Optimization Study on Hardness of Cold Rolled Strip Steel Based on a Data-Driven Approach

Siyi Du; Sihan Qiu

doi:10.62051/xrg7jz77

Authors

Siyi Du
Sihan Qiu

DOI:

https://doi.org/10.62051/xrg7jz77

Keywords:

Cold rolled strip steel, Hardness optimization, Pearson correlation coefficient, Genetic Algorithm.

Abstract

As a key index to measure product quality and processability, the hardness of cold-rolled strip steel has an important influence on the production efficiency of iron and steel enterprises. However, due to the coupling between the process parameters at each stage of the continuous annealing process, and the setting of traditional process parameters depending on the experience of the operator, the lack of scientific and systematic optimization methods, which easily leads to the problem of high hardness fluctuation and low resource utilization. In this study, the data-driven method is used to conduct a preliminary correlation analysis to further clarify the optimal combination of process parameters for achieving the target hardness. The Pearson correlation coefficient is used to determine that carbon content, quenching furnace temperature, and strip speed significantly affect strip hardness. To further clarify the optimal combination of process parameters to achieve the target hardness, a Genetic Algorithm (GA) was tried in this study. The results show that the optimized process parameter combination can effectively improve the stability and consistency of strip hardness, and provide a scientific basis for the process parameter setting in the cold rolled strip production process, which is helpful for enterprises to improve product quality, reduce cost, and enhance market competitiveness.

Downloads

Download data is not yet available.

References

[1] Guo Y Z, Huo J W, Zhang X H. Modern strip quality inspection technology[J]. Engineering Construction, 2023, 6(10): 122-124.

[2] Fei S C, Li Z Q, Jiang X X. Effect of annealing temperature and cold rolling reduction on annealing texture of low carbon aluminum killed steel[J].Mining and Metallurgical Engineering,2022,42(01):149-151.

[3] Yang Q F, Lai X Z, Du S, et al. Prediction of oxygen consumption in converter steelmaking based on particle size clustering[J]. Acta Automatica Sinica, 2024, 50(1): 132-142.

[4] Wang Y, Sun Y K. Optimization method of rolling parameters of tandem cold rolling mill based on immune genetic algorithm[J]. Information and Control, 2002, 31(2): 189-192.

[5] Ding J G, Guo J H. Width prediction of hot strip based on principal component analysis and random forest algorithm[J]. Journal of Northeastern University (Natural Science), 2021, 42(9): 1268.

[6] Qiu S F, Feng F Y, Xu W T. Research on missing value prediction using multiple models[J]. Journal of Science and Engineering Technology, 2021, 17(1): 31-40.

[7] Yang J, Hu Q, Xiao T, et al. Energy efficiency modeling, process parameter optimization, and sequencing modeling optimization of tandem cold rolling unit based on energy consumption[J]. China Mechanical Engineering,2020,31(14):1724-1732.

[8] Li L G, Liu J, Cao K, et al. Intelligent process diagnosis system for wide strip cold rolling[J].China Metallurgy,2024,34(03):139-148.

[9] Wang Z X, Zhao H T, Xie Y H, et al. Prediction of modulus of composite materials by backpropagation neural network combined with genetic algorithm[J]. Journal of Shanghai Jiaotong University, 2022, 56(10): 1341.

[10] Zhang X H, Guo Y Z, Huo J W. Analysis of strip surface on-line inspection and quality control technology[J]. Engineering Construction, 2023, 6(11): 136-138.

[11] Xuan G N, Cheng R W. Genetic Algorithm and engineering optimization[M]. Tsinghua University Press Limited, 2004.