Study of fault diagnosis model based on multi-class wavelet support vector machines

Compromising wavelet kernel and multi-class least squares support vector machines, this article put forward to a fault diagnosis model based on multi-class wavelet support vector machine. The model heightens auto-adaptive model classification ability by adjustment of scale parameter of wavelet kernel, and make use of remarkable learning ability and generalization ability of small sample of vector machines to improve speed and effectiveness of fault diagnosis. And taking fault diagnosis in heat-recycling system of thermal power plant as an example to analysis.