Control Algorithm On The Sling's Tension Of Low-Gravity Simulating System

With the rapid development of the deep space exploration, space detectors play an important role in it. Following the international routine, the space detector should be checked to see whether its power system can meet the practical demand before it is launched. Therefore, the large low-gravity simulating system is the only choice to check the power system's performance. The subject origins from one of China Aerospace Science and Technology's projects; it aims to design a full control algorithm for the tension servo system which is the large low-gravity simulating system's subsystem to meet its performance demand.The tension servo system is the key subsystem of the large low-gravity simulating system. It is used to reduce partially the space detector's gravity, and thus it can simulate the low-gravity environment on the ground. First, the paper overviews the general low-gravity simulating system inside and outside our authry and highlights the slung low-gravity simulator's characters. Based on the performance requirement, we design a proper servo mechanism and then analyze how it works. Using the system identification and mathematical modeling but ignoring the nonlinearities, we acquired the MIMO system's nominal model.For the nonlinearities in the servo system, we are focused on the nonlinearities which affect the system's performance mostly; they are the sling's resonance, mechanical resonance and Coulomb friction, and then the paper lists their approximate mathematical models. As there are many nonlinearities in the MIMO system, thus we abort the traditional PID control and adopt H∞control theory to design control algorithm which is conducive to the realization of the coordinated control of the two AC torque motors. To overcome the disadvantages which is due to the Coulomb friction, on the base of the H∞algorithm we add the disturbance observer of the friction compensation in order to obtain the desired performance. As the order of the H∞? controller is quite high and it can't be directly applied in the project, applying the balance reduction of model's order technology to the high order controller, we can obtain the lower order controller which is approximate to the original controller and can be directly applied in the project. Therefore it solves the disadvantage of the original controller. At last the paper introduces the framework and final realization of the real time control software in the practical project.