滚动轴承性能的乏信息过程实验评估

发布时间：2018-05-16 04:09

本文选题：滚动轴承 + 乏信息理论　；参考：《河南科技大学》2012年硕士论文

【摘要】：滚动轴承性能指标中动态性能非常重要，主要包括振动和摩擦力矩。振动会加速轴承内部零部件的磨损与疲劳损坏，摩擦力矩是评价轴承运转灵活性的重要指标。研究表明轴承振动和摩擦力矩都具有不确定的强烈波动和趋势变化，且具有非线性特征，属于概率分布和趋势规律都未知的乏信息系统。这不利于振动和摩擦力矩的动态评估，因为动力学方程未知且难以找出确定的非线性模型。因此，如何对这种复杂的振动和摩擦力矩进行有效的评估分析成为亟待解决的问题。本文首先用混沌理论进行相空间重构，使轴承性能时间序列的原始动力学特征得到最优恢复；然后基于相空间，用最大熵自助法分别建立当前样本和先验样本的概率密度函数；再结合贝叶斯统计理论，建立轴承性能特征参数的动态贝叶斯后验概率密度函数；最后提出轴承性能特征参数的点估计和区间估计方法等。至此，乏信息数学模型构建完毕，依此就能有效地实现对轴承振动和摩擦力矩的评估分析。通过对四种不同损伤直径的滚动轴承振动实验数据的评估研究，估计结果表明，振动数据的样本均值都能被估计区间完美地包络，且估计结果和试验结果在数值上很接近。滚动轴承振动的平均不确定度与轴承损伤直径的增大而增大。最后对三种不同转速的滚动轴承摩擦力矩实验数据进行评估研究，估计结果表明，，摩擦力矩数据的样本均值都能被估计区间完美地包络，且估计结果和试验结果在数值上很接近。摩擦力矩的平均不确定度随轴承转速的增大而增大。本论文研究所用到的基本理论和方法具有通用性和可行性。研究表明，本文所构建的乏信息数学模型对滚动轴承振动和摩擦力矩的实验评估分析是正确可靠的，估计结果和试验结果在数值上很接近，满足工程需要，从而为轴承性能的非线性动力学特征演变评估奠定新的理论基础。
[Abstract]:The dynamic performance of rolling bearing is very important, including vibration and friction moment. Vibration can accelerate the wear and fatigue damage of bearing internal parts. Friction torque is an important index to evaluate the bearing running flexibility. The results show that the vibration and friction torque of bearing have strong fluctuation and trend change with uncertainty, and have nonlinear characteristics, which belong to the lack of information system with unknown probability distribution and trend law. This is not conducive to the dynamic evaluation of vibration and friction moment because the dynamic equation is unknown and it is difficult to find a definite nonlinear model. Therefore, how to effectively evaluate and analyze the complex vibration and friction moment becomes an urgent problem. In this paper, the phase space is reconstructed by chaos theory, and the original dynamic characteristics of bearing performance time series are recovered optimally, then the probability density functions of current samples and prior samples are established by maximum entropy self-help method based on phase space. The dynamic Bayesian posteriori probability density function of bearing performance characteristic parameters is established based on Bayesian statistical theory. Finally, the point estimation and interval estimation methods of bearing performance characteristic parameters are proposed. So far, the mathematical model of lack of information has been constructed, so that the evaluation and analysis of bearing vibration and friction moment can be realized effectively. Based on the evaluation of the vibration experimental data of four kinds of rolling bearings with different damage diameters, the estimated results show that the sample mean values of the vibration data can be perfectly enveloped by the estimated interval, and the estimated results are very close to the experimental results. The average uncertainty of rolling bearing vibration increases with the increase of bearing damage diameter. Finally, the experimental data of friction torque of three rolling bearings with different rotational speeds are evaluated. The estimated results show that the sample mean values of friction torque data can be perfectly enveloped by the estimated interval. The estimated results are very close to the experimental results. The average uncertainty of friction torque increases with the increase of bearing speed. The basic theories and methods used in this paper are universal and feasible. The research shows that the mathematical model constructed in this paper is correct and reliable for the experimental evaluation and analysis of rolling bearing vibration and friction moment, and the estimated results are very close to the experimental results, which meet the engineering needs. Thus, a new theoretical basis is established for the evolution and evaluation of nonlinear dynamic characteristics of bearing performance.
【学位授予单位】：河南科技大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH133.33

【参考文献】