A novel aircraft brake disturbance recognition model
Zongxia Jiao,Ning Bai,Dong Sun,Xiaochao Liu,Yaoxing Shang,Shuai Wu
DOI: https://doi.org/10.1016/j.ast.2020.106337
IF: 5.457
2020-12-01
Aerospace Science and Technology
Abstract:The aircraft relies on the brake system to decelerate during landing. Compared with the air, the aircraft is not only affected by the aerodynamic force, but also by the ground nonlinear disturbance. At the same time, signals that aircraft brake controller can use for real-time control are limited, and the poor working environment of the wheel speed sensor makes it difficult to guarantee the accuracy. These disturbances directly affect the braking efficiency of the aircraft. This article generalizes different types of disturbance based on the response of the control. The brake system signals are statistically analyzed to extract two characteristic variables that are most sensitive to disturbance. Aircraft brake disturbance recognition model (ABDRM) consisting of data pre-processor, continuous hidden Markov classifier and Bayesian filter is proposed by means of probability. It can recognize the state of the aircraft when it is disturbed during braking. Furthermore, it is logically optimized for compatibility with existing brake control law. The data collected by simulation under different disturbance conditions and inertial test bench is used for verification. These results indicate that the model is effective and accurate in principle and engineering, especially in the disturbance of slow variance. This provides opportunities for further optimization of the aircraft brake anti-disturbance control law.
engineering, aerospace