Research on multi-source data-driven modeling method for aero-hydraulic system parts oriented to adaptive control

As the core power transmission and control unit of modern aircraft, the performance of aviation hydraulic system is directly related to flight safety and operational stability. Especially under the extreme working conditions of high pressure, high frequency and high load, the dynamic characteristics of hydraulic system parts are strongly nonlinear, time-varying and multi-physical field coupling. The traditional modeling method based on mechanism model faces challenges, and it is difficult to accurately describe the actual working conditions and capture the nonlinear dynamic characteristics under extreme working conditions. Therefore, a multi-source data-driven modeling method for aero-hydraulic system parts oriented to adaptive control is proposed in this study. In this method, multi-source dynamic data (pressure, flow, temperature and vibration signals) of hydraulic system parts are collected synchronously, and fuzzy rules are extracted by combining subtraction clustering and fuzzy C-means algorithm (FCM), and a dynamic model based on T-S (Takagi-Sugeno) fuzzy model is constructed. At the same time, the model-free adaptive control (MFAC) strategy is introduced, and the Jacobian matrix in the traditional model is replaced by pseudo-gradient estimation to realize real-time optimization of control parameters. The experimental results show that the T-S fuzzy model has high prediction accuracy under various working conditions, and the root mean square error (RMSE) is significantly better than the traditional mechanism model. The MFAC control strategy has the advantages of low overshoot, short adjustment time and small steady-state error in the sinusoidal command test of actuator position tracking, showing stronger dynamic adaptability and robustness. This research provides theoretical support and technical scheme for high-precision modeling and adaptive control of aviation hydraulic system.