An aircraft brake control algorithm with torque compensation based on RBF neural network  

作　　者：Ning BAI Xiaochao LIU Juefei LI Zhuangzhuang WANG Pengyuan QI Yaoxing SHANG Zongxia JIAO 

机构地区：[1]School of Automation Science and Electrical Engineering,Beihang University,Beijing 100191,China [2]Science and Technology on Aircraft Control Laboratory,Beihang University,Beijing 100191,China [3]Key Laboratory of Advanced Aircraft Systems(Beihang University),Ministry of Industry and Information Technology,Beijing 100091,China [4]Research Institute for Frontier Science,Beihang University,Beijing 100191,China [5]Ningbo Institute of Technology of Beihang University,Ningbo 315800,China [6]Tianmushan Laboratory,Xixi Octagon City,Yuhang District,Hangzhou 310023,China

出　　处：《Chinese Journal of Aeronautics》2024年第1期438-450,共13页中国航空学报（英文版）

基　　金：supported by the National Natural Science Foundation of China(Grant No.52205045);the National Key Research and Development Program of China(Grant No.2021YFB2011300);the Young Elite Scientists Sponsorship Program by CAST(Grant No.YESS20200063)。

摘　　要：The wheel brake system of an aircraft is the key to ensure its safe landing and rejected takeoff.A wheel’s slip state is determined by the brake torque and ground adhesion torque,both of which have a large degree of uncertainty.It is this nature that brings upon the challenge of obtaining high deceleration rate for aircraft brake control.To overcome the disturbances caused by the above uncertainties,a braking control law is designed,which consists of two parts:runway surface recognition and wheel’s slip state tracking.In runway surface recognition,the identification rules balancing safety and braking efficiency are defined,and the actual identification process is realized through recursive least square method with forgetting factors.In slip state tracking,the LuGre model with parameter adaptation and a brake torque compensation method based on RBF neural network are proposed,and their convergence are proven.The effectiveness of our control law is verified through simulation and ground experiment.Especially in the experiments on the ground inertial test bench,compared to the improved pressure-biased-modulation(PBM)anti-skid algorithm,fewer wheel slips occur,and the average deceleration rate is increased by 5.78%,which makes it a control strategy with potential for engineering applications.

关 键 词：Aircraft landing Brakes Disturbance observer Neural networks Runway identification 

分 类 号：V249.1[航空宇航科学与技术—飞行器设计]