Study On Fault Mechanism Of Cold Starting Process In Vehicle Diesel Engine And Its Intelligent Diagnosis

Cold starting process in vehicle diesel engine is a complicated process and in a bad working condition, and there are always faults deviating from normal condition. But a good fault diagnosis expert system can do real-time fault diagnosis and alarm to cold starting process in vehicle diesel engine with a lot of working status of it.This will be conducive to operation and maintenance personnel to make the corresponding measures in time to improve the reliability and security of it, therefore improve the working life of diesel engine.In this paper, the fault mechanism of cold starting process in vehicle diesel engine is researched using the thermodynamic theory, and an intelligent fault diagnosis system for cold starting process in vehicle diesel engine is researched and developed based on these researches with the application of fuzzy neural network theory and expert system technology, to meet the current needs of intelligent fault diagnosis for cold starting process in vehicle diesel engine, the main work and innovation of this paper are as follows:(1) Abnormity of air valve clearance between piston and cylinder, leakage of air valve, wear of three modules, blockage of injection nozzle, leakage of exhaust valves, noise, wear and other fault mechanisms of cold starting process in vehicle diesel engine are researched and analyzed, and the influence from the thermal parameter of cold starting process in vehicle diesel engine is assessed, which provides a good theoretical base to extract feature parameters in intelligent diagnosis process of cold starting process in vehicle diesel engine.(2) A new model of thermal parameters of cold starting process in vehicle diesel engine is put forward considering leakage of compressing working substance, heat loss, as well as clearance volume and other parameters for the first time. This model shows, when only considering leakage of air and heat loss in the compressing process of cold starting in vehicle diesel engine, the temperature of cylinder in vehicle diesel engine will be estimated on the high side, and the normal ignition will be influenced. The model provides a solid theory foundation to diagnose the abnormity of ignition of cold starting process in vehicle diesel engine and solve its problems.(3) The combination and mutual complementation of deep knowledge and shallow knowledge is adopted in the intelligent fault diagnosis system for cold starting process in vehicle diesel engine, the inference engine is established using inaccurate inference method; In order to satisfy the requirement of the given precision, the connection weight of the networks is studied and adjusted using the backpropagation training algorithm (BP algorithm); The knowledge base of fuzzy neural network fault diagnosis expert system for cold starting process is established using the trained connection weight, threshold, central value and width of Gaussian function; The faulty condition of the key parameters in cold starting process in diesel engine is reasoned according to the direction of forward and backward mixed inference.(4) An intelligent fault diagnosis system for cold starting process in vehicle diesel engine is established according to the working process of cold starting process in vehicle diesel engine using the development tool Visual Basic6.0. The results of commission of this system show that, the knowledge base of this system is reasonable, the inference engine is quick and efficient, and the accuracy is high.