Hot Metal Temperature Prediction at Basic-Lined Oxygen Furnace (BOF) Converter Using IR Thermometry and Forecasting Techniques

Abstract

In oxygen steelmaking, the charge calculation strongly depends on hot metal temperature prediction. Although a hot metal temperature drop from the blast furnace in a steel plant may be too complex to be accurately modeled in detail, the combined use of sensors and statistical models can improve temperature estimation and result in better cost, quality and productivity, as well as lower emissions. In order to develop a simple but robust method for hot metal temperature forecasting, the suitability of infrared thermometry and time series forecasting has been studied. Simultaneous infrared thermometer measurement and video recording was used for designing the processing of the thermometer signal. The resulting temperature estimations are in good agreement with disposable thermocouple measurements giving an error of 11 °C with 60% reliability (chances of obtaining a successful output). Conversely, the time series approach was based mainly on the AutoRegressive Integrated Moving Average (ARIMA) model in which five additional process variables were introduced as exogenous predictors, as well as using a moving window of past observations for continuous model training. The resulting error was 15 °C with more than 90% reliability. Combining measuring and modeling approaches reduced the error to 13 °C with 100% reliability, thereby providing a hybrid procedure that has long-term stability and is self-adaptive to varying production scenarios.

Hot Metal Temperature Forecasting at Steel Plant Using Multivariate Adaptive Regression Splines

“…In a previous modelling work, time series were applied to solve this problem. It was shown that the introduction of exogenous predictors into an auto-regressive integral moving average (ARIMA) model improved its performance [21]. Conversely, the potential benefit of introducing the L last observations of the predicted variable as input variables will be analyzed in the present work.…”

“…Since the hot metal pouring may extend over a significant period of time, there is not a clear time limit between torpedo and ladle. Considering that the heat losses in torpedoes and ladles were found to be similar [21], both holding times can be grouped together without excessive simplification.…”

“…The past measured values, Y, are used for model training and to assess the performance of model predictions,Ŷ. More details for variable selection and their characteristics are given in Díaz et al [21].…”

“…The GCV penalty per knot was chosen as d = 2, as suggested by Hastie [44] for additive models. Finally, the initial choices for w and L, were based on previous experiences of the authors [21,22].…”

“…In a previous work, the authors studied the feasibility of infrared thermometry and time series forecasting for accurate prediction of the hot metal temperature at the steel plant [21]. A combination of both methods proved to be reliable and long-term stable, as well as self-adaptive to changing production scenarios.…”