摘要
将一种改进的遗传操作与人工神经网络相结合的混合算法应用于电力变压器的故障诊断,有效地解决了常规BP算法易陷入局部极小、收敛速度慢和基本遗传算法早熟等缺点。实例仿真结果表明,该算法具有较快的收敛速度和较高的计算精度,满足电力变压器故障诊断的要求。
In this paper, a hybrid algorithm which combines improved genetic manipulation with artificial network is applied to fault diagnosis of power transformer. It overcomes the .shortcomings of entrapment in local optimum, slow convergence of ordinary BP algorithm and the premature convergence of basic GA effectively. Simulation diagnosis results of practical examples show that the convergence rate of the proposed method is faster and the computational result is more accurate than others, the proposed method fulfils the requirements of fault diagnosis of power transformer.
出处
《高电压技术》
EI
CAS
CSCD
北大核心
2003年第7期3-6,共4页
High Voltage Engineering
基金
福建省教育厅基金(编号:JB01035)