Semi-Physical Simulation of Fan Rotor Assembly Process Optimization for Unbalance Based on Reinforcement Learning

An aero engine fan rotor is composed of a multi-stage disk and blades. Excessive unbalance the after assembly main cause vibration. In process, blade sequencing optimization set phase are important for reducing overall unbalance. To address this problem, paper proposes semi-physical simulation method based on reinforcement learning to optimize balance in process. Firstly, mass moments individual blades, diagonal moment difference introduced as constraint build single-stage sorting model, used find optimal path so that optimal. Then, basis initial set, model established, Finally, collection data during rotor, least-squares fit calculate real-time achieve The feasibility effectiveness proposed verified by experiments.