Craft Parameters Optimization of Melt-Transportation in Polyester Fiber Production Based on Improved RVEA

摘要

Melt-transportation is an important process in polyester fiber production containing multiple production processes and complicated structure, its craft parameters are set generally based on engineer's experience. Multi-objective optimization is proved to be an efficient method for such problems. And we built a five-objective optimization model based on five performance indicators of melt-transportation process to optimize the craft parameters. Due to the increase of the objectives' dimensions, algorithms based on non-dominated relationship cannot select individuals with good convergence and diversity for the population. Therefore, we improve the angle penalized functions of RVEA, where we adjust the penalized functions according to the diversity of the current population. The experiment results show that the improved RVEA performs competitively compared with the more challenging SDTLZ1-SDTLZ4 test suite based on DTLZ series, and the result of melt-transportation optimization is of guiding value for polyester fiber production.