Scholarlink

<p>According to the calculation of a 300t dephosphorization converter in a steel plant, the utilization rate of top-added lime is 74.07%, while that of bottom-injected lime powder is 82.64%. Thus, the influencing factors of converter lime utilization rate are studied. For the first time, multiple linear regression is used to quantitatively analyze the degree of influence of multiple factors on lime utilization rate, including hot metal temperature, [Si] content in molten iron, slag basicity, (FeO) content in slag, and (MgO) content in slag. The explicit functional relationship between lime utilization rate and each influencing factor is obtained, and the evaluation model of lime utilization rate is established. Based on this, a prediction model of lime utilization rate is established by using BP (back propagation) neural network. In the prediction model of conventional process, the accuracy rate of absolute errors within ± 0.04 and ± 0.08 is 61.53% and 84.63%, respectively. In the prediction model of the bottom blowing O<sub>2</sub>-CO<sub>2</sub>-CaO process, the accuracy rate is 70% and 90% when the absolute error is within ± 0.03 and ± 0.06.</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

Evaluation and Prediction Model of Lime Utilization Rate of Conventional/Bottom Blowing O2-CO2-CaO Process in Dephosphorization Converter

  • 期刊论文

  • 发表日期：

    2025年4月1日

查看全文
  • Xin Ren,
  • Chunyang Wang,
  • Guangsheng Wei,
  • Chao Feng,
  • Rong Zhu,
  • Kai Dong

摘要

According to the calculation of a 300t dephosphorization converter in a steel plant, the utilization rate of top-added lime is 74.07%, while that of bottom-injected lime powder is 82.64%. Thus, the influencing factors of converter lime utilization rate are studied. For the first time, multiple linear regression is used to quantitatively analyze the degree of influence of multiple factors on lime utilization rate, including hot metal temperature, [Si] content in molten iron, slag basicity, (FeO) content in slag, and (MgO) content in slag. The explicit functional relationship between lime utilization rate and each influencing factor is obtained, and the evaluation model of lime utilization rate is established. Based on this, a prediction model of lime utilization rate is established by using BP (back propagation) neural network. In the prediction model of conventional process, the accuracy rate of absolute errors within ± 0.04 and ± 0.08 is 61.53% and 84.63%, respectively. In the prediction model of the bottom blowing O2-CO2-CaO process, the accuracy rate is 70% and 90% when the absolute error is within ± 0.03 and ± 0.06.