【摘要】 对钢包精炼炉(以下缩写为LF)终点温度进行精确预测，有助于降低能源消耗，提高冶炼效率。为提高LF终点温度预测的准确性，先使用孤立森林算法检测并剔除数据集中的异常值，再使用随机森林算法分析各影响因素与终点温度的相关性，最后建立基于麻雀搜索算法优化的BP神经网络模型。基于国内某钢厂LF生产车间实际生产数据对模型进行验证，并与其他模型进行对比。结果表明，文章提出的模型预测的终点温度精度在±5℃内的炉次占比达到了92%，优于其他方法，可以为LF实际生产温度控制提供参考。更多还原

【Abstract】 As a critical secondary refining equipment in steel production,the LF( Ladle Furnace)plays a key role in temperature regulation. Accurate prediction of the LF endpoint temperature helps reduce energy consumption and improve refining efficiency. To improve the prediction accuracy of the LF endpoint temperature,the Isolation Forest algorithm was first used to detect and remove outliers from the dataset. Then,the Random Forest algorithm was applied to analyze the correlations between various influencing factors and the endpoint temperature. Finally,a BP( Back Propagation) neural network model optimized by the Sparrow Search Algorithm was developed. The model was validated using actual production data from the LF workshop of a domestic steel plant and compared with other models. The results show that the proposed model achieves a prediction accuracy within ± 5 ℃ for92% of the data in the workshop dataset,outperforming other methods. This model can serve as a reference for temperature control in actual LF production.更多还原