Hybrid Intelligent Model For High Speed And High Precision Spindle Temperature Field Prediction

With the propose of 2025 intelligent manufacturing,the development of intelligent CNC machine tools has been put on the agenda.The intelligent high speed motorized spindle,as one of the core components of CNC machine tools,has a crucial role in machining efficiency and accuracy,and has become an important factor for intelligent manufacturing.During the working process,the temperature rise of high speed motorized spindle is affected by various conditions,causing nonlinear thermal deformation and low service,therefore,it is of great significant to predict the temperature rise of the spindle to keep the stability of the spindle.In this thesis,taking 170MD30Q15 motorized spindle as research object,the cooling and lubrication parameters of the spindle were analyzed which affect temperature distribution greatly,and the prediction model of heat transfer coefficient optimization was established based on hybrid intelligent motorized spindle temperature field.The main research work were as follows:(1)According to motor knowledge and the theory of heat transfer,the heat exchange for motorized spindle was analyzed.Based on the temperature distribution analysis and the boundary conditions were obtained.And through the analysis of the finite element simulation of spindle temperature equation obtained that the boundary conditions of prediction model was heat source and heat transfer coefficient.(2)The temperature test system for motorized spindle was designed and built.The parameters such as cooling water temperature,cooling water flow rate,air supply pressure and oil supply interval could be adjusted.According to single factor test method,the influence of cooling and lubrication system parameters on the temperature field of electric spindle was analyzed.The orthogonal test method was adopted to obtain the combination of working conditions to ensure the minimum temperature rise of the spindle The results showed that the best working conditions included that the cooling water flow rate was 0.35m3/h?0.38m3/h,the cooling water temperature was 17??19?,the compressed air inlet pressure was 0.36MPa?0.39MPa,and the oil supply time interval was 1min?2min.(3)Based on the experimental data and the finite element model,a hybrid prediction model of motorized spindle temperature field was established.The finite element model of 170MD30Q15 motorized spindle was established by COMSOL Multiphysics software.The loss of the motor and the bearing loss were measured by the loss test,and the loss was transformed into the heat source of the finite element model.The boundary conditions of the heat transfer coefficient of the finite element model were obtained by the optimization method.In this thesis,genetic algorithm and least square method were used to optimize the heat transfer coefficient.The Intelligent optimization of heat transfer coefficient was realized by combining the experimental data and the finite element model data.(4)The optimized heat transfer coefficient was used as the boundary condition to predict the temperature field of 170MD30Q15 spindle.By comparing the temperature field simulation data and experimental data:after the least squares optimization,the average relative error of the predicted temperature field was 0.96%;and after the genetic algorithm optimization,the average relative error of the predicted temperature field was 2.05%.Comparing the prediction accuracy of the two methods,the least squares optimization had a higher prediction accuracy,because of transient temperature test data and real-time of the heat transfer coefficient optimization.(5)Used the morse coefficient to analysis of loss sensitivity for the prediction model of motorized spindle optimized by genetic algorithm,and obtained that the morse coefficient was 0.262,and the loss sensitivity model belonging to the sensitive level.In this thesis,a hybrid intelligent prediction model for the temperature field of the high speed and high precision motorized spindle was established,which can provide the theoretical basis for the optimization design of the motorized spindle and lay the foundation for the establishment of the intelligent motorized spindle system.