Motorized Spindle Temperature Intelligent Prediction And Experimental Research

Motorized spindle is the critical component of high speed CNC machine tool. The technological level of CNC machine tool is directly decided by the performance of motorized spindle. High-speed motorized spindle is often influenced by a variety of conditions in the process of spinning. The unsteady thermal characteristics of motorized spindle will cause failure and reduce its service life, while the breakdowns of the motorized spindle are mainly caused by high temperature. What is more, working accuracy may be reduced by the thermal deformation of motorized spindle, and the performance of the motorized spindle may be affected. In order to ensure the precision and working life of the motorized spindle, it is necessary to predict and control the temperature. In the process of operation, the factors those influence the temperature of motorized spindle contain environment temperature, coolant temperature and run time. However, it is highly complex and nonlinear of the change rule of the temperature field. There are some difficulties in measuring the internal temperature of motorized spindle. In order to predict the surface temperature change rule and the internal temperature field of motorized spindle, we do some research in the paper based on the 150MD24Y20 motorized spindle. The main works are shown below.Further research is done to predict the surface temperature of the motorized spindle. Set up the prediction model based on the basic data obtained from real-time detection of the surface temperature of the motorized spindle and build a temperature prediction model to predict the surface temperature of the metal motorized spindle based on BP algorithm and RBF algorithm. The prediction results indicate that both of the methods can solve the highly nonlinear problems of motorized spindle effectively. While, there are still some errors due to their defections.In order to perfect the defect of single neural network prediction model, the genetic algorithm to optimize the BP neural network(GA-BP) is established to predict surface temperature of motorized spindle. Matlab is used to simulate, and the prediction results are compared with the experimental results to verify the correctness of the prediction results. The results show that it is more accurate by using GA-BP algorithm to predict the temperature of motorized spindle.In this paper, prediction analysis of internal temperature field of motorized spindle is studied based on the loss experiment and the finite element model of the motorized spindle is established. Heat generating mechanism and the heat transfer mechanism is analyzed. Heat generation rate of rotor, stator and bearing are calculated by loss data obtained from measuring equipment. The coefficients of heat transfer from different parts of the motorized spindle are calculated by empirical formula. Then heating rate and the coefficients of heat transfer are loaded to the finite element model of the motorized spindle. Finally, the temperature distribution and temperature change rule are obtained by ANSYS software. Compared the experimental results to the distribution of temperature field from finite element model illustrate the correctness of the ANSYS simulation. By the prediction research for internal temperature of motorized spindle, the temperature field distribution is analyzed to provide theoretical basis for predicting the overall temperature and transient temperature of motorized spindle.It is shown in this paper, GA-BP algorithm is more accurate than single neural network algorithm when establish a surface temperature prediction model of motorized spindle. The internal temperature field distribution model based on ANSYS is established. It not only can be used to predict temperature change trend of motorized spindle but also can be the theoretical basis for the analysis of thermal characteristics of motorized spindle.