关键词: 小波分析, 时间序列, 动态神经网络, 故障, 预测

Abstract: As the power plant prediction of steam turbine rotor vibration time series is difficult, the wavelet decomposition to realize trend prediction is proposed. Some non-stationary time series can be decomposed into several approximate stationary time series with wavelet decomposition. Decomposed time series are forecasted with auto-regression model, to obtain forecasting results of the original time series. Experiments with a power plant vibration signal show that the local and overall effect of the algorithm is better than neural network approaches. The result shows rotor vibration time series forecasting accuracy of this model.

Key words: wavelet analysis, time series, dynamic neural network, fault, predict