Fault Diagnosis Of Diesel Engine Integrated Exhaust Aftertreatment System Based On Multi-source Information Fusion

With the increasingly strict emission regulations,it is difficult to meet the regulatory requirements relying on a certain after-treatment system.The National 6 regulations have implemented in 2019,and the traditional single after-treatment system cannot meet the requirements of the regulations.Therefore,it is necessary to rely on the integrated aftertreatment system to solve the emission problem.However,the integrated after-treatment system has a harsh working environment and complex structure,and there is a coupling effect between the subsystems.The working conditions of the diesel engine change at any time.Therefore,in order to quickly and accurately diagnose the degradation of the system,it is necessary to adopt a method is to perform degradation diagnosis on the system.This article will use multi-source information fusion technology to diagnose the degradation of(DOC+DPF+SCR+AOC)integrated after-treatment system.The GT-Power engine simulation software is used to build the simulation model of the road cycle of diesel vehicles,diesel engine,and the working process of the integrated after-treatment system.Using the simulation model,by changing the structural parameters and chemical reaction parameters in the integrated after-treatment system structural model,different types of degradation and the degree of degradation are simulated.And under steady-state operating conditions,the changes of parameters such as temperature difference,pressure difference,and conversion efficiency of each component under different speeds and loads are studied.Through analysis,the parameters with greater sensitivity is selected as the degradation characteristic parameter.Method of degradation diagnosis based on the parameter is proposed,and the effects of the degradation of each subsystem on the parameters of other subsystems are studied.Using the simulation model,study the changes of the integrated after-treatment system’s exhaust parameters under different degradation and road conditions,study the impact of the degradation on other subsystems under road conditions,and compare with the results obtained under steady-state conditions.Through analysis,the characteristic parameters of degradation diagnosis under road conditions are determined,and whether misdiagnosis diagnosis is required is analyzed.In addition,it is found that when the subsystem is degraded,the pressure difference of the front subsystem is greatly affected,and the temperature difference of the rear subsystem is greatly affected.Research on information entropy fusion technology from the aspects of information entropy measurement index,feature space division method,feature information entropy extraction method,etc.,build a probabilistic neural network model,use the average value of feature information entropy and diagnostic feature parameters to train the probabilistic neural network The information entropy neural network and the mean neural network are used to detect the diagnosis results and compare the diagnosis results.Through analysis,the information entropy fusion and neural network technology can be used to diagnose the degradation of the integrated after-treatment system under a certain accuracy,and when the information entropy is added as the diagnostic feature value,the diagnostic accuracy of the diagnostic model can be improved.