Research On Electronic Controlled Engine Fault Diagnosis System Using Ai Based On Labview Platform

With the development of the world automobile industry, our automobile industry has greatly development. The performance, structure and comfort level of the automobile have greatly improved. Especially, the electronic controlled engine is widely used in the automobile in recent years. Compared with the traditional engine, the fault diagnosis will be more and more difficult. So the fault diagnosis system using Artificial Intelligence (AI) has become one of the most important subjects in the vehicle inspection field. In this dissertation, the structure of the electronic controlled engine is described, and the fault diagnosis model is made after the features of sensors are analyzed. Finally, the software of the fault diagnosis system is developed.Firstly, the structure and features of auto sensors are described. Some signal pre-treatment circuit and sampling circuit are designed, and the sensor data collection platform based on CAN bus is built. In order to communicate conveniently between PC and platform, the CAN-USB adapter is designed.In view of the complexity of electronic control engine failure, the Back-propagation (BP) Neural Network which is self-learning and adaptive is used as the core of the fault diagnosis system. During the BP network is designed, the network structure is established after the comparison of five networks. In order to improve the efficiency of BP training, Ant Colony Optimization algorithm is used to train the BP network. After the fault diagnosis model is built, a set of data is input the model to prove the correctness of fault identification. From the results, it is proved that the design of BP net model is correct. Finally, the software of electronic controlled engine fault diagnosis system is developed in the LABVIEW platform. The man-machine interface is designed, and the system can be controlled at the interface by users.