华南理工大学学报(自然科学版) ›› 2012, Vol. 40 ›› Issue (11): 94-100.

• 动力与电气工程 • 上一篇    下一篇

高功率因数LCC谐振电路的多目标优化

张治国1 谢运祥1 袁兆梅2   

  1. 1.华南理工大学 电力学院,广东 广州 510640; 2.福建龙净环保股份有限公司,福建 龙岩 364000
  • 收稿日期:2011-08-05 修回日期:2011-12-14 出版日期:2012-11-25 发布日期:2012-10-01
  • 通信作者: 张治国(1977-) ,男,博士生,主要从事高频谐振变换器建模与控制研究. E-mail:zhang1569@126.com
  • 作者简介:张治国(1977-) ,男,博士生,主要从事高频谐振变换器建模与控制研究.
  • 基金资助:

    国家火炬计划项目( 2007GH060026)

Multi-Objective Optimization of LCC Resonant Circuit with High Power Factor

Zhang Zhi-guoXie Yun-xiangYuan Zhao-mei2   

  1. 1. School of Electric Power,South China University of Technology,Guangzhou 510640,Guangdong,China; 2. Fujian Longking Co.,Ltd.,Longyan 364000,Fujian,China
  • Received:2011-08-05 Revised:2011-12-14 Online:2012-11-25 Published:2012-10-01
  • Contact: 张治国(1977-) ,男,博士生,主要从事高频谐振变换器建模与控制研究. E-mail:zhang1569@126.com
  • About author:张治国(1977-) ,男,博士生,主要从事高频谐振变换器建模与控制研究.
  • Supported by:

    国家火炬计划项目( 2007GH060026)

摘要: 为了优化具电容型滤波器的LCC 串并联谐振变换器的输入/输出性能,首先,基于傅里叶级数的动态相量法建立了LCC 谐振电路的稳态数学模型,得到了等价电路.然后,在等价电路的基础上得到了反映变换器输入/输出性能的数学表达式.最后,提出一种基于Pareto 的多目标遗传算法来对变换器的输入/输出性能进行优化.该算法采用包含密度信息的细粒度赋值策略,通过最小化目标函数获得了一组均匀分布的最优解集,可用于指导电路设计.仿真和实验结果验证了所提出的优化方法的正确性.

关键词: 串并联谐振变换器, 动态相量模型, 等价电路, 遗传算法, 多目标优化

Abstract:

In order to improve the input /output performance of the series-parallel LCC resonant converter with capacitive output filter,first,a steady-state mathematical model of the LCC resonant circuit is established by using the dynamic phasor based on the Fourier series,and the corresponding equivalent circuit is obtained. Then,the mathematical expressions to describe the input /output performance are deduced based on the equivalent circuit. Finally,a Pareto-based multi-objective genetic algorithm is employed to optimize the input /output performance. By adopting the fine-grained fitness assignment strategy with density estimation and by minimizing the objective functions,the algorithm helps to achieve massive and well-distributed optimal solutions for circuit design. Simulated and experimental results indicate that the proposed optimization method is correct,

Key words: series-parallel resonant converter, dynamic phasor model, equivalent circuit, genetic algorithm, multi-objective optimization