Quality parameter adaptive optimization for spinning process using dynamic non-dominated sorting algorithm

Abstract

Intelligent textile equipment can discover potential patterns in the production process through data mining, and utilize these patterns through intelligent optimization, ultimately achieving intelligent and automated textile production. This paper focuses on the spinning process parameters optimization under changing spinning conditions and proposes a dynamic non-dominant ranking parameter quality adaptive optimization algorithm. Firstly, the changing factors of spinning conditions is analyzed to explore the influence of the change of spinning process conditions on the quality optimization of spinning parameters. Then the factors of spinning process condition changes is transformed into mathematical dynamic constraints and constructing an adaptive optimization model for spinning parameter quality. Basis on this, the response mechanism of spinning environment is established to readjust the optimization direction according to the change of spinning conditions, and the DNSGA-II is used to solve the quality adaptive optimization model. A case study is designed to validate the effectiveness of the proposed method. Compared with other data-driven methods, the proposed method offers enhanced performance in terms of providing corresponding parameter optimization combinations for different spinning conditions.