STATE DETECTION OF OXYGEN GENERATION SYSTEM IN SPACE STATION VIA STOCHASTIC PROJECTION KALMAN FILTER

The oxygen generation system by using electrolytic water in space station is a key physical and chemical recycling equipment. Its operating state directly determines the activation frequency of the standby oxygen cylinder and endangers the life safety of the astronauts. For high-precision control of oxygen production and system fault diagnosis, we need to obtain high-accuracy estimation of the inner state of the system, e.g. partial oxygen pressure in the electrolysis unit. In this paper, a novel state estimation algorithm named stochastic projection Kalman filter is proposed. Compared with EKF and UKF algorithm, the new algorithm can obtain higher estimation precision and better convergence.