CCRL is a research framework for continual fault diagnosis in aero-engines. It integrates contrastive representation learning with reinforcement learning strategies to support incremental fault pattern discovery while maintaining stable diagnostic performance on previously learned conditions.
