This project develops the intelligent centering system, which focuses on the centering process of optical glass lenses in medical electronics area. Centering is the key process of glass lens manufacture to control the optical axis precision. Since the lens with precise optical axis is necessary in a medical microscope. The development of the intelligent centering system can improve the national process ability of optical glass lens to reach the level of medical electronics.

As the optical axis precision is improved, scratches and edge cracks are easily generated in centering process. Therefore, this project developed the clamping optimization and multi-sensing technology for process monitoring and prediction. The optimized clamping structure was designed with the mathematical model. The stress analysis was done with the parameters including materials, geometries and clamping forces. The clamping stress of the lens was simulated to predict the scratches of lens surface. The results of simulation were then imported into the support vector regression (SVR) to enhance the efficiency of analysis. On the other hand, the experiment of clamp trimming was designed to evaluate the effects of parameters on the roughness of clamp edge. The clamping optimization was realized. The scratch specification can be reached to <L1x0.001 by the clamping optimization model. Furthermore, the defect prediction and parameter optimization model were developed by the genetic algorithm (GA). The process efficiency can be improved with the quality of wheel grinding that meets the high-end specification. The quality was reached to edge crack < E0.1 and scratch < L1x0.001.