Research Article
CNN-LSTM-based Study on the Dynamic Characteristics of Steel Protection Slag Crystallization
@INPROCEEDINGS{10.4108/eai.2-6-2023.2334668, author={Sisi Dong and Zilong Cheng and Ziwei Cheng}, title={CNN-LSTM-based Study on the Dynamic Characteristics of Steel Protection Slag Crystallization}, proceedings={Proceedings of the 2nd International Conference on Information Economy, Data Modeling and Cloud Computing, ICIDC 2023, June 2--4, 2023, Nanchang, China}, publisher={EAI}, proceedings_a={ICIDC}, year={2023}, month={8}, keywords={steel protective slag crystallization behavior dynamic features cnn-lstm feature sequences}, doi={10.4108/eai.2-6-2023.2334668} }- Sisi Dong
Zilong Cheng
Ziwei Cheng
Year: 2023
CNN-LSTM-based Study on the Dynamic Characteristics of Steel Protection Slag Crystallization
ICIDC
EAI
DOI: 10.4108/eai.2-6-2023.2334668
Abstract
To improve steel production efficiency and product quality, enhance market competitiveness, secure the development of the steel industry, and reduce the high cost and loss brought by manual operations to the steel industry, this paper proposes a CNN-LSTM-based model for studying the dynamic characteristics of protection slag crystallization in steel. For the accuracy and practicality of protection slag crystallization prediction, the dynamic characteristics of protection slag crystallization are studied at the data level, the CNN-LSTM network model is built and the LSTM model is used to capture the temporal information of the behaviour and to predict the protection slag crystallization process. The experimental results show that the CNN-LSTM-based model can better analyse and predict the crystallization behaviour using sequence images. Compared with the traditional supervised and unsupervised learning methods, the model has higher accuracy and practicality.
