系统可靠性分析中环境因子评估方法研究及软件实现
发布时间：2018-09-08 21:14
【摘要】：高可靠长寿命产品的可靠性分析和寿命预测是一个热点问题,如何获得其工作环境下足够多的寿命数据是难点,也是今后一个重要的研究方向。经过研究表明,通过加速寿命试验在可高应力环境下可获得更多的产品数据,环境因子可解决如何将高应力环境下的数据折算至正常工作环境下的问题,用环境因子按照一定的折算原则对数据进行折算,可扩充高可靠长寿命产品在工作环境下的数据信息,得到的数据可用来对产品进行可靠性分析或寿命预测等。本文主要研究环境因子的估计方法,重点讨论基于产品性能退化数据的环境因子的估计问题、基于多温度条件下产品性能退化数据的环境因子的预测问题和基于环境因子的寿命预测问题。本文的主要研究内容和成果如下:1、基于性能退化试验的环境因子的评估。本文讨论了性能退化试验相较传统的寿命试验的优势,对性能退化试验中定数截尾试验和定时截尾试验做了详细介绍,并对性能退化数据的特点进行了研究,说明了用性能退化数据估计环境因子的合理性。本文结合试验数据的分布,采用最佳线性估计法(BLUE)对环境因子进行估算,并举例详细的介绍了估算过程。2、基于性能退化数据和加速寿命试验的环境因子评估。本文利用产品在高应力下加剧退化的特性,研究了定数截尾情况下,将性能退化试验与加速寿命试验相结合来估计环境因子的方法。由于本文仅考虑温度应力,所以在性能退化数据的基础上,给出了基于阿伦尼斯(Arrhenius)模型建立环境因子估计模型和模型参数的估计方法。为说本文提出的方法的精度,给出了一种新的验证方法。经过实例验证发现,此方法得到的环境因子具有较高的精度。3、基于环境因子的寿命预测。通过加速寿命试验获可得高应力条件下产品的性能退化数据,本文基于这些数据给出了新的寿命预测方法。本文先通过环境因子的估计模型预测工作条件下的环境因子,接着估计产品在工作环境下的性能退化数据,利用产品在工作温度下的原有数据对折算后的数据进行修正,经过与原有数据的不断综合,最终可获得工作环境下性能退化数据。当工作温度下有一组产品样本时,结合数据的分布信息估计产品的平均寿命;当工作温度下仅有单个样本时,估计此样本的失效寿命。最后通过实例详细的介绍了估算过程,并给出了估计结果,证明了方法的可行性。4、环境因子估计和寿命预测的软件设计。本文基于MATLAB软件完成了环境因子的估计系统,实现了基于环境因子的产品估计过程,并具有获得数据退化轨迹,验证数据分布、计算伪失效寿命,估计结果验证等功能。
[Abstract]:The reliability analysis and life prediction of high reliability and long life products is a hot issue. How to obtain enough life data in its working environment is a difficult point and an important research direction in the future. The research shows that through accelerated life test more product data can be obtained under high stress environment, and environmental factors can solve the problem of how to convert data under high stress environment to normal working environment. Using environmental factors to convert data according to certain conversion principle can expand the data information of high reliability and long life products in the working environment. The obtained data can be used for reliability analysis or life prediction of products. In this paper, we mainly study the estimation method of environmental factors, and focus on the estimation of environmental factors based on product performance degradation data. The environmental factor prediction problem and the life prediction problem based on the environmental factor under the multi-temperature condition of product performance degradation data are discussed. The main contents and results of this paper are as follows: 1. Evaluation of environmental factors based on performance degradation test. In this paper, the advantages of performance degradation test compared with traditional life test are discussed, and the characteristic of performance degradation data is studied. The rationality of using performance degradation data to estimate environmental factors is explained. Combined with the distribution of test data, the best linear estimation method (BLUE) is used to estimate the environmental factors, and an example is given to illustrate the estimation process of environmental factors based on the performance degradation data and accelerated life test. In this paper, the method of estimating environmental factors by combining performance degradation test with accelerated life test under the condition of constant truncation is studied by using the characteristics of product degradation under high stress. Since only the temperature stress is considered in this paper, based on the performance degradation data, an estimation method of environmental factor estimation model and model parameters based on Arrhenis (Arrhenius) model is presented. In order to describe the accuracy of the proposed method, a new verification method is presented. An example shows that the environmental factor obtained by this method has a high accuracy of. 3, and the life prediction based on the environmental factor. The data of product degradation under high stress are obtained by accelerated life test. Based on these data, a new life prediction method is presented in this paper. In this paper, the environmental factors under working conditions are predicted by the estimation model of environmental factors, and then the performance degradation data of the products under working environment are estimated, and the converted data are corrected by using the original data of the products at the working temperature. After continuous synthesis with the original data, the performance degradation data under the working environment can be obtained. When there is a set of product samples at the working temperature, the average life of the product is estimated by combining the distribution information of the data, and the failure life of the sample is estimated when there is only a single sample at the working temperature. Finally, the estimation process is introduced in detail through an example, and the estimation results are given. The feasibility of the method. 4, the software design of environmental factor estimation and life prediction are proved. In this paper, an environmental factor estimation system based on MATLAB software is developed, and the process of product estimation based on environmental factor is realized. It has the functions of obtaining data degradation trajectory, verifying data distribution, calculating pseudo-failure life, and validating the result of estimation.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TB114.3
本文编号：2231719
[Abstract]:The reliability analysis and life prediction of high reliability and long life products is a hot issue. How to obtain enough life data in its working environment is a difficult point and an important research direction in the future. The research shows that through accelerated life test more product data can be obtained under high stress environment, and environmental factors can solve the problem of how to convert data under high stress environment to normal working environment. Using environmental factors to convert data according to certain conversion principle can expand the data information of high reliability and long life products in the working environment. The obtained data can be used for reliability analysis or life prediction of products. In this paper, we mainly study the estimation method of environmental factors, and focus on the estimation of environmental factors based on product performance degradation data. The environmental factor prediction problem and the life prediction problem based on the environmental factor under the multi-temperature condition of product performance degradation data are discussed. The main contents and results of this paper are as follows: 1. Evaluation of environmental factors based on performance degradation test. In this paper, the advantages of performance degradation test compared with traditional life test are discussed, and the characteristic of performance degradation data is studied. The rationality of using performance degradation data to estimate environmental factors is explained. Combined with the distribution of test data, the best linear estimation method (BLUE) is used to estimate the environmental factors, and an example is given to illustrate the estimation process of environmental factors based on the performance degradation data and accelerated life test. In this paper, the method of estimating environmental factors by combining performance degradation test with accelerated life test under the condition of constant truncation is studied by using the characteristics of product degradation under high stress. Since only the temperature stress is considered in this paper, based on the performance degradation data, an estimation method of environmental factor estimation model and model parameters based on Arrhenis (Arrhenius) model is presented. In order to describe the accuracy of the proposed method, a new verification method is presented. An example shows that the environmental factor obtained by this method has a high accuracy of. 3, and the life prediction based on the environmental factor. The data of product degradation under high stress are obtained by accelerated life test. Based on these data, a new life prediction method is presented in this paper. In this paper, the environmental factors under working conditions are predicted by the estimation model of environmental factors, and then the performance degradation data of the products under working environment are estimated, and the converted data are corrected by using the original data of the products at the working temperature. After continuous synthesis with the original data, the performance degradation data under the working environment can be obtained. When there is a set of product samples at the working temperature, the average life of the product is estimated by combining the distribution information of the data, and the failure life of the sample is estimated when there is only a single sample at the working temperature. Finally, the estimation process is introduced in detail through an example, and the estimation results are given. The feasibility of the method. 4, the software design of environmental factor estimation and life prediction are proved. In this paper, an environmental factor estimation system based on MATLAB software is developed, and the process of product estimation based on environmental factor is realized. It has the functions of obtaining data degradation trajectory, verifying data distribution, calculating pseudo-failure life, and validating the result of estimation.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TB114.3
【参考文献】
相关期刊论文 前3条
1 周延昆;威布尔环境因子的置信估计[J];科学通报;1985年09期
2 王伟;付永锋;;环境因子定义的等价性探究[J];通化师范学院学报;2009年12期
3 徐福荣;正态变量之商及其在环境因子分析中的应用[J];宇航学报;1984年01期
，本文编号：2231719
本文链接：https://www.wllwen.com/guanlilunwen/gongchengguanli/2231719.html
