发电机局部放电非线性模式识别现状及展望

局部放电是导致发电机定子绝缘劣化的重要因素,在线监测局部放电信号并采用非线性方法识别放电类型,能够及时发现绝缘内部局部缺陷及放电发展程度,防止事故发生。阐述了发电机局部放电产生的类型、特点、在线监测方法以及放电过程的非线性特征,介绍了几种非线性模式识别方法的构成原理与特点,包括基于人工神经网络、小波分析和分形理论的局部放电模式识别,并对近年来各非线性识别方法及其组合的主要研究成果进行了总结与评述。最后还对未来发电机局部放电模式识别的研究方向进行了展望,指出快捷方便的非线性识别方法依然是研究的重点。

Present Situation and Prospect of Non-linear Pattern Recognition on PD in Generator Stator

The Partial Discharge(PD) is considered as a very important factor which results in insulation deterioration of generator stator.So,monitoring partial discharge signals online and identifying discharge types with nonlinear methods could be used to find out internal partial defects and the relevant discharge development degree in time,and then prevents(equipment) from the coming faults.Several aspects concerning PD,such as type and characteristics of PD,online monitoring methods and nonlinear characteristics in PD process are taken in this paper.Three nonlinear pattern recognition methods are introduced in composing principle and peculiarity.They are the methods based on Artificial Neural Network(ANN),wavelet theory and fractal theory.Furthermore,the main research results of the above nonlinear pattern recognition methods and their combination in recent years are reviewed and discussed.Based on these discussions,the direction in the future research on pattern recognition based PD monitoring is looked ahead and it is pointed out that the easy and fast pattern recognition by(non)linear methods is still the focal point of the research.