Research On Light-Duty Gasonline Vehicle Remote Monitoring And Fault Diagnosis Technology

In order to meet increasing emission standard requirements and further improve automobile performance, automotive enterprises now apply electronic control technology in the automobiles increasingly, thus making the structure of automobile electronic control system more and more complicated. It becomes more and more difficult to judge the causes of faults and determine the locations of faults. Based on the study of vehicle remote monitoring and intelligent diagnosis technology, this thesis establishes an vehicle remote monitoring and intelligent diagnosis system, and its core technology is studied.First of all, according to extended protocol of vehicle fault diagnosis ISO14230and ISO15765, a universal data collection platform based on K line and CAN bus is designed and developed. In order to solve the traditional monitoring and diagnosis method is restricted by geographical position, a method for remote diagnosis data acquisition is proposed, which is based on Internet and3G. According to data transmission process and network transmission characteristics, the data transmission strategy and server monitoring model are designed to achieve remote monitoring data acquisition.And then, this paper design a intelligent diagnostic model, which is divided into three parts:data level fusion diagnosis based on RBF neural network; characteristic level fusion diagnosis based on support vector machine and principal component analysis; decision level fusion diagnosis based on D-S evidence theory. Output signal failures and on-off faults of coolant temperature sensor, oxygen sensor, manifold absolutely pressure sensor are simulated, real-time status data of the vehicle are acquired by data collection platform. By comparing and analyzing automobile intelligent diagnostic model that is based on multi-information fusion at data level, characteristic level and decision level, the effectiveness of diagnostic model is validated, a variety of combinations in practical application are proposed.Finally, a vehicle remote monitoring and fault diagnosis system is also designed and built, combining diagnostic communication interface, PC and smart phone client, vehicle remote monitoring and fault diagnosis center. Combined with the actual functional requirements, this paper does research on architecture models of the system. Combined with the multiple information fusion diagnosis models, and design the intelligent diagnosis method based on multi-agent. At the same time, improve the processing performance of the current handheld diagnosis method by the use of PC and smart phones with automobile fault diagnosis is implemented for the client, realizing the auto fault diagnosis anytime and anywhere.