The Design Of Automotive Engine Fault Diagnosis And Prediction System Based On ARM9

With the rapid development of society, automotive industry has become an important symbol of modern society. Automobile and electronic technology has been closely combined with the development of computer and electronic technology. The automotive structure is more complex and performance will be better. As the automotive structure and technology progressing, automobile fault diagnosis must be based on new theories and testing methods to meet the requirements of modern automotive fault diagnosis so as to reduce the exhaust of automotive emissions and noise pollution, save energy, reduce the maintenance costs, extend automotive life and make the automobile safe, reliable and stable operation.This thesis proposed and developed a automotive fault diagnosis and prediction system based on the ARM9 processor. The system is a vehicle-mounted detection equipment, it can read the real time typical data of automotive engine running, analyze and process these data and determine the operational status and the probability of possible engine fault to achieve the function of fault diagnosis and prediction.The automotive fault diagnosis and prediction was introduced from hardware and software two aspects in this thesis. For the hardware, described the hardware structure of system, analyzed the diagram of each hardware module and completed the production of hardware PCB board. For the software, gave the software architecture of system, introduced the development process of the Windows CE operating system and gave the development process of data transmission based on CAN bus communication protocol.The system based on the traditional neural network combined fuzzy theory with neural network to collect fault information effectively. Lead into inertia and change the learning rate to improve the convergence speed of neural network. Genetic algorithm was used by neural network to solve the learning process into a local optimum problem. The improved neural network was used by the system for automotive engine misfire diagnosis. The results of experiment indicate that automotive engine fault diagnosis and prediction system based on ARM9 can effectively carry out misfire fault diagnosis.