Research On The Control Technology For The Interferometer Subsystem In Space-borne Infrared Spectrometer

Fourier transform spectrometer (FTS) has been widely used in the atmospheric sounding as a powerful detector, which is important for the meteorological analysis and ecological conservation. The performance of the FTS is limited by the optical path difference and the stability of the optical path difference velocity (OPDV) in the interferometer, and its reliability in space. In order to improve the FTS's performance, the hardware and simulation platform for a rotary interferometer are designed in this thesis. Moreover, the control strategies for the OPDV and the methods of fault-tolerance in space have also been studied.Firstly, the principle of the FTS and its control requments are instrouced. Then, the simulation platform composed with the optics, machinery, electron and control is built in the Synopsys Saber, which provides the foundation for the study of the control strategies and space environment.In order to control the OPDV high stably in the rotary interferometer with time-varying parameters, we propose a fuzzy integral sliding mode control (FISMC) in this thesis. The steady track error of the OPDV is reduced by the integral in the sliding mode contol (SMC). Moreover, the complexity of the fuzzy rules is simplified by using the sliding mode surface as the input of the fuzzy logic control. In addition, the chattering is suppressed effectively with the help of the output of the fuzzy logic control, which is used to regulate the switch control of the SMC, and the stability of the OPDV is also improved.On basis of the characteristic modeling, we propose a characteristic model-based adaptive discrete-time sliding mode control (ADSMC). By using the characteristic modeling, a characteristic model including the OPDV is modeled in the discrete-time domain. Its performance is equal with the original system, and the discretization effect on the SMC is also reduced. Then, the SMC is designed based on this model, and the time-varying parameters are estimated by the gradient method. Under the control of ADSMC, the output of the control is smooth, and the tracking accuracy of the OPDV in the discrete-time domain is further improved. We model the disturbance moments caused by the microgravity and microvibrograph after analyzing their influences on the interferometer, and design the control approaches based on the above two SMCs. Then, both of the maximum immunities are estimated by the simulations, and the performance of the ADSMC is better than that of FISMC.Focusing on the transient faults occurring in space, such as anticipated failures, unanticipated failures and single event upset, a two-level neural network-based intelligent fault tolerant control is proposed. At the performace level, a model-based error detection is designed for fault detection. The anticipated failures and unanticipated failures are diagnosed by the pattern recognition and RBF neural network, respectively. The SMC designed by reaching law is used to handle the failures. At the code level, the control flow errors are detected by modifying the key codes with CFCSS and closed-loop detecting at the performance level. By using this approach, the disadvantage of the practical application of CFCSS can be ignored. The safety of the interferometer is improved.Finally, the harware platform for the control of the rotary interferometer is designed. Then, several experiments using the ADSMC are carried out on this platform, and its effectivenesses are verified.