基于小波变换的工况区域识别

针对生产过程的特点 ,对过程工况区域识别问题进行了描述 ,在分析小波变换的基础上 ,研究了利用小波变换的多尺度分析特性 ,提取生产过程检测信号的多尺度定性变化特征 ,并分析和证明了提取的工况区域特征具有描述工况区域的稳定性和完备性。在此基础上 ,结合模式归纳分类算法 ,提出了一种基于小波变换的工况区域识别方法。应用实例研究 ,证实了所提方法的可行性和有效性

Process performance region recognition based on wavelet transform

Characteristics of real industrial processes were used to analyze the problem of process performance region recognition and the qualitative pattern series expression was introduced. Analysis of the wavelet transform was used to extract the multi scale qualitative features of the process measurement signals and to prove that the extracted qualitative features could completely describe the process performance region. A new method of process region recognition based on wavelet transform was proposed using the pattern inductive algorithm. Theoretical analysis and real application results verify that the proposed method is feasible and effective.