空气开关电弧电流密度分布的智能反演方法研究  

Investigations on Intelligent Inversion Methods of Current Density Distribution of Electrical Arc in Air

作　　者：张鹏飞[1] 张国钢[1] 杨博宇[1] 董金龙[1] 耿英三[1] 

机构地区：[1]西安交通大学电力设备电气绝缘国家重点实验室,陕西省西安市710049

出　　处：《中国电机工程学报》2013年第33期160-168,20,共9页Proceedings of the CSEE

基　　金：中央高校基本科研业务费专项资金资助项目(XJJ2011019)~~

摘　　要：作为一种非介入式测试手段,基于电磁逆问题求解的磁测试技术在开关电弧等离子体的试验研究中表现出了较好的应用前景。但逆问题的求解有其自身的特点和困难,如非线性、病态性、计算规模大等,因此,为了满足逆问题应用于实际的需要,须在克服病态性的基础上,寻求求解逆问题的高效算法。该文针对空气开关电弧电流密度分布反演重构电磁逆问题的数学模型,基于蚁群、粒子群、模拟退火等优化方法建立了3种智能反演算法,并在对比分析各算法性能的基础上,对适用性较强的基于单点搜索的模拟退火反演算法的关键参数进行了优化。实例分析表明,电弧电流密度分布智能反演方法有效而可靠,能够为电弧内部机理的研究提供一条有效的途径。As a non-intrusive means, magnetic testing technology based on inverse electromagnetic problem solution has shown a promising prospect in the experimental study of electric arc plasma. But the solution of the inverse problem has its own characteristics and difficulties, such as non-linear, ill-posedness, and large scale. So in order to meet the needs of inverse problem applied in practical engineering, it is necessary to explore effective algorithms for solving the inverse problem on the premise of overcoming the ill-posedness. In the paper, for the mathematical model, of inverse electromagnetic problem of current density distribution reconstruction of electric arc in air, three intelligent inversion algorithms were established based on ant colony, particle swarm and simulated annealing respectively. Meanwhile, after doing a comparative analysis of the performance of each algorithm, the key parameters of the SA inversion algorithm with strongest applicability were optimized. Results of case analysis illustrate that the proposed intelligent inversion method for the arc current density distribution reconstruction is effective and reliable, which can provide a useful way for the research of the arc internal mechanism.

关 键 词：开关电器 空气电弧 电流密度分布 智能反演 

分 类 号：TM564[电气工程—电器]