An intelligent online detection approach based on big data for mechanical properties of hot-rolled strip
by JinXiang Chen; Ziming Fan
International Journal of Modelling, Identification and Control (IJMIC), Vol. 37, No. 2, 2021

Abstract: An LightGBM prediction model based on big data is presented in order to online detect the mechanical properties of hot-rolled strip in this paper, which can achieve the greater accuracy than both the existing prediction approaches and hardware detection method for the local strips. A dataset of mechanical properties of hot-rolled strip is constructed firstly by collecting a steel plant's hot-rolled process control parameters, which includes 17,000 samples, and every sample contains 17 input characteristics and three output mechanical property parameters. Based on the dataset, an LightGBM intelligent prediction model is established and trained to predict the three mechanical properties of the hot-rolled strip steels. 17,000 data of hot rolling mill are used to verify the effectiveness of the model. Results show that the prediction accuracy for tensile strength, compressive strength and elongation are 0.99971, 0.99835, and 0.99631, respectively. Especially, the prediction accuracy for elongation is higher than the existing methods.

Online publication date: Tue, 11-Jan-2022

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Modelling, Identification and Control (IJMIC):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com