| The CNC machine tool on-machine measuring system is an intelligent device integrating product processing and measurement,which can effectively shorten the product development cycle and improve the production efficiency of the enterprise.However,the existing CNC machine tools on-machine measuring system generally have the problem that the measurement accuracy cannot meet the processing accuracy of more than three times,which has led to the low penetration rate of CNC machine tools on-machine measuring system in China.Therefore,it is of great significance to study the error compensation technology of on-machine measurement system of CNC machine tools to improve its measurement accuracy.In this paper,the improved experiment platform is used as the model,and the research is carried out from three aspects:analyzing the error source,establishing the error model and designing the compensation system,so as to improve the measurement accuracy of CNC machine tools on-machine measuring system.Based on the structural characteristics of CNC machine tools on-machine measuring system,the main error components are analyzed,and the three main factors that affecting the measurement accuracy are determined,namely temperature,speed,and X/Y coordinate position.In order to solve the prediction problem of single error under the influence of multiple factors,a genetic algorithm-back propagation(GA-BP)neural network was used for modeling and prediction,and compared with the prediction result established by the BP neural network,the GA-BP neural network has higher prediction accuracy and better robustness.In order to establish a high-precision comprehensive error prediction model,the comprehensive error model of the experiment platform was improved according to the correlation principle between errors and Abbe error theory.According to the requirements of error compensation of CNC machine tools on-machine measuring system,digital signal processor(DSP)is selected as the core processor of the error compensation system.In order to acquire the three factors that affect the accuracy of the experiment platform in real time,eight temperature acquisition circuits,two encoder acquisition circuits and two grating signal acquisition circuits are designed.In addition,two hardware compensation circuits are designed according to the principle of feedback interruption method.In order to realize the control of the compensation hardware circuit,the compensation system is programmed and developed through the DSP integrated development environment.First of all,the data of the three influencing factors are obtained through the collection program,and then the influencing factors are substituted into the single error model to calculate the individual error values,and then the single error values are imported into the comprehensive error model to calculate the comprehensive error value of each motion axis in the current state.Finally,the compensation pulse program is used to convert the error value into the corresponding pulse number and insert it into the feedback loop to realize the error compensation function.According to the structural characteristics of CNC machine tools on-machine measuring system,the existing experiment platform is improved.The verification experiments of various functions of the compensation system are carried out by using the experiment platform.According to the experimental phenomenon,the compensation system can collect relevant influencing factors in real time.The prediction accuracy and real-time performance of the comprehensive error model established in the DSP meet the requirements,and the compensation system can insert compensation pulses in the grating system.Figure[71]Table[3]Reference[83]... |