Novel gray wolf optimization algorithm-based aero-engine nonlinear predictive control method

Abstract

The invention relates to a novel gray wolf optimization algorithm-based aero-engine nonlinear predictive control method. The method includes the following steps of: prediction model establishment: Gaussian white noises act on an aero-engine as input data, so that corresponding output data are obtained, as for the acquired input data and output data, a BP neural network method is utilized to offline train a corresponding neural network model, and a prediction model is established through using a recursive method on the basis of the neural network model; feedback correction: feedback correction is performed on the output value of the prediction model according to the deviation of the output value of the prediction model at a k time point and the actual output value of the engine; and rolling optimization: a novel gray wolf optimization algorithm is adopted to perform solving, with the difference of the output value of the prediction model and the output fixed value of the engine adopted as input, rolling optimization is performed, and therefore, optimal control quantity fuel flow can be obtained. Compared with an ordinary aero-engine control method, the method has better robustness and constraint processing capacity.

Description

technical field [0001] The invention belongs to the technical field of aero-engine control, and in particular relates to a non-linear predictive control method for an aero-engine based on a novel gray wolf optimization algorithm. Background technique [0002] The aero-engine system is a complex aerodynamic thermodynamic system with a wide range of work and complex working conditions. In the actual system, there are inevitably a large number of uncertainties such as disturbances and unmodeled dynamics. The aero-engine control system is required to have a strong Robustness, and in practice, there will be many physical constraints in the aero-engine system, which requires the control system to have good processing constraints. Predictive control can take into account the above two points and get a good control effect. The important factor to ensure the performance of predictive control is the optimization ability of the algorithm used in rolling optimization. Therefore, in pred...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a non-linear predictive control method for aero-engines based on a novel gray wolf optimization algorithm with high convergence, fast calculation speed, and strong processing constraints, which effectively solves the problem of processing constraints in aero-engine predictive control. Difficult to guarantee requirements and real-time

Images