Font Size: a A A

Research On Fault Diagnosis For Micro-Nano Spacecraft Swarm

Posted on:2018-06-19Degree:MasterType:Thesis
Country:ChinaCandidate:J Y WangFull Text:PDF
GTID:2322330536487914Subject:Navigation, guidance and control
Abstract/Summary:PDF Full Text Request
With the increasing development of space technology,it is important to ensure the functionality,stability and intelligence of spacecraft for complicate flying mission.The concept of micro-nano spacecraft swarm has been put forward to meet these demands and the intelligent fault diagnosis technology plays an important role in the spacecraft swarm system.Considering the micro-nano spacecraft swarm system as the object and the fault diagnosis technology as the core,the research of intelligent fault detection algorithm for swarm and the design of formation reconfiguration are respectively carried out in this paper.The main tasks are as follows:(1)A new fault detection method for subsystem of spacecraft swarm based on wavelet neural network observer is proposed.By the research of feedforward wavelet neural network with single hidden layer,the proposed observer is designed and the stability of the observer is proved.Then,the fault of attitude control system is studied as the research object and the residual generating method of the wavelet neural network observer are used to determine whether there is a fault.(2)A fault detection method for swarm system based on dynamic neural network is proposed.In order to solve missed detection problem of faulty spacecraft,the relative data of adjacent spacecraft are used for realizing the fault detection system combining the mathematical model of reaction wheel with strong learning ability and nonlinear approximation property of dynamic neural network.(3)The formation reconfiguration is proposed for fault reconstruction of swarm.When faulty spacecraft is found in swarm,the linear programming method is used to make faulty spacecraft fly away and backup spacecraft fly into the swarm for realizing the formation reconfiguration task by considering the fuel optimal and collision avoidance problem.The simulation results show that the fault occurred in subsystem and swarm system can effectively be detected by the proposed intelligent fault detection method.And formation configuration can be carried out by STK and MATLAB simulation after the fault occurred.The proposed fault diagnosis scheme for swarm system is proved to be valid and feasible,which can ensure the reliability and stability of the flying mission.
Keywords/Search Tags:Micro-Nano spacecraft swarm, fault diagnosis, fault detection, wavelet neural network, dynamic neural network, formation reconfiguration
PDF Full Text Request
Related items