A Design Method of Variable Cycle Engine Controller Based on Improved Deterministic Policy Gradient Algorithm

Abstract

The invention belongs to the technical field of aero-engine control, and relates to a design method of a variable-cycle engine controller based on an improved deterministic strategy gradient algorithm. The invention adopts the sample priority playback technology and the target neural network technology to improve the deterministic strategy gradient algorithm, and uses it as the multivariable control algorithm of the variable cycle engine. The improved deterministic policy gradient algorithm can speed up the training speed of the neural network, ensure the stability of the updating process of the neural network parameters, and realize the combined adjustment of all adjustable components of the variable cycle engine. The variable cycle engine steady-state multivariable controller designed in the present invention does not need to decouple the variables and has better portability. The added sample priority playback mechanism is conducive to improving the sample utilization rate and speeding up the neural network training process. At the same time, the dual Actor-Critic structure design uses the target neural network to calculate the real neural network gradient, making the neural network training process easier to converge.

Description

technical field [0001] The invention belongs to the technical field of aero-engine control, and in particular relates to a design method of a variable-cycle engine controller based on an improved deterministic strategy gradient algorithm. Background technique [0002] The aero-engine provides the power needed for the aircraft to fly and is an essential equipment for the aircraft to perform various flight tasks, while the control system of the aero-engine is an important guarantee for the stable, reliable and safe operation of the engine. Compared with the traditional aero-engine, the variable-cycle engine can change the thermodynamic cycle process of the engine in a wide range by adjusting multiple variable geometric components, and realize the free switching of supersonic, high-thrust, subsonic, and low fuel consumption modes. It can be regarded as a combined device of turbojet engine and fan engine, which integrates the advantages of turbofan and turbojet engine, and is th...

AI Technical Summary

Problems solved by technology

The control system of traditional fuel closed-loop control and geometric mechanism open-loop scheduling cannot meet the control requirements of variable cycle engines, and there are problems such as the need to establish a linear model with small deviations, fewer control variables, and complex controller structures.

Images