An intelligent multi-objective optimization method for transverse profile grinding processes of large shafts
Xuekun Li,Meng Ye,Zihan Tang,Liping Wang,Jiayu Pan,Dong Wang
DOI: https://doi.org/10.1007/s00170-024-14309-w
IF: 3.563
2024-09-08
The International Journal of Advanced Manufacturing Technology
Abstract:Large shaft grinding processes are required to generate multiple quality indices with good consistency, such as profile dimension accuracy, surface roughness, and optical glossiness. Generating profiles for large shafts in the grinding process requires the grinding wheels to move transversely under the desired path, which includes rough grinding, semi-finish grinding, finish grinding, and spark-out grinding stages, and each grinding stage needs to go through multiple passes under different process parameters. Therefore, the design and selection of process parameters for each grinding stage is always challenging to meet both quality and efficiency requirements. In this study, a multi-objective optimization method for transverse profile grinding processes of large shafts is developed. In the proposed method, multiple grinding quality indices and grinding efficiency are taken as optimization objectives, and grinding surface quality, grinding depth, and grinder machining performance are taken as constraints. By utilizing the dynamic factor improvement strategy and the adaptive grid density improvement strategy, an improved particle swarm algorithm is developed to optimize the grinding process parameters. The grinding dimensional error compensation method and the surface quality consistency control method are integrated into the improved particle swarm algorithm to realize the accurate prediction of grinding dimensional accuracy and surface quality. Based on the proposed method, the optimized grinding process can be obtained, and the effectiveness of the method is verified by grinding experiments.
engineering, manufacturing,automation & control systems