Abstract:  Aiming at the difficulty of multi-person collaborative operation modeling and the low efficiency of disassembly sequence solution in the maintenance of large and complex equipment, the disassembly model of the asynchronous parallel disassembly sequence planning and solving algorithm were studied, and an asynchronous parallel disassembly sequence planning method based on an improved gravitational search algorithm (GSA) was proposed. Firstly, the disassembly work area interference and geometric constraints between components were studied to establish a priority constraint model. Based on the priority relationship and the disassembly target parts, the minimum set of parts to be disassembled was searched backward. Then, the mathematical model of asynchronous parallel disassembly sequence planning was established, and the objective function was built with the optimization goal of minimizing disassembly time. Encoding and decoding methods were constructed for the asynchronous parallel disassembly sequences. In addition, the updated formula of the individual evolution in GSA was reconstructed, and the escape operator was designed to jump out of the local optimum. Finally, taking the hydraulic turbine main shaft seal as an example, and the effectiveness of the proposed method was proved by comparing with the traditional algorithm. The results show that improved GSA achieves less time cost than comparison algorithms and is more effective in dealing with complex equipment disassembly.
Key words:  mechanical equipment maintenance; gravitational search algorithm(GSA); combination optimization; multi-person collaborative operation; model for disassembly; solving algorithm