基于改进极限学习机的转炉出钢合金化锰收得率预测模型

来源期刊:中南大学学报(自然科学版)2021年第5期

论文作者:刘青 周凯啸 林文辉 孙建坤 冯小明 方炜

文章页码:1399 - 1407

关键词:转炉;出钢合金化;元素收得率;正则化极限学习机;改进粒子群算法;预测模型

Key words:converter; tapping alloying; element yield; regularized extreme learning machine; improved particle swarm algorithm; prediction model

摘    要:针对转炉炼钢出钢合金化过程合金的加入量偏差较大的问题,为更精确地控制合金加入量,以某钢厂冶炼HRB400钢出钢过程加入硅锰合金为例,建立基于极限学习机算法的Mn元素收得率预测模型,并引入正则化方法和改进粒子群算法(IPSO)对极限学习机算法进行优化,以提高模型的泛化能力和预测精度。研究结果表明:Mn元素收得率预测相对误差在5%和3%以内的命中率分别为95%和80%,准确性高于BP神经网络及人工经验的预测结果。照此种方式控制硅锰合金加入量可以满足成品钢的成分要求,且每炉次硅锰合金加入量较人工经验值平均减少20 kg,可带来每年400万元的经济效益,能够为现场生产提供参考。

Abstract: Due to the problem of large deviations in the addition of alloys during the alloying process of converter steelmaking, in order to more accurately control the amount of alloy added, taking a steel plant smelting HRB400 steel tapping process adding silicon manganese alloy as an example, a prediction model of Mn yield was established based on extreme learning machine algorithm, and regularization methods and improved particle swarm optimization(IPSO) were introduced to optimize the extreme learning machine to improve the generalization ability and prediction accuracy of the model. The results show that the prediction hit rates of the Mn element yield within relative error of 5% and 3% are 95% and 80%, respectively, and the accuracy is better than the prediction results of BP neural network and artificial experience. On this basis, the amount of silicon-manganese added in this way can meet the composition requirements of the finished steel, and the amount of silicon-manganese alloy added per furnace is reduced by 20 kg on average compared with the artificial experience value, which can bring the economic benefit of 4 million yuan per year. This can provide a reference for on-site production.

相关论文

  • 暂无!

相关知识点

  • 暂无!

有色金属在线官网  |   会议  |   在线投稿  |   购买纸书  |   科技图书馆

中南大学出版社 技术支持 版权声明   电话:0731-88830515 88830516   传真:0731-88710482   Email:administrator@cnnmol.com

互联网出版许可证:(署)网出证(京)字第342号   京ICP备17050991号-6      京公网安备11010802042557号