复杂系统状态预测的小波包方法

基于小波包变换和混沌理论对复杂系统状态预测方法进行了研究.首先应用小波包变换对系统的特征参数序列进行3层分解,得到第3层从低频到高频8个频率成分的时序;然后,对8个时序作进一步分析,以确认它们都存在混沌特性,再应用混沌理论分别建立8个时序的预测模型,分别对8个时序进行预测;最后,基于小波包理论将混沌模型预测的结果予以小波包重构,实现对系统特征参数序列的预测.实例研究表明,该方法具有较高预测精度,可有效地应用于复杂系统的状态预测和故障趋势预测分析中.

Forecasting method of complex system condition based on wavelet packet transformation

Based on wavelet packet transformation and chaos theory, the research on forecasting method of complex system condition is made. Firstly, by using wavelet packet transformation, system feature reference data series are decomposed into eight time series parts from low frequency to high frequency. And the further analysis of decomposition indicates that there exists a chaos feature in the eight time series. Then, by using chaos theory, the chaotic forecasting models are established to respectively forecast the eight time series. Finally, the forecasting results of chaotic models are reconstructed based on wavelet packet theory. By doing so, the forecasting of system feature reference data series can be made. Our result demonstrates that the proposed method is of high precision, and can be applied to condition forecasting and fault trend forecast analysis of complex system effectively.