无人机涡喷发动机的神经网络自适应PID控制

推进技术 ›› 2003, Vol. 24 ›› Issue (6): 517-520.

无人机涡喷发动机的神经网络自适应PID控制

马静,杨育武,王镛根

西北工业大学航空动力与热力工程系;陕西西安710072;西安远东公司军品研究所;陕西西安710072;西北工业大学航空动力与热力工程系;陕西西安710072

发布日期:2021-08-15

Neural network PID adaptive control in pilotless aircraft turbojet engine

Dept. of Aeroengine Engineering, Northwestern Polytechnical Univ., Xi’an 710072, China;The Military Products Research Inst. of Xi’an YuanDong Company, Xi’an 710073, China;Dept. of Aeroengine Engineering, Northwestern Polytechnical Univ., Xi’an 710072, China

Published:2021-08-15

摘要/Abstract

摘要： 将基于RBF神经网络的辨识网络与基于BP网络的控制器相结合,组成自适应PID神经网络控制系统。RBF神经网络采用离线学习在线修正权值和阈值,为加快收敛速度,应用带惯性项的梯度下降法。大量仿真结果表明,RBF网络较ELM,标准BP及改进的BP等网络具有明显优点。对某型无人机涡喷发动机控制系统的仿真结果表明此控制方式具有鲁棒性好、响应速度快、稳态误差小等优点。

关键词: 无人驾驶飞机;涡轮喷气发动机;人工神经元网络;自适应控制

Abstract: The work presented here aims establishing the self-adaptive PID neural network control on the basis of the combination of the identification network of the RBF neural network and the controller of the BP neural network. RBF neural network adopts the off-line training and the on-line adaptation of weight and threshold value. In order to speed up the convergence, the grads descent method with inertia item was used. Massive stimulation proves the superiority of the RBF network to the ELM and the standard and the ameliorated BP. Stimulation to the pilotless aircraft turbojet engine proves several advantages of this controlling method, such as good robustness, sensitive response, and minimal stable error.

Key words: Pilotless aircraft;Turbojet engine;Artificial neural network;Adaptive control

马静,杨育武,王镛根. 无人机涡喷发动机的神经网络自适应PID控制[J]. 推进技术, 2003, 24(6): 517-520.