This content is not included in your SAE MOBILUS subscription, or you are not logged in.

One Approach to the Aircraft Brake Control System Numeric Identification Method

Journal Article
2015-01-2693
ISSN: 1946-391X, e-ISSN: 1946-3928
Published September 27, 2015 by SAE International in United States
One Approach to the Aircraft Brake Control System Numeric Identification Method
Citation: Novakovic, N., "One Approach to the Aircraft Brake Control System Numeric Identification Method," SAE Int. J. Commer. Veh. 8(2):302-309, 2015, https://doi.org/10.4271/2015-01-2693.
Language: English

Abstract:

Aircraft anti-skid brake control system is considered one of the most complex aircraft systems whose performance depends not only on subsystem parameters but rather on many other external conditions and physical parameters which are difficult to control and predict. Over the years aircraft brake control system performance and fault diagnostics have been simulated and analyzed from various aspects. Based on the task to enhance aircraft brake control system diagnostic methods, this article presents one approach to mathematical modeling and a numeric identification method of the hydro-mechanical brake control components.
For any complex system behavioral or performance analysis approach, system modeling and simulation are the most common tools. Most often, the complete system model is unknown, and only simple segments of the unknown system or a small number of subsystem components may be known in a form of transfer function with static and dynamic characteristics. For that reason, mathematical modeling and system identification methods have evolved and become a part of greater control systems theory.
In this article, using a time domain input and output parameters from flight test aircraft, hydro-mechanical elements of the brake control system have been modeled as linear, time-invariant dynamic subsystem with unknown constant parameters. For the model parameters identification, a numeric algorithm has been developed and implemented based on Lüders-Narendra's adaptive observer. Finally, simulated and real system dynamic responses were compared and evaluated. With regards to dynamic performance, the results of the simulation demonstrate the model is stable and accurate in comparison with real system test data.