Research On Bonnet Polishing Process Based On Five Axis Hybrid Machine

Optical technology has been well used in national defense, aerospace, aviation, information technology and many other fields. The optical elements also have gradually changed into aspheric spares. Therefore, the processing method of these new elements becomes a seriously issue that needs to be studied on. Bonnet polishing can resolve this issue both effectively and precisely.Bonnet polishing is a new precise method that is suitable for polishing aspheric surface and freeform surface. This method is not only easy to handle but also brings high machining efficiency and surface accuracy as well as low processing cost. In China, lots of research institutes have done some researches on bonnet polishing process, but no further research has been done on the core technology. This thesis studies on the influence of bonnet polishing process parameters and polishing path on surface precision of the work piece. Then, considering the profile accuracy and surface roughness, this thesis optimizes process parameters and path planning. With the optimization of a series of process of bonnet polishing, Surface quality could finally reach to a high level.According to Preston equation and Hertz contact theory, this thesis analyzes the velocity distritution and pressure distribution of the processing surface and build velocity distritution model, pressure distribution model and material removal model. By analyzing these models, this thesis finds the influence mechanism of bonnet rotational speed, processing speed and curvature radius of the work piece on polishing process. This thesis chooses orthogonal experiment to research on process parameters fixing on the material removal rate and surface roughness, aiming at finding the influence laws of these two factors caused by the five parameters. Then, the best combination of parameters which can get high removal rate and low surface roughness can be found. This thesis also accomplishes the function design and analysis of the path planning algorithm. In order to meet the demands, neural network control is chosen as the method. After finishing the structure design of neural network control and training the algorithm for the demands, the algorithm is used for polishing simulation with different paths under different conditions. The accuracy of the algorithm can be tested according to the result of the simulation.Last, based on all studies before, this thesis conduct experiments on workpieces and obtain high precise processing surface, which verifies the correctness of the material removal model, optimum result of process parameters and path planning algorithm.