提出了将经济学领域中的协整理论应用于动态(非平稳)采用协整理论分析了发动机在正常工作状态下若干非平稳输出变量,工程系统的状态监测与故障诊断问题;基于汽车发动机仿真数据,验证了这些变量在正常工作状态下满足协整关系,即长期均衡的关系;进一步模拟了发动机输出轴扭矩传感器故障。在这种故障状态下,发现发动机的各检测变量之间的协整关系已被破坏;运用变量轮换-降阶协整模型法,确定了转矩变量是破坏这种协整关系的源头;结果表明协整理论在动态工程系统状态监测与故障诊断领域有良好的应用前景。
Applying the method of cointegration, which is a new algorithm developed in economics area for modelling non-stationary economical and financial processes, to condition monitoring and fault diagnosis for non-stationary systems. Based on the data from the professional engine simulate software, dyno2003, and using method of cointegration testing to analysis the engine's several output variables which under the normal working condition, then, validate these variables can pass the cointegratiun test, in other words, these non-stationary variables have long-term balanced relation. Further, simulate the engine output shaft sensor's fault, find that the cointegration relation between output variables under this condition has been destroyed; utilizing method of variable exchange to confirm the torque variable is the cause of losing the Cointegration relation. The results demonstrate that the cointegration test method has a bright future in engineering area for condition monitoring and fault diagnosis for dynamic engineering systems.