数 字 孪 生 驱 动 的 航 空 发 动 机 转 子 精 密 堆 叠 装 配

The extremely high structural complexity of the aero-engine，together with its strict requirements of high performance and stability pose big challenges to aero-engine manufacturing technology. Assembly is a key procedure of aero-engine manufacturing that directly affects product quality，wherein rotor is the key power component of aeroengine，in that its assembly accuracy and quality have great influence on aero-engine performance. Rotor is as⁃ sembled by multiple parts in series，and the assembly error can be arisen by such multiple factors as error accumula⁃ tion and connection structure deformation. Aiming at the issue of aero-engine rotor assembly error，a data-driven as⁃ sembly error optimization method is studied. The mechanism of assembly process is analyzed combining the digital twin technology. Research is carried on digital twin assembly optimization technology from three aspects，including the overall technical framework of rotor assembly unit digital twin，digital analysis model of assembly process，and com⁃ prehensive evaluation and decision methods of assembly process. A digital analysis model for assembly process is proposed integrating analysis models of multi physical factors. Then data-driven rotor assembly process monitoring and process optimization can be realized. Through application verification of rotor assembly，the proposed optimization method reduces the assembly error by 57.4% compared to the maximum possible error. The digital twin driven rotor assembly optimization method is then proved to be effective for practical problems，providing a new approach for solv⁃ ing the quality prediction and optimization problems of aero-engine rotor assembly.