In terms of the fault diagnosis for non-stationary engineering systems,the cointegration coefficients matrix was proposed as the characteristic parameter if the system variables are integrated of first order.System variables should first be verified to be integrated of first order through unit root test,then Johansen method was used to estimate the cointegration coefficients matrix.The classification algorithm of supporting vector machine (SVM)was used to train the model and test the accuracy of classification.Thus,a new approach applied in non-stationary system was established towards finding a new characteristic parameter for fault diagnosis.This new approach has been applied to the fault diagnosis of a simulated hydraulic flap servo system,using 5 system variables including input orders,flap angle,etc.as the cointegration variables.The test results indicate that the cointegration coefficients matrix model has great fault diagnosis ability for typical non-stationary systems.%对于非平稳工程系统，若系统变量是一阶单整的，则可以建立变量间的协整关系模型，通过理论分析证明，协整系数矩阵可作为故障空间的特征参数用作系统的故障诊断。非平稳系统变量首先要通过单位根检验以证明其一阶单整，然后使用 Johansen 检验法估计出协整系数矩阵，即特征参数，使用一对多分类的支持向量机（SVM）算法进行故障数据的训练和测试。使用液压舵面故障仿真系统作为试验平台，使用输入指令、舵面角度等5个一阶单整的系统变量作为协整变量，其估计出的协整系数矩阵作为特征参数，结果表明协整系数矩阵作为特征参数，SVM作为数据分类方法，具有很好的故障分类效果。
展开▼
