Analysis on intelligent control method of energy saving and emission reduction in whole sintering process
YU Han1,2, ZHAO Man-kun3, PAN Zhi-cheng2, YU Hong-bing1, CAO Yu-xin1, LI Ying-jie1
1. School of Environmental Science and Engineering, Nankai University, Tianjin 300350, China; 2. National Enterprise Technology Center, Haitian Water Group Co., Ltd., Chengdu 610095, Sichuan, China; 3. College of Intelligence and Computing, Tianjin University, Tianjin 300350, China
Abstract:This study proposed the model and method for emission reduction of air pollutants and energy saving in the whole sintering process. According to the node characteristics of material flow and energy flow, the sintering process was divided into sintering source system, process system, and end treatment system. Based on the deep belief network, the "dual-balance" constraint method was used to train the ingredients-mineralization prediction model for the intelligent control of the source ingredients. By optimizing the fuel ratio, fuel consumption and pollutant emissions can be reduced from the beginning. Based on the deep neural network, the wind box negative pressure and temperature prediction coupling model was constructed, which reduced the electrical energy consumption during the sintering process. Reduction agent for terminal denitrification was adjusted according to the information from the pollutants at the beginning and during the sintering process. Finally, an intelligent auxiliary diagnosis and decision-making system for energy saving and emission reduction in the whole sintering process was established. In the application of this system, solid fuel consumption, electricity consumption, NOx emission, SO2emission and particulate matter were reduced by 18.9%, 21.9%, 43.6%, 14.0%, and 20.1%, respectively. The effect of this system on energy saving and emission reduction was determined significant.
于晗, 赵满坤, 潘志成, 于宏兵, 曹玉鑫, 李英杰. 烧结全过程节能减排智能控制方法分析[J]. 中国冶金, 2020, 30(12): 112-118.
YU Han, ZHAO Man-kun, PAN Zhi-cheng, YU Hong-bing, CAO Yu-xin, LI Ying-jie. Analysis on intelligent control method of energy saving and emission reduction in whole sintering process[J]. China Metallurgy, 2020, 30(12): 112-118.
Hinton G E, Osindero S, Teh Y W. A fast learning algorithm for deep belief nets[J]. Neural Computation, 2006, 18(7):1527.
[7]
Fukushima K, Miyake S. Neocognitron: A Self-organizing Neural Network Model for A Mechanism of Visual Pattern Recognition[M]. Berlin, Heidelberg:Springer, 1982: 267.
[8]
Huang G, Liu Z, Weinberger K Q, et al. Densely connected convolutional networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Hawaii:IEEE,2017: 3.
[9]
Ashish Vaswani, Noam Shazeer, Niki Parmer, et al. Attention is all you need[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach:IEEE, 2017: 5998.
[10]
HU J, SHEN L, SUN G. Squeeze-and-excitation networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Salt Lake City:IEEE, 2018: 7132.
[11]
Hochreiter S, Schmidhuber J. Long short-term memory[J]. Neural Computation, 1997, 9(8):1735.
[12]
Cho K, Van Merrienboer B, Gulcehre C, et al. Learning phrase representations using RNN encoder-decoder for statistical machine translation[J]. Computer Ence, 2014,1(14):1179.