控制量使用不同的编码长度,能同时处理连续和离散空间的变量,再引入遗传算法中的锦标赛选择机制,锦标赛选择对适应值的非线性变换具有不变性,能更稳定地控制选择压力,即使陷入局部最优也具有很高的逃离概率。通过限制信息素上限也能有效防止信息素累积过多。使用IEEE-14、IEEE-30和IEEE-57系统对算法进行测试,验证了以上改进的有效性。
Reactive power optimization is a mixed integer programming problem.A tournament based selection method from genetic algorithm is proposed to improve the performance of the algorithm.On the basis of the existing ant colony algorithm,an ant colony algorithm based on coding technology is introduced.Different length of code is employed to encode different types of control varibles.In this way both continuous and discrete varibles can be handled.A tournament based selection method from Genetic Algorithm is reintroduced as the tournament based selection is invariant to nonlinear transform of fitness values,the selection pressure can be easily adjusted and the probability of escaping from local optimum is improved.A limitation strategy of pheromone value is implemented to prevent the values from over-increasing.The proposed algorithm is tested under IEEE-14,IEEE-30 and IEEE-57 systems to show the efficiency of the improvements.
出处
《电力系统保护与控制》
EI
CSCD
北大核心
2010年第12期80-85,共6页
Power System Protection and Control
关键词
电力系统
无功优化
蚁群算法
锦标赛选择
混合整数规划
power system
reactive power optimization
ant colony algorithm
tournament selection
mixed integer programming
