摘要
配电电网装设无源滤波器以补偿无功和抑制谐波的设计中,滤波器的LC参数是一个多约束、非线性的规划问题,其参数优化最为关键。针对目前现有优化设计方法中,假设条件较多、寻优能力不强、寻优速度较慢等问题,利用改进遗传算法的全局寻优能力,提出了一种新的无源滤波器设计方法,即无源滤波器的成本、无功补偿和滤波效果3个目标全局优化。通过适应度函数的阈值制约以及以不同概率进行染色体选择操作,使得种群向3个目标最佳协调点进化;通过混沌算子解决早熟收敛的问题。工程算例的仿真计算表明了应用遗传算法设计滤波方案的优越性和对遗传算法改进的有效性。
The passive power filter is applied in power distribution network to suppress the network harmonic and compensate the reactive power,the optimization of LC parameter becomes important in the filter design.A new design method for passive power filter based on genetic algorithm(GA) is adopted,which macroscopically optimizes the cost,harmonic suppression and the effect of filtering as three objectives.By means of threshold limitation and chromosome selection with different probabilities,the three objectives are optimized.A chaos operator is introduced to overcome the premature convergence.The results of test and simulation in a real project reveal its superiority and effectiveness.
出处
《高电压技术》
EI
CAS
CSCD
北大核心
2007年第12期177-182,共6页
High Voltage Engineering
基金
湖南省自然科学基金资助项目(03YJJ5005)
湖南省教育厅科学研究资助项目(05C002)。~~
关键词
无源滤波器
改进遗传算法
多目标
设计
优化
混沌算子
适应度
passive power filter
refined genetic algorithm
multi-objective
design
optimization
chaos operator
degree of adaptability