Method for online building of state variable model of aircraft engine based on precise partial derivative

Abstract

The invention discloses a method for online building of a state variable model of an aircraft engine based on a precise partial derivative. The method comprises the following steps of step A, determining an expression form of the state variable model of the aircraft engine; step B, acquiring a uniform discrete state variable form capable of expressing steady-state points and dynamic points based on Taylor expansion according to the expression form of the state variable model; step C, determining intermediate variables in a partial derivative computation process according to variables in the discrete state model needing to be established, and adopting an analysis method to establish a corresponding partial derivative model on the basis of aircraft engine component-level models and a chain type derivative rule; and step D, combining the component-level models and the partial derivative model to carry out joint computation through a multiple-pass algorithm, acquiring a state variable model coefficient matrix and an initial value of a current working point, and acquiring a state variable model of the corresponding working point. According to the method, the accurate state variable model can be computed on line at any working point.

Description

technical field [0001] The invention belongs to the field of system simulation and control in aerospace propulsion theory and engineering, and specifically relates to an online establishment method of an aeroengine state variable model based on accurate partial derivatives. Background technique [0002] The mathematical model of aero-engine has great value in aero-engine control, simulation and engineering application. Due to its simple calculation and certain theoretical support, the state variable model is representative in the mathematical model of aeroengines. [0003] In view of the importance of the state variable model, two main methods for establishing the state variable model of an aeroengine have been developed, namely the fitting method and the partial derivative method. The fitting method includes using genetic algorithm, least square method and other optimization algorithms to directly fit the corresponding experimental data. The partial derivative method main...

AI Technical Summary

Problems solved by technology

If the model used is accurate enough, it can effectively estimate the output change curve of several moments after the current moment. If the prediction model is obtained by linear interpolation of finite state variable models or parameter scheduling, the accuracy of unmodeled points is difficult to guarantee. , and it is difficult to adapt to many uncertainties of the engine, which destroys the advantages of the MPC control algorithm

Images