共查询到19条相似文献,搜索用时 78 毫秒
1.
2.
3.
4.
5.
6.
7.
“双碳”背景下,提升焦炭质量是保证钢铁行业高质量发展的研究重点之一,而炼焦行业存在着在线实时监测难、焦炭质量预测模型泛化能力差等问题。为此,提出一种通过自适应全局搜索算法,即改进鲸鱼优化算法(WOA)与长短期记忆(LSTM)循环神经网络综合建模的方法来解决这一问题。首先选取出配合煤中可反映焦炭质量的可测参数,再运用主成分分析(PCA)去除变异性小的冗余因子后,得到预测因子,将其作为LSTM网络的外部输入;通过加入自适应惯性权重以及最佳扰动更新改进WOA,从而训练LSTM网络的超参数,采用均方根误差(RMSE)和R-squared 进行算法检验;最后将改进后的AGWOA-LSTM模型与典型的LSTM、WOA-LSTM模型进行对比,以验证本方法的优越性。结果表明AGWOA-LSTM模型预测焦炭质量具有精度高、运行速度快等特点。研究对焦炭生产具有一定的理论指导意义。 相似文献
8.
通过关联配合煤中硫、灰分、挥发分的质量分数等主要的煤质指标,利用基于Adma算法为优化器的GRU神经网络模型,不断对模型参数进行调整后,通过sigmoid激活函数判断模型准确率的多标签多分类方法,建立了焦炭质量预测模型。结果表明:当三层GRU网络的隐层神经元数量为(64,64,64);学习率为0.01;样本批次大小为64;样本训练次数为50;丢弃率为0.3时,得到了模型的最优参数,此时模型预测准确率达到97%。采用GRU神经网络多标签多分类焦炭预测模型不仅具有高精度、低损失函数等特点,而且针对小样本配煤数据预测焦炭质量可以达到很好的效果,对实际的配煤炼焦具有一定的参考意义。 相似文献
9.
随着高炉的大型化和富氧喷煤技术的应用,对焦炭质量提出了更高的要求,利用单种煤和配合煤的各种试验室测定的性能指标来预测焦炭质量,可以减少小焦炉配煤试验,以最少的试验室投入,确 相似文献
10.
基于支持向量机的焦炭质量预测模型 总被引:3,自引:2,他引:3
采用机器学习方法中的支持向量机技术来预测焦炭质量,不需要了解煤成焦机理,综合考虑配合煤特性和焦炉加热制度的影响,而且克服了人工神经网络预测精度低的缺点,在取得最小拟合误差的同时可以得到最小的预测误差;给出了一种基于组态王6.0与VB6.0的焦炭质量在线预测系统框架. 相似文献
11.
通过对西北地区煤和内地部分煤种的基本性质、灰成分分析,以及对西北煤配煤后所得的焦炭进行焦炭反应性和光学组织的研究,探讨了灰成分对焦炭性能和微观结构的影响.在分析无机矿物质的催化作用基础上,研究了高碱金属含量煤的炼焦机理及焦炭特殊性的原因. 相似文献
12.
13.
14.
15.
16.
17.
An approach to blast furnace coke quality prediction 总被引:3,自引:0,他引:3
Although coke cold drum mechanical strength has historically been the most relevant coke quality parameter, currently coke reactivity and post-reaction strength (CRI/CSR) are the most important parameters used to assess blast-furnace coke quality. Many models of coke quality prediction have been proposed, most of which are based on coal characteristics and limited to the same coal geographic origin, but as yet there is no universally applicable prediction formula. The present work describes a simple model of coke CRI/CSR prediction based on the assumption that the CSR of a coke produced from a blend of coals can be predicted from the CSR obtained from the cokes of the individual coals through the application of the additivity law. The additivity law was also applied to the coke cold mechanical strength indices derived from the Irsid test, which are widely employed by the European coke industry as complementary coke quality indicators. 相似文献
18.
Nonlinear model predictive control based on support vector machine and genetic algorithm 总被引:1,自引:0,他引:1
This paper presents a nonlinear model predictive control (NMPC) approach based on support vector machine (SVM) and genetic algorithm (GA) for multiple-input multiple-output (MIMO) nonlinear systems. Individual SVM is used to approximate each output of the controlled plant. Then the model is used in MPC control scheme to predict the outputs of the controlled plant. The optimal control sequence is calculated using GA with elite preserve strategy. Simulation results of a typical MIMO nonlinear system show that this method has a good ability of set points tracking and disturbance rejection. 相似文献