论文部分内容阅读
给出了发动机故障诊断的 Monte Carlo通解算法 ,该算法可以有效地解决发动机故障诊断算法中由于故障方程存在多重共线性所引起的误诊、漏诊与多解问题。首先给出求解故障方程的 Monte Carlo算法 ,该算法能够保证得到满足故障方程的全部合理解 ;能够简单地用于各种故障相关性准则 (例如各种残差向量范数准则 )以及亚定故障方程的求解问题 ;并且算法简单易行 ,无需为每个特定情况专门编制计算程序。其次提出了基本解与通解的概念 ,并且给出了利用主成分分析与利用主因子模型求基本解的两种有效算法。利用基本解与通解算法可以将由于多重共线性引起的复杂故障诊断结果表示为简单明了的形式 ,有利于对诊断结果作出正确决策。用计算机模拟方法对算法的有效性进行了分析 ,对于 JT9D发动机气路方面的 2 4个实际故障样本 ,所给出的算法的确诊率为 86%～ 92 %
The Monte Carlo general solution algorithm for engine fault diagnosis is given. This algorithm can effectively solve the problems of misdiagnosis, misdiagnosis and multiple solutions caused by the existence of multicollinearity in the fault equations of engine fault diagnosis algorithm. Firstly, the Monte Carlo algorithm for solving fault equations is given. The algorithm can guarantee to obtain all reasonable solutions to the fault equations. It can be easily used in various fault correlation criteria (such as various residual vector norm criteria) and sub-adventures Solving equations; and the algorithm is simple and easy, and there is no need to formulate a calculation program for each specific situation. Secondly, the concepts of basic solution and general solution are proposed, and two effective algorithms of using principal component analysis and principal factor model to solve basic solution are given. Using the basic solution and the general solution algorithm, the complex fault diagnosis result caused by multiple collinearity can be expressed as a simple and clear form, which is beneficial to make correct decision on the diagnosis result. The effectiveness of the algorithm was analyzed by computer simulation. The accuracy of the algorithm was 24% for 24 actual fault samples of JT9D engine gas circuit.