针对难加工材料的刀具可靠性建模及其分析方法研究

发布时间：2018-07-05 17:42

本文选题：刀具 + 难加工材料　；参考：《电子科技大学》2017年硕士论文

【摘要】：《中国制造2025》中明确提出“在航天、汽车、机床等重点领域集中开展数控机床专项成果的应用验证,全面推进国产数控系统、功能部件和刀具的研发和配套应用,最终实现机床智能化”。刀具作为数控加工的一个重要部件,其可靠性直接影响到整个加工过程的加工效率和稳定性。因此,开展刀具可靠性分析,具有重要的理论和现实意义。对刀具进行可靠性分析时,需要获得刀具退化数据。为了获得刀具退化数据,本文针对难加工材料进行刀具切削加工实验,由于加工工况、现场环境等各方面因素的限制,难以获得大量的实验数据。针对此“样本小、数据乏”的难题,引入Bayes(贝叶斯)方法对经验信息进行综合。在该理论基础上,本文围绕刀具退化数据开展刀具可靠性分析,具体研究内容主要包括以下三个方面:(1)为了得到刀具在某一固定时刻可靠度值,本文提出了一种基于退化量分布模型的刀具可靠性分析方法。该方法通过确定某一固定时刻刀具退化数据分布,结合Bayes理论,计算刀具在某一固定时刻可靠度值。为了得到刀具可靠度随时间的变化趋势,本文在假设刀具失效分布已知的基础上,对刀具进行可靠性分析。(2)为了模拟出刀具的退化过程,本文提出了一种基于退化过程模型的刀具可靠性分析方法。该方法在未考虑个体差异的条件下,基于Bayes理论,对刀具退化模型中的参数进行更新,实现了对刀具退化过程的模拟和刀具可靠性分析。同时,由于同一刀盘上的刀具退化过程存在差异,本文提出了一种考虑个体差异的随机过程模型的可靠性分析方法,该方法得到了刀具可靠度随时间的变化趋势。(3)为了对上述两种方法进行验证,本文实施了难加工材料数控加工实验。此实验在考虑了切削用量对切削效率影响的条件下,设计了三因素三水平正交实验。每组切削用量下,进行10次切削加工,收集了大量的实验数据,并运用此数据对上述方法进行验证。此外,在此实验中除了采集到刀具磨损量数据外,采集到的实验数据还包括:振动信号,表面粗糙度和工件切屑数据。基于此数据对刀具磨损状态进行多角度监测,形成了比较全面的刀具切削加工数据库。通过对刀具磨损量分析,证明本文提出的方法可以很好地对刀具进行可靠性评估,为数控加工工艺规划和机床刀具智能选择提供了理论支撑。
[Abstract]:The "made in China 2025" clearly states that "in key areas such as aerospace, automobiles, machine tools, and other key areas, we will concentrate on the application and verification of special achievements in numerical control machine tools, and comprehensively promote the development and matching application of domestic numerical control systems, functional components and cutting tools." Finally realize the intelligent machine tool "." As an important part of NC machining, tool reliability directly affects the machining efficiency and stability of the whole machining process. Therefore, it has important theoretical and practical significance to carry out tool reliability analysis. It is necessary to obtain tool degradation data for tool reliability analysis. In order to obtain the tool degradation data, this paper carries on the tool cutting experiment to the difficult machined material, because of the processing condition, the field environment and other factors' limitation, it is difficult to obtain a large amount of experimental data. In order to solve the problem of "small sample and lack of data", Bayes (Bayesian) method is introduced to synthesize the empirical information. Based on this theory, the tool reliability analysis is carried out around the tool degradation data. The specific research contents include the following three aspects: (1) in order to obtain the tool reliability value at a fixed time, In this paper, a tool reliability analysis method based on degenerate distribution model is proposed. In this method, the tool degradation data distribution at a fixed time is determined, and the reliability of the tool at a fixed time is calculated by combining Bayes theory. In order to obtain the change trend of tool reliability with time, this paper analyzes the tool reliability based on the assumption that the tool failure distribution is known. (2) in order to simulate the process of tool degradation, In this paper, a tool reliability analysis method based on degenerate process model is proposed. Based on Bayes theory, the parameters of tool degradation model are updated, and the simulation of tool degradation process and tool reliability analysis are realized. At the same time, due to the difference of tool degradation process on the same cutter head, this paper presents a reliability analysis method of stochastic process model considering individual differences. The change trend of tool reliability with time is obtained by this method. (3) in order to verify the above two methods, the NC machining experiment of difficult machining material is carried out in this paper. In this experiment, considering the effect of cutting parameters on cutting efficiency, a three-factor three-level orthogonal experiment was designed. Under each set of cutting parameters, 10 cutting processes were carried out, a large number of experimental data were collected, and the above method was verified by using this data. In addition, in addition to the tool wear data, the experimental data include vibration signal, surface roughness and chip data. Based on this data, the tool wear state is monitored from various angles, and a more comprehensive tool cutting database is formed. Through the analysis of tool wear, it is proved that the method presented in this paper can evaluate the reliability of the tool well and provide theoretical support for NC machining process planning and tool intelligent selection.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TG71

【参考文献】