The assembly process of body-in-white (BIW) welding is one of the four key processes of automobile production, whose running quality crucially affects the production cost and quality of the vehicles. Diagnosing failures accurately and in real time is an important means and key technology to ensure BIW welding process to run efficiently and stably and it can effectively reduce the production cost and improve the quality. A system of real-time monitoring and fault diagnosis for BIW welding process is constructed, which is able to monitor the causes of failures in real time to avoid them, and able to monitor and analyze the state data online to recognize faults and diagnose them.Firstly, combined with the characteristics of BIW welding assembly process, the causes and mechanisms of failures such as equipment failure, quality failure and personnel accidents is analyzed to determine the scheme of welding process monitoring and the task of fault diagnosis. Secondly, the structure of the BIW welding process real-time monitoring and fault diagnosis system is analyzed. And combined with the needs of monitoring the process, the technologies of the industrial control systems’communication, sensors, vehicle identification and configuration software which is development environment of data acquisition and process monitoring system are researched. And then on the basis of Fuzzy Petri Net, the method of online equipment fault diagnosis and the logic process of the fault diagnosis system are studied. And this method fuzzes the causes of equipment failures, describes the transitive relation among the equipment failures by means of fuzzy production rule, and diagnoses equipment failures online through a formal reasoning algorithm. Lastly, according to the technology and method researched, a BIW welding process real-time monitoring and fault diagnosis system is developed and applied in a welding workshop, and this provides guarantee and technology support for the efficient and stable operation of the BIW welding process. |