Long short term comprehensive prediction of sinter FeO components based on EEMD and machine learning
ZHANG Zhen1, TANG Jue1, CHU Mansheng1, LIU Zhenggen1, LI Fumin2, LÜ Qing2
1. School of Metallurgy, Northeastern University, Shenyang 110819, Liaoning, China; 2. College of Metallurgy and Energy, North China University of Technology, Tangshan 063210, Hebei, China
Abstract：Iron making depended on "70% of raw materials, 30% of operation". Sinter was the main raw material for blast furnace iron making, and FeO composition was an important index affecting the reduction, strength and particle size of sinter ore, as well as an important factor affecting blast furnace iron production and fuel ratio.Therefore, timely and accurate FeO content of sinter ore played a significant role in guiding the smooth production of blast furnace ironmaking. To tackle the problems of delayed detection results and poor accuracy of FeO composition of sinter ore, an comprehensive long and short-term prediction model of FeO composition by EEMD and machine learning was proposed and established. Exploratory analysis was conducted for sintering data, and according to the characteristics of sintering data, box line diagrams and sliding windows were used to process the data to ensure the data value and to consolidate the data foundation for modeling. The comprehensive model consists of two modules. Long-term prediction model. Applying EEMD (ensemble empirical mode decomposition)to decompose fluctuating FeO component data, which could reduced the complexity of input data, and perform advance forecasting of FeO components within 3 h with Bi-LSTM(bi-directional long short-term memory). Short-term prediction module. Integrating EEMD, feature selection and extraction methods to construct derived features and enhance the learning ability of the model for input and target data to ET(extra-trees) the FeO composition prediction for the next hour. Under the validation of the unknown sintered data test set, it was found that EEMD-assisted machine learning modeling could significantly improve the accuracy and stability of FeO composition prediction, and the MAPEand MSE of EEMD-Bi-LSTM and EEMD-ET models were around 1% and 0.027 with errors close to zero values. The highest hit rate of the prediction interval could reach more than 94%, and the trend of FeO composition prediction was consistent with the real situation. This result contributed to achieve accurate advance control of FeO composition trends and values.
张振, 唐珏, 储满生, 柳政根, 李福民, 吕庆. 基于EEMD和机器学习的烧结矿FeO成分长短期综合预报[J]. 钢铁, 2023, 58(8): 32-40.
ZHANG Zhen, TANG Jue, CHU Mansheng, LIU Zhenggen, LI Fumin, LÜ Qing. Long short term comprehensive prediction of sinter FeO components based on EEMD and machine learning[J]. Iron and Steel, 2023, 58(8): 32-40.
[1] 牛乐乐, 刘征建, 张建良, 等. 铁矿粉矿物组成对烧结矿冶金性能的影响[J]. 钢铁, 2019, 54(9):27. (NIU L L, LIU Z J, ZHANG J L, et al. Effect of mineral composition of iron ore powder on metallurgical properties of sintering ore[J]. Iron and Steel, 2019, 54(9):27). [2] 郄亚娜, 李宇壮, 白诗皓, 等. 碱度对富矿粉厚料层烧结固相反应及矿物生成的影响[J].中国冶金, 2023, 33(2):38. (QIE Y N, LI Y Z, BAI S H, et al. The influence of alkalinity on sintering solid phase reaction and mineral formation of rich mineral powder[J]. China Metallurgy, 2023,33(2):38.) [3] 单长冬, 张建良, 王耀祖, 等. 富氧条件下高精粉配比对烧结指标及冶金性能的影响[J].钢铁, 2023, 58(1):22. (SHAN C D, ZHANG J L, WANG Y Z, et al. The influence of high precision powder ratio on sintering index and metallurgical properties under oxygen-rich conditions[J]. Iron and Steel, 2023,58(1): 22.) [4] 陈小龙,高立华.富氢还原对烧结矿低温还原粉化的影响[J].冶金能源,2022,41(4):27.(CHEN X L ,GAO L H. Effect of rich hydrogen reduction on low temperature reduction disintegration of sinters[J]. Energy for Metallurgical Industry,2022,41(4):27.) [5] 刘颂, 赵亚迪, 甘丽, 等. 烧结系统智能制造与大数据技术应用探讨[J]. 钢铁, 2021, 56(10):54. (LIU S, ZHAO Y D, GAN L, et al. Discussion on intelligent manufacturing and application of big data technology in sintering system [J]. Iron and Steel, 2021,56 (10): 54.) [6] 孙卫华, 焦吉成, 李率民, 等. 热连轧钢材力学性能在线预报的研究与应用[J]. 钢铁, 2022, 57(8):168. (SUN W H, JIAO J C, LI S M, et al. Research and application of on-line prediction of mechanical properties of hot-rolled steel[J]. Iron and Steel, 2022,57(8): 168.) [7] 张振, 李欣, 刘颂, 等. 基于多类别生产状态的烧结矿转鼓指数预测模型[J]. 中国冶金, 2022, 32(1):27. (ZHANG Z, LI X, LIU S, et al. Prediction model of sinter turn drum index based on multi-category production status[J]. China Metallurgy, 2022,32(1): 27.) [8] 范天骄,柯长松,徐永权,等.基于物质-能量流协同的铁水智能调度系统研究与开发[J].冶金能源,2022,41(3):9.(FAN T J,KE C S,XU Y Q, et al. Research and development of intelligent dispatching system for molten iron based on matter-Energy flow coordination[J]. Energy for Metallurgical Industry,2022,41(3):9. [9] LIU S, ZHANG Z, ZHAO J, et al. Research and application of the sintering whole process intelligent manufacturing system[J]. Ironmaking and Steelmaking, 2023: 1.DOI:10.1080/03019233.2023.2168929. [10] 王龙, 冀秀梅, 刘玠. 人工智能在钢铁工业智能制造中的应用[J]. 钢铁, 2021, 56(4):1. (WANG L, JI X M, LIU J. Application of artificial intelligence in intelligent manufacturing in the iron and steel industry[J]. Iron and Steel, 2021, 56(4): 1.) [11] 李宏扬, 刘小杰, 李欣, 等. 高炉炼铁工业互联网平台的应用[J]. 钢铁, 2021, 56(9):10. (LI H Y, LIU X J, LI X, et al. The Application of the internet platform for the blast furnace ironmaking industry[J]. Iron and Steel, 2021,56(9): 10.) [12] LIU S, LIU X, LÜ Q, et al. Comprehensive system based on a DNN and LSTM for predicting sinter composition[J]. Applied Soft Computing, 2020(95): 106574. [13] 赵军, 李宏扬, 李欣, 等.基于智能模型的高炉煤气利用率预测[J]. 中国冶金, 2021, 31(3):93. (ZHAO J, LI H Y, LI X, et al. Forecast of blast furnace gas utilization rate based on the intelligent model[J]. China Metallurgy, 2021, 31(3): 93.) [14] 刘小杰, 张玉洁, 李欣, 等.基于大数据的高炉炉温监测预警系统[J]. 中国冶金, 2023, 33(2):98. (LIU X J, ZHANG Y J, LI X, et al. Blast furnace temperature monitoring and early warning system based on big data[J]. China Metallurgy, 2023, 33(2): 98.) [15] 刘颂, 赵亚迪, 张振, 等.基于集成学习的高炉压差预报模型研究[J]. 电子测量技术, 2022, 45(2):31. (LIU S, ZHAO Y D, ZHANG Z, et al. Research on blast furnace differential pressure prediction model based on integrated learning[J]. Electronic Measurement Technology, 2022, 45(2): 31.) [16] 张智峰, 刘小杰, 李欣,等.基于MIV-GA-BP模型预测烧结矿FeO成分[J].中国冶金, 2022,32(10):75. (ZHANG Z F, LIU X J, LI X, et al. Prediction of sinter FeO composition based on the MIV-GA-BP model[J].China Metallurgy, 2022,32(10):75.) [17] 王宏涛, 储满生, 赵伟, 等. 高炉炼铁低碳化和智能化技术发展现状[J].河北冶金, 2018(9):1. (WANG H T, CHU M S, ZHAO W, et al. The development status of low carbonization and intelligent technology of blast furnace ironmaking [J]. Hebei Metallurgy, 2018(9):1.) [18] 李壮年, 储满生, 柳政根, 等. 基于机器学习和遗传算法的高炉参数预测与优化[J]. 东北大学学报(自然科学版), 2020, 41(9):1262. (LI Z N, CHU M S, LIU Z G, et al. Prediction and optimization of blast furnace parameters based on machine learning and genetic algorithm[J]. Journal of Northeastern University(Natural Science), 2020, 41(9):1262.) [19] 郝良元,刘福龙,牛佳星,等.提高高炉智能控制水平的分析与思考[J].冶金能源,2021,40(5):55. (HAO L Y, LIU F L, NIU J X,et al.Analysis and thinking on improving intelligent control level of BF[J]. Energy for Metallurgical Industry, 2021,40(5):55.) [20] LI H, LIU X, LI X, et al. Causality analysis and prediction of blast furnace state based on convergence cross mapping[J]. Ironmaking and Steelmaking, 2022, 49(9):875. [21] LIU S, ZHAO Y, LI X, et al. An online sintering batching system based on machine learning and intelligent algorithm[J]. ISIJ International, 2021, 61(8): 2237. [22] LIU S, LYU Q, LIU X J, et al. Synthetically predicting the quality index of sinter using machine learning model[J]. Ironmaking and Steelmaking, 2019, 47(7):828. [23] 刘馨, 张卫军, 石泉, 等.基于数据挖掘与清洗的高炉操作参数优化[J]. 东北大学学报(自然科学版), 2020, 41(8):1153. (LIU X, ZHANG W J, SHI Q, et al. Optimization of blast furnace operation parameters based on data mining and cleaning[J]. Journal of Northeastern University(Natural Science), 2020, 41(8):1153.) [24] 陈少飞,刘小杰,李宏扬,等.高炉炼铁数据缺失处理研究初探[J]. 中国冶金, 2021, 31(2):17. (CHEN S F, LIU X J, LI H Y, et al. Study on missing data processing in blast furnace ironmaking[J]. China Metallurgy, 2021, 31 (2): 17.) [25] 陈少飞, 刘小杰, 李宏扬, 等. 高炉炼铁数据离群筛选办法[J]. 钢铁研究学报, 2021, 33(6):467. (CHEN S F, LIU X J, LI H Y, et al. Outlier screening method of blast furnace ironmaking data[J]. Journal of Steel Research, 2021, 33(6):467). [26] ZHAO J, CHEN S, LIU X, et al. Outlier screening for ironmaking data on blast furnaces[J]. International Journal of Minerals, Metallurgy and Materials, 2021, 28(6): 1001. [27] LI H, LIU X, LI X, et al. Prediction model for vanadium content in vanadium and titanium blast furnace smelting iron based on big data mining[J]. ISIJ International, 2022, 62(11):2301. [28] WU Z, HUANG N E. Ensemble empirical mode decomposition:A noise assisted data analysis method[J]. Advances in Adaptive Data Analysis, 2009, 1(1):1. [29] FURLANCETO D C, OLIVEIRA L S, MENOTTI D, et al. Bias effect on predicting market trends with EMD[J]. Expert Systems with Applications, 2017(82): 19. [30] TIAN G, WANG Q, ZHAO Y, et al. Smart contract classification with a Bi-LSTM based approach[J]. IEEE Access, 2020 (8): 43806. [31] AHMAD M W, REYNOLDS J, REZGUI Y. Predictive modelling for solar thermal energy systems: A comparison of support vector regression, random forest, extra trees and regression trees[J]. Journal of Cleaner Production, 2018(203): 810.
2023, Vol. 58 
