One stop service for steel industry

Electric arc furnace end temperature forecast

Developing the short process of electric arc furnace steelmaking is an important strategic way to realize the green development of the steel industry. The end point control of electric arc furnace steelmaking determines the quality of steel tapping and smelting efficiency, especially the end point temperature control. Establishing an electric arc furnace end-point temperature prediction model to predict the end-point temperature in advance will help to adjust the smelting process in a timely manner and achieve fast and efficient tapping operations.

Electric arc furnace end temperature prediction models are mainly divided into mechanism models and data-driven models. Data-driven models are the current main research direction. However, the existing data-driven model modeling process relies on a large amount of historical data and is difficult to achieve accurate prediction of end-point temperature under small sample data conditions. In this regard, a highly adaptable electric arc furnace end temperature prediction model was established based on the metallurgical mechanism and artificial intelligence algorithm as the core.

The end temperature of the electric arc furnace directly affects the quality of molten steel during tapping and the progress of the refining process. The temperature of molten steel generally requires a brief pause in the smelting process and is measured using a thermocouple. However, multiple measurements will slow down the production pace, and establishing a prediction model can realize soft measurement of the end temperature of molten steel, thereby quickly and accurately meeting steel tapping requirements, shortening the smelting cycle, and improving production efficiency.

Generally speaking, when the end-point temperature prediction error is within ±5°C, the hit rate needs to reach 90% and above to meet the needs of production forecasting. There is still room for improvement in the accuracy of existing models. Most data-driven models use large samples of historical data, and there is less research on small sample data. Some steel plants with newly built electric arc furnaces lack the accumulation of large amounts of data. Therefore, it is a big challenge to establish a high-precision data-driven model when the amount of data is small. In addition, compared with BOF, LF and other steelmaking equipment. At present, the mechanism of electric arc furnace end temperature prediction is poorly integrated with data-driven models. Some scholars only outline the smelting process or briefly analyze the smelting mechanism when modeling, and do not achieve a close combination of mechanism analysis and data-driven modeling.

Therefore, this article will aim to establish a high-precision electric arc furnace end-point prediction model under small sample data. The input parameters of the model will be obtained through mechanism and data correlation analysis. In order to solve the problem that the FCNN algorithm is prone to over-fitting, an early stopping strategy will be introduced for optimization. , improved the applicability and accuracy of the model, and established an end-point temperature prediction model based on the e-FCNN algorithm.

electric arc furnace temperature

1) The energy flow mechanism of the electric arc furnace was analyzed, and the influencing parameters of the end point temperature were obtained. Combined with Pearson correlation coefficient analysis, the model input parameters were determined. On the basis of the FCNN algorithm, an early stopping strategy is introduced for optimization, and a forecast model based on the e-FCNN algorithm is constructed to achieve accurate prediction of the end point temperature under small sample data conditions. When the end-point temperature prediction error of the optimal e-FCNN model is within ±5, ±6, and ±8°C, the hit rate is 93.3%, 96.6%, and 100%.

2) Compare the end-point temperature prediction effects of various machine learning algorithms. The results show that e-FCNN performs best, and ε-SVR and RF also basically meet the needs. Therefore, e-FCNN, ε-SVR and RF can provide algorithm references for building end-point temperature prediction models.

3) Use the optimal e-FCNN end-point temperature prediction model to continuously track the actual production data of 30 heats. The results show that the model has high applicability, the prediction error is within ±6°C, and the hit rate is 96.7%, and it can guide the electric arc furnace tapping operation.

4) Future work will further improve the integration of the mechanism model and the data-driven model, expand the input dimensions of the model, and introduce variables that can represent the submerged arc effect. Research the advanced optimization method of hyperparameters of the model, expand the training samples, and improve the prediction accuracy of the electric arc furnace end temperature.

Facebook
Twitter
LinkedIn
Reddit

As professional one-stop solution provider, LIAONING MINERAL & METALLURGY GROUP CO., LTD(LMM GROUP) Established in 2007, and focus on engineering research & design, production & delivery, technology transfer, installation & commissioning, construction & building, operation & management for iron, steel & metallurgical industries globally. 

Our product  have been supplied to world’s top steel manufacturer Arcelormittal, TATA Steel, EZZ steel etc. We do OEM for Concast and Danieli for a long time.

LMM Main product series

Need For Some Help ?

Professional engineers provide solutions and technical drawings

phone

86(411) 84174804

mail

lmme-business@lmmgroup.com.cn

Special product design, please send specific data and drawings to our mailbox or form.

As professional one-stop solution provider, LIAONING MINERAL & METALLURGY GROUP CO., LTD(LMM GROUP) Established in 2007, and focus on engineering research & design, production & delivery, technology transfer, installation & commissioning, construction & building, operation & management for iron, steel & metallurgical industries globally. 

Our product  have been supplied to world’s top steel manufacturer Arcelormittal, TATA Steel, EZZ steel etc. We do OEM for Concast and Danieli for a long time.

LMM GROUP Certification

LMM GROUP Service

Get A Free Consultation
And Estimate