Research On Abnormal State Diagnosis Of Electric Energy Metering Device

Electric energy metering device is the core equipment of power trade settlement.The reliability and accuracy of its on-site operation is the basis of ensuring fair electricity transaction.It is the interests of the vast number of power customers and the safe use of electricity.It is of great significance to improve the quality of power supply service and social credibility.The function and operation environment of the device are becoming more and more complex,and the abnormality of the device and power consumption presents the characteristics of sudden,complex,multi-faceted and difficult to recur.Traditional on-site electric energy metering device abnormal diagnosis uses manual periodic on-site verification.The detection,diagnosis and treatment are of low timeliness,poor accuracy and low user acceptance,which is increasingly contradictory to the development trend of safe,stable and mutual trust in the electricity environment.Therefore,how to detect and deal with the abnormal operation of metering device in time and ensure the stable and reliable operation of metering device is of great significance.The main research work of this paper includes:1.The current fault diagnosis methods of mainstream electric energy metering devices are studied,including on-line monitoring technology of electric energy metering devices and fault diagnosis and identification technology for abnormal conditions such as electricity theft by users.The advantages and disadvantages of traditional methods are analyzed and summarized.2.The current mainstream power metering devices and their sampling systems are studied.By analyzing the function principle of electric energy metering device and combining with sampling data,the common fault types and sampling characteristics of electric energy metering device under various fault conditions are summarized.3.A fault diagnosis model is established.The relationship between the change of sampling value of electric energy metering device and the abnormal operation state of electric energy metering device is studied.Based on the change of five kinds of direct sampling values and the correlation between indirect sampling values and faults,a Bayesian network model for fault diagnosis of electric energy metering devices is established.4.The fault analysis method of electric energy metering device is put forward.According to the established fault diagnosis network model of electric energy metering device.Based on Bayesian logical reasoning,a method of transformer fault oil chromatographic analysis based on Naive Bayesian network is proposed.Bayesian network is used to analyze the characteristics of uncertain energy metering devices.The fault prediction and prediction of electric energy metering device are given.By training and testing the established diagnostic model with historical data,it is verified that the accuracy of this method is much higher than that of the existing diagnostic methods in the fault diagnosis of power metering devices,and it meets the requirements of Engineering field.5.Using the proposed fault modeling and analysis method of electric energy metering device,based on Genie tool,the fault cases of an electric energy metering device in a domestic power grid are analyzed.Compared with the diagnosis conclusions of traditional methods,the effectiveness of this method is verified.