Research On Aircraft Anti-skid Brake Control System Based On T-S Fuzzy Neural Network

Aircraft brake system is an important equipment of an aircraft, which has a crucial impact on the safety of aircraft taking-off and landing. With the development of aviation industry, the standard of brake system performance has been set higher. Improving the control law of anti-brake is a significant way to improve brake performance. An adaptive fuzzy neural network brake controller based on Takagi-Sugeno fuzzy neural network is presented aiming at the characteristic of aircraft anti-skid brake system.Paper first analyzes the development of the domestic and international aircraft brake system and anti-skid control algorithm. After studying the working principle of anti-skid brake system, paper compares several common control methods, selects the control method based on the slip rate. On the basis of analyzing principle of anti-skid brake system, numeric model of every part of hydraulic anti-skid brake system is constructed in the paper, including dynamic model of aircraft, dynamic model of wheel, tire and runway model, landing gear model, hydraulic system model and brake equipment model. Considering the nonlinearity and uncertainty of braking process, paper analyzes the advantages and disadvantages of fuzzy control and neural network and then designs an adaptive fuzzy neural network brake controller based on Takagi-Sugeno fuzzy neural network. This kind of adaptive fuzzy neural network combines the ability of fuzzy control dealing with parameter uncertainty and the adaptive ability of neural network, so the network can handle the problems mentioned before therefore enhance the performance of anti-skid brake system.Finally, the numeric model of the proposed system and the adaptive control algorithm has been verified and simulated in MATLAB/Simulink software. Braking performance on dry runway, wet runway, ice runway and mixed runway is analyzed. The results have been compared to the traditional PID+PBM algorithm’s results. Paper also gives the noise analysis and dynamic performance analysis, the results demonstrate that the adaptive fuzzy neural network controller based on T-S fuzzy neural network algorithm can improve the performance of anti-skid brake system obviously.