【作者】 黄海凤；

【导师】 许明恒；

【作者基本信息】 西南交通大学 ， 机械电子工程， 2013， 博士

【摘要】 数控机床是加工制造业的关键装备,数控制造技术属于先进制造技术之一,其水平直接反应一个国家制造业的实力。制造业又是关系国计民生和国家安全的支柱产业。滚珠丝杠副是数控机床不可或缺的关键部件,其性能状态直接影响数控机床的加工质量和生产效率。为提高国产数控机床的整体技术水平,保证制造业健康发展,研究滚珠丝杠副的性能退化机理,评估滚珠丝杠副的性能衰退过程显得尤为必要。滚珠丝杠副的使用寿命约10年左右,丝杠副工作过程中性能逐渐衰退,传动精度逐渐降低,并最终导致丝杠副故障,甚至报废。为实时准确评估滚珠丝杠副的性能状态,制定智能维护维修方案和备件计划,本文首先研究了滚珠丝杠副的性能退化机理,分析了产生磨损、振动和冲击的影响因素,并建立了滚珠丝杠副性能评估体系,具体工作如下：对滚珠丝杠副进行受力分析,首先针对不同结构的滚珠丝杠副分析滚珠和滚道的接触载荷特性、曲率变化、接触区域摩擦特性及滚珠的受力特点分析滚道和滚珠的受力和运动,研究接触应力和摩擦力矩对丝杠副的磨损、疲劳、振动的影响,并对滚珠进行运动分析,揭示滚珠丝杠副性能退化的机理。根据滚珠丝杠副的结构、性能退化特点和工作过程,提出丝杠副性能退化评估的目的和性能退化评估原则,系统科学地分析性能退化规律,采用定性分析和定量分析结合的方法实现丝杠副实时性能评估。分析丝杠副的精度参数和性能参数与性能的对应关系,建立滚珠丝杠副性能退化评估指标集,采用三层分级评估结构建立滚珠丝杠副性能评估系统结构和性能退化评估体系。基于VMC1500等型号的数控机床建立滚珠丝杠副性能退化试验方案,根据信号特点确定振动传感器、电流传感器的型号及采集系统等硬件。应用动态模糊神经网络建立滚珠丝杠副性能退化评估模型,实现信号特征与丝杠副性能状态的映射关系,鉴于样本的分布对网络预测精度有很大影响,因此采用正交试验和性能指标试验相结合的试验方法。研究表明,滚珠丝杠副性能退化为缓变过程,退化曲线是连续的,为准确提取性能退化特征并解决海量数据问题,建立动态聚类数据简化算法,通过调整动态聚类阈值调整评估精度和系统的动态响应特性。通过对信号的时域分析、频域分析及小波分析,采用能量谱分析振动信号、电流信号的能量分布作为性能退化评估模型的输入参数。同时结合指标的定性分析提高了性能退化评估精度。为解决数控机床工作过程中无法实时测量摩擦力矩的难题,假设丝杠副空载时,驱动电机输出扭矩全部用来克服摩擦力矩,应用这种方法建立了电流信号和摩擦力矩之间的映射关系。将摩擦力矩作为性能评估模型输入量,提高性能退化评估精度。依据性能退化和磨损程度的评估结果,建立了智能维护维修实用模型。根据该模型,可以确定数控机床滚珠丝杠副等关键部件的智能维护维修方案,合理制定备件计划,有利于精简维护维修成本和备件成本,并且有效提高了数控机床的使用效率。上述研究对提高滚珠丝杠副的性能,减少维护维修成本和备件成本,有效提高数控机床的加工性能和效率,具有重要的理论意义和工程实用价值。更多还原

【Abstract】 NC machine tool is the key equipment in manufacturing industry, and NC manufacturing technology is one of advanced manufacturing technology. The developing state of NC manufacturing technology can directly show the manufacturing ability of a country. And the manufacturing industry is also the pillar industry related to people’s livelihood and national defense security.Ball screw pair (BSP) is the key part of NC machine tool, and its performance states directly affects machining quality and production efficiency of NC machine tool. To research the performance degradation (PD) mechanism of BSP is extremely important for improving the performance of BSP, realizing fault prediction and intelligent maintenance of machine tools.The service life of BSP may be10years. During its working condition, performance of BSP declines gradually, so that the precision of transmission is reduced gradually and BSP faults occur, or even scrapped in the end. In order to accurately evaluate the performance of BSP on time, intelligent maintenance plan and spare parts plan, the PD mechanism of BSP is firstly studied in this thesis. And then, the influence factors of wear, vibration and impact and so on are analyzed. Based on above researches, the BSP performance evaluation system is established. Main contents of this thesis are following:Force analysis of BSP. Firstly, aimed at BSP with different structures, the contact load characteristic between ball and its rollaway nest, the curvature change of rollaway nest, the friction characteristic of contact area and the loading features of the ball are analyzed. The characteristics of loading and motion between balls and its raceway are also studied. Then, the effects of wear, fatigue and vibration due to contact stress and friction torque are studied and the kinematic analysis of the ball is also carried out. The PD mechanism of BSP is uncovered by above researches.On the bases of the structure, PD characteristic and working process of BSP, evaluation goal and evaluation principle on PD of BSP are proposed. Based on sufficiently analysis of whole system PD processing, on time performance evaluation of the BSP is carried out by means of combination of qualitative analysis and quantitative analysis.PD evaluation index set of the BSP is set up by correspondence analysis between the performance and the accuracy parameters of BSP. The performance evaluation system structure and the PD evaluating system of BSP are established by means of three layers classification evaluation. Test program of BSP PD is built up based on VMC1500and other type NC machine tools.In order to realize the mapping relationship of signal characters and performance states, PD evaluating model of BSP is established by dynamic fuzzy neural network. Because the prediction accuracy of network is seriously affected by sample distribution, the test method of orthogonal test conjunted with performance index test is employed.The study is shown that the PD of BSP is a gradually reducing process, well the BSP degenerated curve is continuously, the degradation curve of ball screw, therefore, is continuous. In order to obtain PD character and solve the massive data problem, data simplification algorithm with dynamic clustering is established. Evaluation precision and dynamic response characteristics can be adjusted by adjusting the threshold of dynamic clustering.The vibration signal of energy spectruml and the energy distribution of current signal are taken as input parameters of PD evaluation model by means of time domain analysis, frequency domain analysis and wavelet analysis for the signals. PD evaluation precision is improved by qualitative analysis of some indexes at the same time.In order to solve the difficult problem that the friction torque can not be meassured in real time in NC machine tool working conditions, the hypothes is taken as that the output torque of drive motor is completely used to overcome friction torque while BSP is in no load condition. In this way, mapping relationships between current signals and friction torque is built and PD evaluation precision is also improved by taking friction torque as the input index.Intelligent practical model used for NC machine tool preserving and maintenance is proposed according to the BSP evaluation results of PD and attrition rate. On the basis of this model, intelligent maintenance schedules of NC machine tool key parts such as BSP can be confirmed and the spare parts planning can be set out reasonably. These are in favor of reducing the cost of maintenance and the cost of spare parts, and also increasing service efficiency of NC machine tools effectively.Above research results have important theoretical significance and engineering practical value for enhancing BSP performance, reducing costs of maintenance and spare parts, and effectively increasing the processability and processing efficiency of NC machine tool.更多还原