航空发动机故障诊断主因子选取相似系数法

针对目前航空发动机气路故障诊断主因子模型中计算量大、故障方程多重共线性的问题,提出了主因子选取的相似系数法。通过计算故障模式和测量参数偏差之间的相似系数来选取故障方程的主因子,避免了对全部可能的主因子组合的计算。讨论了利用方法求解时的方程共线性问题,并对比了直接利用主因子模型求解的计算结果,结果表明,提高了故障诊断结果的可判断性,使故障诊断更加迅速可靠。

Similar Coefficients Method of Primary Factors Selection for Aero-engine Fault Diagnosis

For the common problems existed in models of main factors in fault diagnosis of aero engines that how to decrease calculation amounts and overcome questions of multicollinearity in error equations,the method of similar coefficients in choosing the main factors presented in the paper.Based on models of aero engine air path diagnosis,this method was used to compute the difference between the modes and the measured parameters,which was utilized to choose the main factors of error equations and led to avoid computing all possible combinations of the main factors.The comparison of the results between this method and other methods of direct models of the main factors,together with the research into questions of multicollinearity in dealing with the equations of this method showed that the method decreased the multicollinearity of error equations,more time-saving and reliable.