基于灰自助模型下的双渐开线齿轮振动疲劳寿命预测

发布时间：2018-12-05 19:33

【摘要】：乏信息即贫信息,指研究对象呈现的特征信息不完备或不充分,缺乏先验知识。在低碳经济背景下,齿轮传动逐渐向轻量化、高功率密度方向发展,其面临的强度和可靠性问题更加突出。长期以来,齿轮疲劳性能评价主要依靠经典力学、疲劳寿命试验,并以威布尔、指数分布或对数正态分布为假设,进行可靠性分析;传统的疲劳试验分析方法以大样本经典统计理论为基础,不仅需要大量的试验工作投入,而且人力、物力耗费巨大,发展相对滞后;随着新能源、新材料与重载交通运输机械的快速发展,动力传输机械的极端工况对相关齿轮的疲劳寿命与可靠性提出了更高的要求,传统的齿轮疲劳性能评估理论已无法满足社会的需求,迫使人们不得不面对大量乏信息系统的处理与评价。作者以双渐开线齿轮为研究对象,基于乏信息理论对该齿轮在随机振动情况下对接触疲劳寿命的影响进行研究,具体内容如下:(1)将乏信息系统理论引入到双渐开线齿轮疲劳强度性能试验评估中,突破以经典几何、力学理论和大样本试验统计为基础的P-N-S模型。基于乏信息理论,分别建立灰色系统理论、自助法及蒙特卡洛法的数学模型,通过对双渐开线齿轮参数进行评估,分析三种模型的特性。(2)结合灰色GM(1,1)模型和Bootstrap抽样原理的优点,提出一种基于乏信息理论的动态估计预测的新方法,并建立灰自助预测模型。利用MAT LAB软件对灰自助模型进行可靠度仿真,对比均匀分布、Weibull分布等统计法,分析灰自助模型的可靠性。(3)利用灰自助模型对双渐开线齿轮的随机振动误差进行预测,并分别与灰色GM(1,1)模型、蒙特卡洛法的预测值进行对比,确定灰自助模型的可行性与准确性。(4)基于乏信息理论,在灰自助模型中引入峰值概率密度函数和应力功率谱密度法,研究双渐开线齿轮在随机振动情况下对接触疲劳寿命的预测。(5)利用灰自助模型预测双渐开线齿轮轴向、周向及径向振动过程中疲劳寿命,通过对比试验数据模拟出该模型的误差分布范围,分析该模型预测结果的准确性,验证该模型能否解决双渐开线齿轮乏信息疲劳寿命问题。
[Abstract]:Lack of information, that is, poor information, means that the characteristic information presented by the research object is incomplete or insufficient, and lacks prior knowledge. Under the background of low carbon economy, the gear transmission is developing towards the direction of light weight and high power density, and the problems of strength and reliability are more prominent. For a long time, the fatigue performance evaluation of gears mainly depends on classical mechanics, fatigue life test and reliability analysis based on Weibull, exponential distribution or logarithmic normal distribution. The traditional fatigue test analysis method is based on the classical statistical theory of large sample. It not only needs a large amount of experimental work investment, but also consumes a lot of manpower and material resources, and its development is relatively lagging behind. With the rapid development of new energy, new materials and heavy haul transportation machinery, the extreme working conditions of power transmission machinery require higher fatigue life and reliability of gear. The traditional theory of gear fatigue performance evaluation can not meet the needs of the society, forcing people to face a large number of processing and evaluation of lack of information system. The influence of double involute gear on contact fatigue life under random vibration is studied based on the theory of lack of information. The main contents are as follows: (1) the theory of lack of information system is introduced into the fatigue strength test evaluation of double involute gear, which breaks through the P-N-S model based on classical geometry, mechanics theory and large sample test statistics. Based on the theory of lacking information, the mathematical models of grey system theory, self-help method and Monte Carlo method are established, and the parameters of double involute gear are evaluated. The characteristics of the three models are analyzed. (2) combining the advantages of the grey GM model and the principle of Bootstrap sampling, a new method of dynamic estimation and prediction based on the lack of information theory is proposed, and the grey self-help prediction model is established. The reliability of the grey self-help model is simulated by MAT LAB software, and the reliability of the grey self-help model is analyzed by comparing the statistical methods such as uniform distribution and Weibull distribution. (3) the random vibration error of double involute gear is predicted by using the grey self-help model. And compared with the grey GM (1) model and Monte Carlo method, the feasibility and accuracy of the grey self-help model are determined. (4) based on the lack of information theory, The peak probability density function and the stress power spectrum density method are introduced into the grey self-help model to study the prediction of contact fatigue life of double involute gears under random vibration. (5) the grey self-help model is used to predict the axial direction of double involute gears. In the process of cyclic and radial vibration, the error distribution range of the model is simulated by comparing the experimental data, the accuracy of the prediction result is analyzed, and it is verified that the model can solve the problem of fatigue life of double involute gear without information.
【学位授予单位】：青岛科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TH132.41

【参考文献】