基于遗传编程的绝缘内部局部放电缺陷模式识别

APPLICATION OF GENETIC PROGRAMMING FOR PARTIAL DISCHARGE PATTERN RECOGNITION OF INSULATION INSIDE DEFECT MODELS

采用新的模拟进化技术——遗传编程,进行局部放电模式识别以区分不同的绝缘内部缺陷类型。制作了4种结构的人工缺陷模型以模拟发电机定子中典型的绝缘内部放电,从局部放电试验中获得二维和三维谱图特征,计算局放信号的矩特征值。首先用模糊方法将局部放电信号的矩特征表示为关于对象不确定知识的模糊特征, 作为放电数据的预处理。再由遗传编程分类表达式进化生成局部放 电缺陷类型判别函数, 并采用递增式学习规则以提高最佳特征对局部放电缺陷分类的效果。另外,将Bootstrap统计模拟技术与遗传编程结合,以克服从小样本数据中进行知识获取的“瓶颈”。人工缺陷模型试验数据的测试结果表明,该方法在局部放电缺陷类型识别中得到了良好的识别效果。[( )-263.2( )]

In this paper, a new method based on genetic programming (GP) is applied to the automatic pattern recognition of partial discharge (PD) for insulation defect models. Four types of artificial defect models are used to simulate typical inside discharges in generator stator winding. The PD experiments are performed in order to obtain the moment features calculated from the 2-D and 3-D charts. Attributes related with the insulation defects are first transformed into fuzzy attributes, and then a set of PD defect discriminant functions are evolved based on incremental learning strategy and a distance-based fitness function for partial discharge source classification. In addition, in order to eliminate the bottleneck of insufficient sample size, a kind of statistical resampling technique called bootstrap is incorporated as a preprocessing step into genetic programming. The experimental results show the good ability in automatic pattern recognition of partial discharge for insulation defects.[( )-263.2( )-263.2( )]