Loading...

Satellite Attitude Actuator Fault Detection and Identification Using Adaptive Neuro-Fuzzy Inference System (ANFIS)

Moosavi, Saba | 2023

32 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 56780 (45)
  4. University: Sharif University of Technology
  5. Department: Aerospace Engineering
  6. Advisor(s): Asadian, Nima
  7. Abstract:
  8. This thesis focuses on the detection and estimation of faults occurring in one of the reaction wheel axes of a three-axis stabilized satellite. The wheel axes are aligned with the satellite's body axes, and the emphasis is on the attitude (rather than position) of the satellite. The primary goal of the subsystem is to determine and control the satellite's attitude, transitioning from its initial state to a desired Earth-pointing state. It is assumed that environmental disturbances, including gravitational disturbances and Drag related disturbances, are considered. Based on this, a PID controller is designed after modeling the nonlinear kinematics and dynamics of the satellite using Euler angles and quaternions. Then the performance of the control system in the absence of a fault has been checked by the simulation results. As the next step by adding a fault to the actuator, an Adaptive Neuro-Fuzzy Inference System designed to detect and estimate the fault. During several trials and errors and numerous analyzes on the results obtained at each stage, a system was improved and trained which, while having the least complexity and as a result the fastest response, has the best and most acceptable results. By testing the fault occurrence modes in different wheel axes and considering various values for it, and trying to diagnose this fault By ANFIS the accuracy of the fault detection system was confirmed. Finally, the robustness of this method checked in several cases. The first case is when maneuver is changed from nadir pointing to slew maneuver, the second case is when there is an uncertainty in the products of inertia, and the third case is when there is an uncertainty in the diagonal moments of inertia. In all three cases, the performance of this trained fault estimation system was checked and confirmed on the satellite. In general, the result of this research in the use of Adaptive Neuro-Fuzzy Inference System for detecting and estimating the fault of reaction wheel actuator in a satellite has been favorable
  9. Keywords:
  10. Divert and Attitude Control System (DACS) ; Attitude Determination ; Reaction Wheel (RW) ; Fault Detection and Diagnosis (FDD) ; Adaptive Neuro-Fuzzy Inference System (ANFIS) ; Fault Detection and Isolation (FDI) ; Satellite Attitude Control

 Digital Object List

 Bookmark

No TOC