一种数据驱动的湿法烟气脱硫系统的故障诊断方法

采用主元分析法(PCA)对火电厂湿法烟气脱硫系统进行故障诊断.利用PCA建立系统故障诊断模型,通过计算平方预测误差(SPE)来检测系统是否发生故障,若有故障发生则用Q贡献率法来分离故障,识别发生故障的原因.通过采集福州某电厂湿法烟气脱硫系统的历史数据进行Matlab仿真并在组态王中显示故障诊断曲线,表明用PCA法对湿法烟气脱硫系统故障具有良好的诊断效果.

A data-driven fault diagnosis of wet flue gas desulfurization system

In this paper,a method using principal component analysis(PCA) for fault diagnosis of the wet flue gas desulfurization system in thermal power plants is proposed.This method establishes the model of fault diagnosis by PCA and detects the system fault by calculating square prediction error(SPE).If some faults are detected,the Q contributing rate method is adopted to separate the fault and then identify the source of fault.By gathering some historical data from the wet flue gas desulfurization system of a power plant in Fuzhou for Matlab simulation and the display of fault diagnosis curve in Kingview,it shows that the PCA method poses a good ability in fault detection of the wet flue gas desulfurization system.