Research On Fault Diagnosis Technology Of Mine Drainage System Based On Improved Fuzzy Petri Net

Coal plays an important role in the economic development of our country.As the main energy consumption,it is critical to economic development and daily life.However,in the production of coal mining,the gushing water greatly affects the production and operation of the coal mine.Therefore,a safe,reliable and stable mine drainage system,and also can be diagnosis its faults is the necessary equipment for mining safely production.This thesis presents an improved fuzzy Petri net fault diagnosis(FPN)method for mine drainage system,which aims to improve the accuracy of fault diagnosis method.At first,we analyze the performance and common failures of mine drainage system,with the actual operating conditions of a mine drainage system in Shandong,we can design an automatic monitoring system for this drainage system,which can collect flow,pressure,vibration signal and other multi-source heterogeneous information.For the characteristics of nonlinearity between fault and fault source,choosing sparse Autoencoder theory to build a multi-dimensional information association model,subtracting the redundant information and improving the accuracy of multi-source heterogeneous data perception information.Because the Petri net can diagnose the failure of the mine drainage system during operation,we designed the mine drainage system fault diagnosis model based on fuzzy Petri net.The initial parameter of the model is based on artificial experience to set,which makes a certain impact on the results of the fault diagnosis.We choose BP neural network algorithm to optimize the parameters in the FPN model to reduce the error of fault diagnosis and to achieve the reliability and real-time performance of mine drainage system fault diagnosis.In order to verify the effectiveness of the method,we use the MATLAB to diagnose the fault data,and compare with the traditional fault diagnosis results of the reverse recursive method.The simulation results show that the FPN model can not only shorten the time for fault diagnosis,but also can diagnose the fault location and the transmission path of the fault in the process of mine drainage system fault diagnosis with higher accuracy.Through actual production application,it is verified that the proposed method can achieve the purpose of quickly diagnosing faults...