SAC-Optimized Active Disturbance Rejection Controller for Thermal Regulationin Aerospace Applications

Efficient temperature regulation is essential for maintaining the sensitivity and noise performance of spaceborne infrared detectors under varying observation modes.This paper proposes an Active Disturbance Rejection Controller (ADRC) with dynamic parameter tuning based on the Soft Actor-Critic (SAC) reinforcement learning algorithm, aiming to optimize the temperature control performance and disturbance rejection capability of spacecraft systems.A thermal transfer model was established, and the Soft Actor-Critic (SAC) algorithm was introduced to dynamically tune parameters (system gain, controller bandwidth and observer bandwidth) by maximizing a weighted sum of the reward and policy entropy.This approach enables real-time estimation and compensation of total disturbances.Simulation results demonstrate that, in temperature tracking tasks, the proposed method reduces the integral of absolute error (IAE) by 14.61% and 14.87%, and shortens the settling time by 36.36% and 58.47%, compared to fixed-parameter ADRC and PID controllers, respectively.Under external periodic disturbances, the proposed controller improves control accuracy by 15.6% and 13.6%.Monte Carlo robustness tests further show that, under ±5% parameter perturbations, the method exhibits small fluctuations in IAE, settling time, and overshoot.The proposed SAC-ADRC strategy provides a promising solution for rapid and high-precision thermal regulation of infrared detectors in aerospace applications.