| 崔靖崧,肖凯文,姚毅,李兵,张大义,黄行蓉.基于尺寸链不确定性量化的涡轮轴承不对中
故障失效概率预测[J].航空发动机,2026,52(1):168-174 |
| 基于尺寸链不确定性量化的涡轮轴承不对中
故障失效概率预测 |
| Prediction of Turbine Bearing Misalignment Failure Probability Based on Dimension ChainUncertainty Quantification |
| |
| DOI：10.12482/ISSN.1672-3147.20231115001 |
| 中文关键词: 失效概率预测 公差累积计算 装配尺寸链 多项式混沌展开 不确定性量化 涡轮轴承 航空发动机 |
| 英文关键词:failure probability prediction tolerance accumulation calculation assembly dimension chain polynomial chaotic expan⁃
sion uncertainty quantification |
| 基金项目:国家自然科学基金(52105083、52175071,U2341231)、先进航空动力创新工作站(HKCX2020-02-016)资助 |
|
| 摘要点击次数: 2908 |
| 全文下载次数: 481 |
| 中文摘要: |
| 涡轮设备的装配公差和几何公差累积会导致转子不对中问题，进而引发轴承的不对中支反力过大，导致轴承加速磨损、
寿命缩短乃至失效。为了提高发动机涡轮转子试验装置的可靠性，开展了轴承不对中故障失效概率评估。基于不确定性量化方
法来评估轴承失效的概率，优化支承机匣公差分配以提升装置的可靠性。选择不同公差分配的支承机匣模型，建立装配尺寸链进
行公差累积计算；通过蒙特卡洛法和多项式混沌展开（PCE）法，评估不同尺寸链轴承支座的不对中值概率分布；基于仿真确定的
最大允许不对中值，评估不同公差条件下的轴承失效概率，并据此得出最佳的公差分配方案。结果表明：相较于蒙特卡洛法，PCE
法计算不对中值分布的最大误差为9.09%且计算效率提高92.50%。所提出的公差分配方案对于提升发动机涡轮转子试验装置的
可靠性具有重要意义，PCE不确定性量化方法对于公差累积引发的失效概率预测提供了重要的实践参考。 |
| 英文摘要: |
| The accumulation of assembly and geometric tolerance in turbine equipment can lead to misalignment issues in the rotor,
subsequently causing excessive misalignment reaction forces in the bearings. This may result in accelerated wear, reduced lifespan, or even
failure of the bearings. To enhance the reliability of a engine turbine rotor testing apparatus, research was conducted to assess the prob?
ability of bearing failure due to misalignment. Uncertainty quantification methods were used to evaluate the likelihood of bearing failure
and to select the optimal tolerance distribution for the bearing supports. Initially, bearing supports models with different tolerance were
chosen to establish assembly dimension chains for the calculation of tolerance accumulation. Subsequently, the probability distribution of
misalignment values for different dimension chains of bearing supports were assessed using Monte Carlo and Polynomial Chaos Expansion
(PCE) method. Moreover, based on the maximum permissible misalignment value determined by simulations, the failure probability of bear?
ings under different tolerance conditions was calculated to determine the best tolerance distribution plan. The results show that compared
with the Monte Carlo method, the PCE method calculates the misalignment distribution with less error and significantly improves the
computational efficiency. The tolerance distribution scheme obtained is of significant importance for enhancing the reliability of the engine
turbine rotor testing apparatus. The PCE uncertainty quantification method provides an important practical reference for predicting the
failure probability caused by tolerance accumulation. |
| 查看全文 HTML 查看/发表评论 下载PDF阅读器
|