基于K－means的机泵群SISO传感器故障监测

打开文本图片集

中图分类号：TQ053；TP399 文献标志码：A 文章编号：1001-5922（2025）07-0163-05

Abstract：Inorder to improvethe safetyand efficiencyof industrial equipmentoperation，afault monitoring method of single input single output （SISO） sensor of machine pump group based on K - means clustering algorithm was discussed.By analyzing the measurement data of the sensor in the feedback control loop，the K -means algorithm was used to cluster the setpoints，control outputsand processvariables，and the normal data modelunder steady state was established，and thedynamiccentroid updateand fault detection algorithm were designed toidentifyabnormal data points.The experimental data ofthe pump group of an oil production plant of a petrochemical unit were used to verifythat the method could effctivelydetect sensor deviation and accuracy degradation faults，and was sensitive to process changes.The results show thatthe proposed method can detect sensor faults in time and provide reliable technical support for the stable operation of the pump group.

Key words ： pump group； sensor; fault diagnosis； K - means clustering algorithm

通过传感器进行故障监测是确保工程质量和操作安全的关键手段。（剩余8165字）