A Self-learning Algorithm for Space Environment Temperature Control

This paper proposes a parameter self-learning proportional periodic temperature control algorithm that is specific to the characteristics of space environment temperature change. The algorithm is able to identify system parameters and control parameters, and comes up with proper control strategies with respect to environment temperature. It can be applied to objects whose system parameters are unknown, and has the self-adaptive ability to environment temperature change. The application of the self-learning algorithm gets rid of manual intervention completely, and realizes intelligent control. The simulation result shows that this algorithm achieves excellent performances for temperature control with high precision and self-adaptive, and can meet the high precision temperature control requirements, and reduces the heat equilibrium burden of heat control subsystem imposed by on-board devices.