A self-learning framework for adaptive free-form pipe routing in aeroengines
This study proposes a self-learning framework (SLPR) to optimize free-form pipe routing in aeroengines, tackling the NP-hard problem of complex piping. SLPR uses Proximal Policy Optimization (PPO) integrated with a unified rule modeling framework and a potential energy table. This allows the system to learn optimal routes through interaction and rapidly adapt to changing engine designs in dynamic environments.