基于TVAR参数信号模型的反电晕放电检测

将小波神经网络引入时变参数信号模型中,提出一个基于小波神经网络的时变参数信号模型.使用该信号模型对非平稳的反电晕放电信号建模,通过模型参数提取信号的特征,根据提取的特征判别反电晕放电现象是否发生.对实际信号的建模实验结果表明,该参数信号模型在放电信号建模方面具有优良的性能,特别是在区分正常放电与反电晕放电方面性能较好.通过适当整合,本方法可用于静电除尘器运行监控系统.

Detecting Anti-Electric-Corona Discharge Based on TVAR Parametric Model

The wavelet neural network is introduced into the time-varying auto regressive parametric model, so a new time-varying auto-regressive parametric model based on wavelet neural network is presented. Using the model a nonstationary anti-electric-corona discharge signal is modeled. The characters of the discharge signal can be extracted by the model parameter. Based on these characters, it does not know whether the anti-electric-corona discharge phenomenon happens. Simulation results indicate that the TVAR model for modeling a discharge signal has a good performance for distinguishing the normal discharge and the anti-electriccorona discharge. Therefore, by proper coordinating, the method can be used as a monitor system of the electric dust catcher.