Multi-population genetic algorithm for seamless steel tube blank design considering delivery time

Seamless steel tube blank design is the link between customer orders and actual production. For the customer-oriented seamless steel tube blank design problem, delivery time is a key factor as well as traditional round billet constraints and process constraints. This paper focuses on its short-round...

Full description

Saved in:
Bibliographic Details
Published in2022 15th International Symposium on Computational Intelligence and Design (ISCID) pp. 25 - 28
Main Authors Chen, Lu, Wang, Bailin, Yuan, Shuaipeng, Li, Tieke, Zhang, Wenxin
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.12.2022
Subjects
Billets
blank design
Decoding
delivery time
Electron tubes
Encoding
genetic algorithm
multi-objective optimization
parallel evolution
Production
Sociology
Steel
Online AccessGet full text

Cover

More Information
Summary:Seamless steel tube blank design is the link between customer orders and actual production. For the customer-oriented seamless steel tube blank design problem, delivery time is a key factor as well as traditional round billet constraints and process constraints. This paper focuses on its short-round billet cutting problem to minimize the number of round blanks and the penalty of delivery time. According to the characteristics of the problem, a multi-population genetic algorithm is proposed. Encoding and decoding strategies are presented in terms of decision variable features. According to the multi-objective characteristics, it proposes a multi-population parallel evolution strategy, where the population is divided into one main population with multi-objective optimization and two auxiliary populations for each single objective, thereby enhancing the global search ability. The feasibility and effectiveness of the proposed algorithm are verified by different scale data experiments.
ISSN:2473-3547
DOI:10.1109/ISCID56505.2022.00013