Research On Thermal Error Modeling Of Machine Tools Based On Bayesian Network

The main content of this dissertation is the model of thermal errors for CNC machine tools via Bayesian Networks (BN). Based on fully understanding and deeply analyzing the current status of research and application of the thermal error modeling and compensation technique for CNC machine tools, Bayesian Networks modeling is proposed.BN is one of the most important methods to dealing with uncertainty problems. It has complete mathematic support and is based on the probabilistic and statistics theory. Because of its natural knowledge representation, powerful reasoning ability, as well as convenient decision-making mechanism, it is applied in many domains now. The prediction model of this paper is legible in frame work. The probability computation based on BN has strong theory basis and is concise to compute and easy to understand. By the abundance of factor which induces the thermal error, the prediction precision of this model will become higher and the result is more useful. The primary content of this paper includes:In chapter 1, the background and the significance of the research are stated. The research history and current situation of the thermal error modeling are also provided. After the description of the shortcoming of current thermal error modeling, the BN modeling is advised and the main content of the dissertation is presented.In chapter 2, BN methods and theoretical basis are stated, including probability map mode, constitution of BN, learning of BN and reasoning of BN. In the end, Naive Bayesian Classifier (NBC) is stated.In charpet 3, the method of thermal error of machine tool based on BN is stated. First, the method of how to select network is introduced. Then, the method of establishing the Bayesian network is introduced. Finally, briefly analyse the advantage of the modeling.In charpet 4, the experiment of thermal errors of CNC machine tools is introduced. The experiment equipments include CNC lather, the temperature measure system, and laser CCD displacement sensors. The scheme and the data of this experiment are discussed in detail.In charpet 5, the whole process of BN thermal error modeling is described. First, one temperature variable modeling is introduced; Second, four temperature variable modeling is introduced; Third, the whole temperature variable modeling is introduced. Finally, the superiority of BN modeling is validated by the data analysis and the contrast of Least Square (LS) modeling.In charpet 6, the study contents and conclusions of the dissertation are summarized, and the further research works is forecasted.