改进k-means的控制自动化系统工业故障诊断预测

Improved k-means Industrial Fault Diagnosis and Prediction of Control Automation System

为提升自动化系统的运行效果,提出改进k-means的控制自动化系统工业故障诊断预测方法。该方法首先使用改进k-means对TE化工数据聚类处理,依据处理结果完成数据信号的去噪;再基于鲁棒ICA方法计算信号的独立成分,建立故障诊断模型,结合建立的固定监测阈值,完成TE化工过程的故障诊断;系统故障诊断后,继续对系统TE化工过程实施监测,依据监测结果建立故障预测模型,结合LS-SVM回归预测方法实现自动化控制系统TE过程的故障自动化预测。实验结果表明,使用该方法开展控制自动化系统工业故障诊断预测时,故障诊断预测效果好。

In order to improve the operation effect of the automation system,an improved k-means control automation system industrial fault diagnosis and prediction method is proposed.The method first uses the improved k-means to cluster the TE chemical data,and completes the denoising of the data signal according to the processing results;then calculates the independent components of the signal based on the robust ICA method.A fault diagnosis model is established,and the established fixed monitoring thresholds are combined.The fault diagnosis of the TE chemical process is completed;after the system fault diagnosis,then the system TE chemical process is monitored,a fault prediction model is established based on the monitoring results,and the LS-SVM regression prediction method is combined to realize the automatic control system TE process fault automatic prediction.The experimental results show that when the method is used to carry out industrial fault diagnosis and prediction of control automation system,the fault diagnosis and prediction effect is good.