Aiming at the problems of fault diagnosis of hybrid gas insulated switchgear(HGIS)equipment during operation and low detection efficiency of SF6 gas leakage fault,this study designs a gas detection method based on differential photoacoustic spectroscopy.By using molecular infrared absorption spectroscopy technology,this method realizes the gas detection function of PPT and PPB magnitudes,and avoids the interference of the photoacoustic effect of air itself.At the same time,the equip-ment state space model is constructed,and the Kalman filter method is extended based on the particle filter algorithm.The re-sidual value is obtained by comparing the real-time measured value of the state variable with the estimated value,and the system failure is early warned.In addition,an adaptive threshold method is added to detect equipment faults to avoid false alarms caused by model errors and noise.The experimental results show that the system has a good effect on fault detection and fault early warning,and the detection accuracy is higher than 94％.
关键词
故障检测/运维管理/状态空间模型/粒子滤波算法/变量残差
Key words
fault detection/operation and maintenance management/state space model/particle filter algorithm/variable resid-ual
万方数据