Expert system for cylinder grinding control based on surface roughness
GU Ji nan, PAN Tian
（Mechanical Information Engineering Research Center, Jiangsu University, Zhenjiang 212000, China）


Abstract: Aiming at the problems such as high quality, low efficiency and low quality and high cost of the traditional casting process, the grinding process was studied and the grinding process was summarized and optimized. Based on the expert knowledge and operating experience, the expert system for the grinding control of the cylinder surface was designed and developed from the surface roughness during the robot grinding process. The BP neural network model was used to predict the grinding results, and the improvement of the grinding control method was discussed, and the optimization of the machining parameters was realized. In the process of grinding expert system, the main function of the system was fully analyzed and designed in detail, using the Java programming language to achieve system function, using MySQL database for data storage. The results indicate that the system can effectively reduce the surface roughness during grinding and improve the surface quality. This is to reduce the labor intensity, improve the efficiency of grinding, reduce environmental pollution, saving production costs have a very important significance.
Key words: cylinder grinding ;expert system ; BP neural network