Because the sintering process has complex and high-dimensional process variables and many uncertain factors, it is difficult for a single feature selection method to effectively select the best feature set, which affects the prediction accuracy of the model. Therefore, a prediction model of attention mechanism-based bidirectional gated recurrent unit model (BiGRU-Att) sinter drum index based on hybrid feature selection was proposed. Firstly, the maximum information coefficient (MIC) was used to select candidate features from the original feature set. Then, the feature selection method based on simultaneous perturbation stochastic approximation (SPSA-FS) was used to further optimize the candidate feature set. Finally, the best feature set was used as the input of BiGRU-Att to predict the sinter drum index. The results of comparative analysis with multiple models and single feature selection methods show that the hybrid feature selection method proposed in this paper can select the best feature set, and the established model has higher prediction accuracy, providing reliable decision-making support for the sintering process.
针对以上问题,本文提出了基于混合特征选择的BiGRU-Att烧结矿转鼓指数预测模型.首先对采集的烧结数据进行预处理,以得到可以直接建模的有效数据.然后使用最大信息系数(maximum information coefficient,MIC)选出与转鼓指数强相关的候选特征,再运用基于同时扰动随机逼近的特征选择方法(feature selection method based on simultaneous perturbation stochastic approximation,SPSA-FS)对候选特征进一步优化,得出最佳的特征集[8-9].在获得最佳的特征集之后,考虑到烧结过程是一个连续复杂的工业过程,为了更准确地预测转鼓指数,构建了基于注意力机制的双向门控循环单元模型(bidirectional gated recurrent unit model based on attention mechanism, BiGRU-Att),以捕捉烧结过程变量的长期依赖关系,并在每个时间步上动态地分配注意力,更好地关注对转鼓指数重要的时间点或特征[10].通过与其他几种模型和单一特征选择方法进行比较分析,表明本文提出的混合特征选择方法能够筛选出更优的特征集,并且所建模型表现出较高的预测精度.
LiuRan, ZhangZhi-feng, LiuXiao-jie, et al. Prediction of sinter drum index based on convolutional neural network and process theory[J]. Journal of Iron and Steel Research, 2023, 35(6): 651-658.
[5]
LiX, LiuX J, LiH Y, et al. Research on sinter quality prediction system based on Granger causality analysis and stacking integration algorithm[J]. Metals, 2023, 13(2): 13020419.
ZhangZhen, LiXin, LiuSong, et al. Predictive model of sinter drum index based on multi-category production status[J]. China Metallurgy, 2022,32(1): 27-35.
[8]
RenX Y, YangB, LuoN, et al. The prediction of sinter drums strength using hybrid machine learning algorithms[J]. Computational Intelligence and Neuroscience, 2022, 2022: 4790736.
HaoLiang-yuan, LiuFu-long, LiuXiao-jie, et al. Prediction model of sintered ore drum index based on deep forest algorithm[J]. Metallurgical Industry Automation, 2022, 46(4): 78-85.
[11]
XiaG L, WuZ X, LiuM Y, et al. Prediction interval estimation of sinter drum index based on light gradient boosting machine and kernel density estimation[J]. Ironmaking & Steelmaking, 2023, 50(8): 909-920.
[12]
ApolloniJ, LeguizamónG, AlbaE. Two hybrid wrapper-filter feature selection algorithms applied to high-dimensional microarray experiments[J]. Applied Soft Computing, 2016, 38: 922-932.
[13]
HsuH H, HsiehC W, LuM D. Hybrid feature selection by combining filters and wrappers[J]. Expert Systems with Applications, 2011, 38(7): 8144-8150.
[14]
DuS, WuM, ChenL F, et al. Operating mode recognition of iron ore sintering process based on the clustering of time series data[J]. Control Engineering Practice, 2020, 96: 104297.
[15]
YeniceZ D, AdhikariN, WongY K, et al. SPSA-FSR: simultaneous perturbation stochastic approximation for feature selection and ranking[EB/OL]. (2018-04-16)[2023-12-11].
ChoK, Van Merrie¨nboerB, GulcehreC, et al. Learning phrase representations using RNN encoder-decoder for statistical machine translation[EB/OL].(2014-06-03)[2023-12-11].
[18]
YamakP T, LiY J, GadoseyP K. A comparison between ARIMA, LSTM, and GRU for time series forecasting[C]//Proceedings of the 2019 2nd International Conference on Algorithms, Computing and Artificial Intelligence. Sanya, 2019: 49-55.
[19]
DuanY H, LiH H, HeM Q, et al. A BiGRU autoencoder remaining useful life prediction scheme with attention mechanism and skip connection[J]. IEEE Sensors Journal, 2021, 21(9): 10905-10914.
[20]
UmadeviT, BrahmacharyuluA, SahR, et al. Influence of sinter grate suction pressure (flame front speed) on microstructure, productivity and quality of iron ore sinter[J]. Ironmaking & Steelmaking, 2014, 41(6): 410-417.
[21]
GaoQ J, WangH, PanX Y, et al. A forecast model of the sinter tumble strength in iron ore fines sintering process[J]. Powder Technology, 2021, 390: 256-267.
[22]
YiZ M, LiuQ, QinJ Z. Optimization of sintering strength based on response surface methodology[J]. Transactions of the Indian Institute of Metals, 2021, 74(12): 3085-3092.