Satellite group fault diagnosis method based on multi-information fusion

• Main Authors: ZHANG XIUYUN, ZONG QUN, DOU LIQIAN, LENG JIAJUN, LIU WENJING
• Format: Patent
• Language: Chinese; English
• Published: 27.02.2024
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention discloses a satellite group fault diagnosis method based on multi-information fusion, which is applied to the technical fields of reinforcement learning, federal learning and multi-agent. Comprising the following steps: acquiring sensor data during satellite operation; establishing a satellite attitude dynamic model under the influence of the flywheel fault according to the acquired sensor data, and establishing a virtual simulation environment; a satellite local fault diagnosis network model is established and trained; cooperative training of a satellite local fault diagnosis network model is carried out by using a federal center aggregation means, and aggregation and updating are completed; and obtaining a fault diagnosis result according to the updated satellite local fault diagnosis network model. According to the satellite group fault diagnosis method based on federated learning, in consideration of limited types of faults generated during operation of a single satellite in a formation, in