Research On Improving Positioning Accuracy Of Line Scanning Gear Measurement Experimental System

The line scanning gear measuring instrument is mainly composed of a threecoordinate workbench,a turntable,a mandrel and a line structured light sensor,which can quickly and accurately scan the full information of the gear without contacting the surface of the gear.In order to develop a high-precision line-scanning gear measuring instrument,the error sources are analyzed.The measuring accuracy of the line-scanning gear measuring instrument is not only related to the accuracy of the wired structured light sensor itself,but also related to the dynamic performance and positioning accuracy of the three-coordinate table equipped with the line-scanning sensor.Therefore,in order to improve the measuring accuracy of the line-scanning gear measuring instrument,the error sources are analyzed,the error compensation model is established,the motor control algorithm is optimized,and the experimental verification is carried out.Firstly,according to the structure of the line scanning gear measuring system,the experimental device platform is built,and the error sources are analyzed,and the main factors affecting the accuracy of the line scanning gear measuring system are determined,namely the temperature of the workbench,the running speed and the coordinate position,and the Abbe error of each axis is calculated.Considering that the above errors are influenced by many factors,the coupling degree is high,and the law is complex,it is difficult to establish an whitening model.Therefore,in this paper,BP(Mind Evolutionary Algorithm Back Propagation)neural network is optimized by thinking evolution algorithm to model and predict each single error of the three-coordinate worktable,and the comprehensive error of the worktable is modeled according to the topological structure low-order body array description method.Secondly,in order to optimize the dynamic performance of the line scanning gear measuring system,the motor control algorithm is improved.According to the PMSM mathematical model,SVPWM algorithm and implementation principle,the PMSM fuzzy PI controller is built.However,due to the fixed universe of the conventional fuzzy controller,the adaptive ability needs to be improved.Therefore,the Grey Wolf Optimization(GWO)is introduced as a variable universe link,and the output adjustment parameters can be adjusted online to quantify and scale factors.A PMSM speed controller based on GWO optimized variable universe fuzzy PI is built to realize the dynamic optimization of the workbench.In order to verify the error compensation model,dynamic performance verification and single error prediction comparison experiments are carried out on the experimental device platform,and actual error compensation experiments are carried out on the experimental platform.The experimental results show that the response time of the controller is 8.36 ms and the disturbance response time is 2.91 ms,and all dynamic indexes are excellent.The root mean square of single error prediction is around 0.6;After compensation,the positioning error of each axis is within 10 μm,and the repeated positioning accuracy is within 3 μm.To sum up,the dynamic performance and positioning accuracy of the experimental platform have been greatly optimized.Figure(102)Table(9)Reference(80)...