Abstract: In order to solve the initial values problem and improve model accuracy dramatically in establishment of turbofan engine SVM using fitting method, a state variable modeling (SVM) method was investigated based on linear least square optimized fitting. The partial derivative method (PDM) was improved in terms of iteration computation in Components Level Model (CLM) to get the initial values. The linear least square optimized fitting method was studied to establish the SVM of turbofan engine according to the law that the dynamic responses of the linear model should be in accordance with that of the nonlinear model at the same steady working point. The results from improved fitting method (IFM) were compared with those from improved PDM (IPDM) and CLM. The results show that the IFM can ensure the convergence of the model, obtain better modeling precision and adapt to the high-level systems modeling. 

Key words: Turbofan engine; State variable model;Partial derivative method;Fitting method

摘要： 为解决拟合法建立涡扇发动机状态变量模型时初值难以选取的问题和有效提高模型的精度,研究了一种改进的优化拟合方法。针对拟合法初值难以选取的问题,采用改进偏导数法获得的优化迭代初值；根据发动机稳态工作点线性化模型的动态响应与该点处的非线性模型动态响应一致的原则,采用线性最小二乘优化拟合法建立某型涡扇发动机状态变量模型。通过在不同状态工作点、采用不同模型结构,与改进偏导数法和非线性部件级模型仿真结果相比较表明:该方法能有效保证模型收敛性,具有较高的建模精度,并适用于高阶系统建模。

关键词: 涡扇发动机;状态变量模型;偏导数法;拟合法 