Research On Online Safety Precaution Technology Of High-medium Pressure Gas Regulator

As the key equipment to regulate and stabilize pressure of the gas pipeline network,gas pressure regulator plays an important role in the gas pressure regulating system.High-medium pressure gas regulator is pressure regulating equipment of the upstream gas transmission and distribution network.Its safe and reliable operation directly is related to the safety of gas supply in the downstream region.Therefore,strengthening the study on safety precaution technology of high-medium pressure gas regulator,evaluation the operation state of gas regulator effectively,early detection failure and realizing precaution of failure,are important to guarantee the normal operation of equipment and personal and property security of downstream residents.Through the survey on typical high-medium gas pressure regulating station,this paper summarizes and analyzes the common fault types,fault causes and fault performance of high-medium pressure gas regulator.At the same time,the continuous wavelet transform and wavelet packet method are used respectively to analyze the fault pressure data.The energy spectrum and the frequency band energy of the fault pressure data are extracted respectively.The applicability and effectiveness of the two methods of extracting the fault characteristics of the gas regulator are analyzed and compared by the distribution rule of the characteristic data.The results show that the method of frequency band energy characteristic based on wavelet packet decomposition is simple and effective,and it is more suitable for fault feature extraction of high-medium pressure gas regulator.Finally,in order to improve the precaution accuracy of gas regulator,on the basis of summing up the frequency band energy value distribution rule of wavelet packet in different fault types,in this paper,the fault diagnosis experience of the expert technicians in the field of gas regulator also is considered.A multi variable safety precaution model combining qualitative analysis with quantitative calculation is established initially.In addition,the method of combining wavelet packet frequency characteristic energy with RBF neural network is applied to the fault identification of high-medium pressure gas regulator,and the feasibility of the method is discussed.The results show that the method of wavelet packet-RBF neural network can realize the classification of different faults of gas regulator.This is a useful exploration for the next step to realize intelligent fault diagnosis and precaution of high-medium pressure gas regulator.