Scholarlink

Abstract <p>The energy consumption of silicon manganese ore blast furnace is a key comprehensive indicator of its production efficiency and operational economy. Addressing the difficulty in directly measuring this metric, this study employs total flue gas heat as an indirect proxy. A series of predictive models—including Linear Regression, XGBoost, ADABoost.R2, ELM, LightGBM, and LSBoost—was initially developed to conduct a comparative benchmarking analysis. The comparison results show that the XGBoost model achieved the best overall performance on the test set (RMSE&#xa0;=&#xa0;0.048645, R<InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(^{2}\)</EquationSource> <EquationSource Format="MATHML"><math> <mmultiscripts> <mrow /> <mrow /> <mn>2</mn> </mmultiscripts> </math></EquationSource> </InlineEquation>&#xa0;=&#xa0;0.89717). To further improve its prediction accuracy, random search and Bayesian optimization were used as benchmarks, and four optimization algorithms, HPKO, COA, FSA, and IVY, were introduced for hyperparameter tuning. The XGBoost model optimized by the HPKO algorithm has the best performance, with the RMSE reduced to 0.037258 and <InlineEquation ID="IEq2"> <EquationSource Format="TEX">\(\hbox {R}^{2}\)</EquationSource> <EquationSource Format="MATHML"><math> <msup> <mtext>R</mtext> <mn>2</mn> </msup> </math></EquationSource> </InlineEquation> increased to 0.93998. Actual production data verification confirms that the optimization model can effectively predict the energy consumption level of silicon manganese ore furnaces, providing a reliable basis for optimizing the smelting process and demonstrating the potential application of machine learning combined with intelligent optimization algorithms in this industrial scenario.</p> Graphical Abstract <p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Research on a Machine Learning-Based Energy Consumption Prediction Model for Silicon Manganese Smelting Furnaces

  • 期刊论文

  • 发表日期：

    2026年6月18日

查看全文
  • Xiangnan Xiong,
  • Xuebo Chen,
  • Changxin Li,
  • Yuan Gao,
  • Zhengjun Yu,
  • Guangchuan Ma

摘要

Abstract

The energy consumption of silicon manganese ore blast furnace is a key comprehensive indicator of its production efficiency and operational economy. Addressing the difficulty in directly measuring this metric, this study employs total flue gas heat as an indirect proxy. A series of predictive models—including Linear Regression, XGBoost, ADABoost.R2, ELM, LightGBM, and LSBoost—was initially developed to conduct a comparative benchmarking analysis. The comparison results show that the XGBoost model achieved the best overall performance on the test set (RMSE = 0.048645, R \(^{2}\) 2  = 0.89717). To further improve its prediction accuracy, random search and Bayesian optimization were used as benchmarks, and four optimization algorithms, HPKO, COA, FSA, and IVY, were introduced for hyperparameter tuning. The XGBoost model optimized by the HPKO algorithm has the best performance, with the RMSE reduced to 0.037258 and \(\hbox {R}^{2}\) R 2 increased to 0.93998. Actual production data verification confirms that the optimization model can effectively predict the energy consumption level of silicon manganese ore furnaces, providing a reliable basis for optimizing the smelting process and demonstrating the potential application of machine learning combined with intelligent optimization algorithms in this industrial scenario.

Graphical Abstract