Turbofan Thrust Prediction Method and Controller Based on Neural Network

Abstract

The invention discloses a turbofan engine thrust prediction method and controller based on a neural network. The method includes: establishing a steady-state multi-input multi-output neural network model of key gas path measurement parameters of the turbofan engine; using the neural network model As a predictive model, a thrust model predictive controller is designed; using the model predictive controller, a closed-loop control system of a turbofan engine is designed, and the tracking performance and constraint performance of the thrust control system are studied. On the basis of considering the existing control constraints of the engine, the present invention uses the steady-state neural network model combined with the SQP optimization algorithm, aims at the minimum error of the engine's unmeasurable thrust performance parameters, designs a thrust prediction controller, and calculates the engine control quantity. The control method The structure is simple, and it plays a positive role in reducing the design workload of the controller and suppressing the degradation of the thrust performance of the controller.

Description

technical field [0001] The invention belongs to the field of aero-engine simulation and control, and in particular relates to a neural network-based turbofan engine thrust prediction method and a controller. Background technique [0002] Due to the continuous improvement of aero-engine technology, the performance requirements of the engine control system are also getting higher and higher. The traditional control method satisfies all constraints at the expense of engine performance, and it is difficult to meet the increasingly higher performance requirements of the engine. Compared with traditional control methods, model predictive control (Model Predictive Control, MPC) can not only deal with the problems of time delay and limit constraints of the system, but also realize optimal control based on the predictive model, so that the control system has good control performance, and Meet control needs. [0003] The coupling of mechanical, fluid and thermal effects in an aeroen...

AI Technical Summary

Problems solved by technology

[0005] The technical problem to be solved by the present invention is to aim at the defects of the background technology, design a turbofan engine thrust prediction controller based on a steady-state multiple-input multiple-output neural network model, and use it to solve the problems of engine constraints, nonlinear characteristics and dynamic model-based Problems with long response times for predictive controllers with good tracking and constraint performance

Images