Theoretical Research And Numerical Simulation On The Temperature Field Of Large Flat Wheel Grinding

The large flat grinding wheel is mainly used for the finishing of standard gears andgear cutting tools, Its precision reflects the ability of a country’s ultra-precision gearmachining technology. China’s gear industrial manufacturing precision is1to2gradeslower than foreign gear manufacturing technology. as the grinding temperature willcause great impact on the manufacturing precision, this paper focus on the grindingtemperature field of large flat wheel grinding.Grinding can greatly improve the accuracy of the workpiece compared with othermaterials-removal process, but the high energy input needed to effectively removeworkpiece material during grinding. as the caused grinding heat of grinding processhave great effects on performance, surface quality of the workpiece and the service lifeof the grinding wheel, the study of grinding zone temperature field becomes veryimportant. Although research on flat grinding temperature field has been relativelyfruitful, the gear tooth surface grinding temperature field of large flat wheel grindinghas not been studied. Therefore, this paper focus on the study of large flat wheelgrinding temperature in the manner of analytical modeling and ANSYS simulation,predicting the results using neural network to reasonably select the grinding processparameters. the main contents of this paper include:(1) Introduce the basic theory of the grinding temperature field. Analyze thetemperature field of large flat wheel grinding with the help of moving heat sourcemethod and image method and derive the grinding temperature mathematical model.Examine the impact of the grinding process parameters on the temperature field.(2) Describe the theoretical basis of the finite element method in the application ofthermal effects. The geometric modeling, meshing and boundary conditions definitionof Large flat wheel grinding are finished by using the finite element software ANSYS.Heat load is applying on the gear tooth surface by using the ANSYS parametric designlanguage to simulate the large flat wheel grinding process. The grinding temperaturefields of gear tooth surface at different moments are finally obtained.(3)Because artificial neural networks have nonlinear approximation properties, thispaper establishes the improved BP neural network based on the Levenberg-Marquardtalgorithm. The use of orthogonal analytical model results as neural network trainingsamples to train the neural network to achieve rapid prediction of large flat wheel grinding temperature, which is used to guide the selection of the grinding processparameters during the actual processing.Finally, this paper summarizes the work which has been done, points out theshortcomings and discusses the future research.