基于二代小波变换的信号去噪及其软测量建模

Signal De-noising based on second generation wavelet transform and soft sensor modeling

化工生产过程中采集到的数据信号通常具有随机性和非平稳性,附加了各种噪声,以至于影响数据建模的拟合效果和泛化性能。本文基于二代小波分析的特点,提出了一种对信号数据进行小波变换阈值去噪的方法。该方法可去除大部分高频随机噪声,提取真实信号,进而提高数据的置信度。将该方法与支持向量机相结合并应用于双酚A反应过程质量指标软测量模型中。仿真结果表明,该方法能有效恢复数据的真实性,提高数据建模的拟合精度与泛化性能。

Because of the nonstationarity and randomicity of data in the chemical production process, it contains different noises inevitably, which will affect the fitting and the generalization capability in data modeling. Based on the characteristics of wavelet analysis, this paper proposed a signal denoising method which combines the second generation wavelet transform with a new threshold function . The method can remove random noises, and extract true signals to improve the credibility of the data. Combined the method with support vector machine and applied to a modeling for the performance figure of BPA productive process, the result indicates that it is effective to regain the factuality of the data and improve the fitting and generalization capability of the soft sensor model.