Research On Fault Cause Reasoning Based On Meteorological Factors

Natural factors,human factors and equipment problems can cause power system faults,such as thunderstruck,the short circuit of transmission and distribution lines,mistake operation or the aging equipment,among which meteorology is the main cause of power system fault.China has a vast territory and diverse meteorological environment.Severe weather disasters bring a large risk for the power system safety operation.As a result,the analysis on the meteorology factors on power system fault is quite important for eliminating the fault and system maintenance.In this context,the multi-source heterogeneous data fusion method is proposed.The proposed method researches high-capacity data fusion technology cross security partitions,which conforms to the security regulations.The information of power system operation,equipment state information and environmental monitoring information of several systems located in different security zones are merged.The early warning framework of power grid fault risk based on large data is also proposed.The analysis method of meteorological causes on power system fault based on BP neural network is proposed.First,a method of timing fuzzy Petri net is used to diagnose the faulty equipment.Then,a neural network is established based on historical meteorological data and power equipment failure rates.This neural network can predict equipment failure rates under certain meteorological conditions.Finally,equipment failure rates under different meteorological conditions are analyzed,and the meteorological reasons of the equipment failure is deduced.According to the actual situation of Jiangsu power grid in China,the fault diagnosis platform and the fault-assisted analysis system based on multi-source data fusion is developed in this paper,where the multi-source data mainly include the information of power grid operation,equipment status and environmental monitoring information.With the establishment of power system fault diagnosis platform based on multi-source data fusion,the redundancy information of multi-source data can be used to achieve the smart fault diagnosis more quickly and accurately.In this work,the safe and stable operation level of a power system can be improved,and the management of distribution network technique can be enhanced.