【摘要】 针对发动机系统结构复杂和运行状态监测信息丰富的特点,考虑到状态监测模型本身的不确定性,研究发动机系统状态监测方法。采用贝叶斯线性模型融合多源状态监测信息,降低状态监测信息的不确定性;建立基于Gamma随机过程和Weibull分布的发动机系统状态监测模型,采用贝叶斯因子方法对发动机系统状态监测模型进行选择与评估;采用贝叶斯模型平均方法对发动机系统状态监测模型进行组合与优化,提高发动机状态监测准确性。通过算例,验证提出方法的有效性。更多还原

【Abstract】 Aiming the characteristics of complex structure and abundant condition monitoring information of engine system,considering the model its own uncertainty,a condition monitoring method for engine system is studied. The Bayesian linear model is adopted to fuse the multi-source condition monitoring information and reduce the condition monitoring information uncertainty. The condition monitoring models of engine system are built based on Gamma random process and Weibull density distribution,respectively. The Bayes factor method is used to select and assess the condition monitoring models of engine system. The Bayes model averaging method is used to combine and optimize the condition monitoring models of engine system. Using the above method,the condition monitoring accuracy of the engine system is improved. The effectiveness of the proposed method is verified with a calculation example.更多还原