摘要
针对铁水硅含量无法直接在线检测的问题,提出了一种基于优化极限学习机的高炉铁水硅含量数据驱动预测模型,该模型利用差分进化算法的全局寻优能力来优化极限学习机的输入权值和隐元偏差,在此基础上建立了基于差分进化算法优化极限学习机(DE-ELM)的高炉铁水硅含量预测模型。所建模型对高炉炉温的实际调控具有较好的指导意义。
Considering the silicon content of the hot metal cannot be directly detected online, a data-driven prediction method for silicon content in hot metal based on the optimized extreme learning machines is proposed. The connection weights of inputs and biases of hidden nodes of the extreme learning machine are optimized by the differential evolution algorithm because of its global optimization ability. Based on the optimization, the prediction model of the differential evolution extreme learning machine(DE-ELM) is constructed. The proposed model provide a great guiding significance to the temperature control of the blast furnace..
出处
《有色冶金设计与研究》
2015年第3期36-38,41,共4页
Nonferrous Metals Engineering & Research
基金
国家自然科学基金重大项目(61290325)
国家自然科学基金创新研究群体科学基金(61321003)
关键词
硅含量
差分进化
极限学习机
高炉
数据
silicon content
differential evolution
extreme learning machine
blast furnace
data