| The main method of processing the higher accuracy ball screw is grinding. With the extensive application of ball screw, the requirements of grinding technology are also rising. High precision, automation and intelligence have become the development trend of grinding. We can not build the mathematical model precisely to control the grinding process, because of the uncertainty, randomness and complexity of the grinding process. So, improving the automation and intelligence of grinding process plays an important role in improving the accuracy of grinding process.In this thesis, an intelligent grinding technology of screw grinder is researched, which realizes the automation and intelligence of grinding process and improves the accuracy of screw grinding. The error which affects the accuracy of grinding is measured real-timely with high-precision sensors. The grinding process monitoring, optimization and control by intelligent grinding system is basis for the data which is counted real-timely by high-speed data acquisition card. The completion of intelligent grinding is based on predicting and compensating the error, with the 840D CNC system.Prediction and compensation online are very important in the intelligent grinding technology of screw which affect the precision of grinding and reflect the intelligent level of screw grinding machine. Papers are focusing on the experiment of error prediction and compensation system for verifying the feasibility of prediction algorithm and online compensation. Prediction algorithm and online compensation were simulated by the application of the screw error which was measured. According to the simulation results of data, the feasibility of the algorithm was analyzed. Using self-built experimental platform, the experiment of predicting the error and compensating online was done. The feasibility and effectiveness of this method was further demonstrated for the realization of screw intelligent grinding by the analysis of the error curve and data. |