基于多维高斯贝叶斯的造纸机械传动系统故障识别  

Fault Identification of Paper Machinery Drive System Based on Multi-Dimensional Gaussian Bayesian

作　　者：张强[1] 冯晖[1] ZHANG Qiang;FENG Hui(Beijing Polytechnic,Beijing 100176,China)

机构地区：[1]北京电子科技职业学院,北京100176

出　　处：《造纸科学与技术》2025年第1期83-87,共5页Paper Science And Technology

基　　金：北京市教育科学规划领导小组办公室、北京市教育科学规划项目(AHDA24075)。

摘　　要：因造纸机械传动系统的振动信号具有非线性、非平稳及强噪声干扰等特点,导致其故障特征干扰维度较多,从而影响了故障识别效果。为提高造纸机械的运行效率,现提出基于多维高斯贝叶斯的造纸机械传动系统故障识别,结合LMD分解法,对传动信号PF分量进行提取,并对其进行相关性分析,通过构建PF分量矩阵,并计算出协方差矩阵,对特征值以及特征向量进行求解,从而根据特征值大小选取出主成分。结合多维高斯贝叶斯算法,利用故障样本估计各故障类别的多维高斯分布参数;计算新样本属于各故障类别的后验概率,选取概率最大者作为识别结果,输出故障识别结果。在实验中,对提出的方法进行了识别精度的检验。最终的测试结果表明,采用提出的方法对造纸机械传动系统进行故障识别时,算法的正向样本覆盖率较高,具有较为理想的识别效果。Due to the vibration signal of papermaking machinery drive system has the characteristics of nonlinearity,non-stationarity and strong noise interference,which leads to more interference dimensions of its fault characteristics,thus affecting the fault identification effect.In order to improve the operational efficiency of paper machinery,the fault identification of paper machinery transmission system based on multidimensional Gaussian Bayes is now proposed.Combined with the LMD decomposition method,the PF component of the transmission signal is extracted and its correlation is analyzed.By constructing the PF component matrix and calculating the covariance matrix,the eigenvalues as well as the eigenvectors are solved,so that the principal components are selected according to the size of the eigenvalues.Combined with the multidimensional Gaussian Bayesian algorithm,the fault samples are used to estimate the multidimensional Gaussian distribution parameters of each fault category;the a posteriori probability that a new sample belongs to each fault category is calculated,and the one with the largest probability is selected as the identification result,and the fault identification result is output.In the experiments,the recognition accuracy of the proposed method is tested.The final test results show that the algorithm has a high coverage of positive samples when the proposed method is used to identify faults in the transmission system of paper-making machinery,and it has a more ideal identification effect.

关 键 词：造纸机械 传动系统 故障识别 多维高斯贝叶斯 识别精度 

分 类 号：TS73[轻工技术与工程—制浆造纸工程]