Journal of Industrial Convergence (산업융합연구)

DAEHAN Society of Industrial Management (대한산업경영학회)
권호 표지
DOI QR코드

DOI QR Code

A Study on the Prediction of Nitrogen Oxide Emissions in Rotary Kiln Process using Machine Learning

머신러닝 기법을 이용한 로터리 킬른 공정의 질소산화물 배출예측에 관한 연구

  • Je-Hyeung Yoo (Candidate, Seoul Business School, aSSIST University) ;
  • Cheong-Yeul Park (Seoul Business School, aSSIST University) ;
  • Jae Kwon Bae (Dept. of Management Information Systems, Keimyung University)
  • 유제형 (서울과학종합대학원대학교 경영학과) ;
  • 박정열 (서울과학종합대학원대학교 경영학과) ;
  • 배재권 (계명대학교 디지털경영학부 경영정보학전공)
  • Received : 2023.05.30
  • Accepted : 2023.07.20
  • Published : 2023.07.28
Download PDF
⟨ Previous Next ⟩

Abstract

As the secondary battery market expands, the process of producing laterite ore using the rotary kiln and electric furnace method is expanding worldwide. As ESG management expands, the management of air pollutants such as nitrogen oxides in exhaust gases is strengthened. The rotary kiln, one of the main facilities of the pyrometallurgy process, is a facility for drying and preliminary reduction of ore, and it generate nitrogen oxides, thus prediction of nitrogen oxide is important. In this study, LSTM for regression prediction and LightGBM for classification prediction were used to predict and then model optimization was performed using AutoML. When applying LSTM, the predicted value after 5 minutes was 0.86, MAE 5.13ppm, and after 40 minutes, the predicted value was 0.38 and MAE 10.84ppm. As a result of applying LightGBM for classification prediction, the test accuracy rose from 0.75 after 5 minutes to 0.61 after 40 minutes, to a level that can be used for actual operation, and as a result of model optimization through AutoML, the accuracy of the prediction after 5 minutes improved from 0.75 to 0.80 and from 0.61 to 0.70. Through this study, nitrogen oxide prediction values can be applied to actual operations to contribute to compliance with air pollutant emission regulations and ESG management.

이차전지 시장의 확대에 따라 니켈 산화광을 로터리 킬른 및 전기로 공법을 이용하여 생산하는 공정이 전 세계적으로 확대되고 있는 상황이며 지속가능한 ESG 경영 확대에 따라 배출가스 내 질소산화물 등 대기오염물질 관리가 강화되고 있다. 건식니켈제련 공정의 주요 설비 중 하나인 로터리 킬른은 광석의 건조와 예비환원을 위한 설비이며 운전 중 질소산화물이 생성되므로 질소산화물 농도 예측 운전이 필요하다. 본 연구에서는 회귀 예측을 위한 LSTM 모델과 분류 예측을 위한 LightGBM 모델을 적용한 AutoML을 사용하여 모델을 최적화 하였다. LSTM을 적용 시 5분 후 예측 값은 상관계수 0.86, MAE 5.13ppm, 40분 후 예측 값은 상관계수 0.38, MAE 10.84ppm의 결과를 얻었다. 분류 예측을 위한 LightGBM 적용 결과 Test 정확도는 5분 후 0.75에서 40분 후 0.61로 상승하여 실제 조업에 활용할 수 있는 수준까지 상승되었고 AutoML을 통한 모델 최적화 결과 5분 후 예측 값의 정확도는 0.75에서 0.80까지, 40분 후의 예측 정확도는 0.61에서 0.70까지 향상되었다. 본 연구를 통해 로터리 킬른 질소산화물 예측 값을 실제 조업에 적용하여 대기오염물질 배출규제 준수 및 ESG 경영에 기여할 수 있다.

Keywords

References

  1. BISULANDU, B. J. R. M., & HUCHET, F. (2022). Rotary kiln process: An overview of physical mechanisms, models and applications. Applied Thermal Engineering, 119637. 
  2. Kambara, S., Takarada, T., Toyoshima, M., & Kato, K. (1995). Relation between functional forms of coal nitrogen and NOx emissions from pulverized coal combustion. Fuel, 74(9), 1247-1253.  https://doi.org/10.1016/0016-2361(95)00090-R
  3. Sohn, H. S. (2021). Current Status of Nickel Smelting Technology. Resources Recycling, 30(2), 3-13.  https://doi.org/10.7844/kirr.2021.30.2.3
  4. Cai, J., Wu, H., Ren, Q., Lin, L., Zhou, T., & Lyu, Q. (2020). Innovative NOx reduction from cement kiln and pilot-scale experimental verification, Fuel Processing Technology, 199, 106306. 
  5. Edland, R., Normann, F., Fredriksson, C., & Andersson, K. (2017). Implications of fuel choice and burner settings for combustion efficiency and NOx formation in PF-fired iron ore rotary kilns, Energy & Fuels, 31(3), 3253-3261.  https://doi.org/10.1021/acs.energyfuels.6b03205
  6. Orooji, Y., Javadi, M., Karimi-Maleh, H., Aghaie, A. Z., Shayan, K., Sanati, A. L., & Darabi, R. (2021). Numerical and experimental investigation of natural gas injection effects on NOx reburning at the rotary cement kiln exhaust, Process Safety and Environmental Protection, 151, 290-298. 
  7. Muzio, L. J., & Quartucy, G. C.(1997). Implementing NOx control: reseach to application. Progress in Energy and Combustion Science, 23(3), 233-266.  https://doi.org/10.1016/S0360-1285(97)00002-6
  8. KAIKAKE, A. (2007). Recent ferronickel smelting operation at Hyuga Smelting Co., Ltd. Journal of MMIJ, 123(12), 686-688. 
  9. Korea Energy Economics Institutes (2022). 2021 Annual Energy Statistics, Ulsan : Korea Energy Economics Institutes. 
  10. Adams, D., Oh, D. H., Kim, D. W., Lee, C. H., & Oh, M. (2020). Prediction of SOx-NOx emission from a coal-fired CFB power plant with machine learning: Plant data learned by deep neural network and least square support vector machine, Journal of Cleaner Production, 270, 122310. 
  11. Tan, P., Xia, J., Zhang, C., Fang, Q., & Chen, G. (2016). Modeling and reduction of NOX emissions for a 700 MW coal-fired boiler with the advanced machine learning method, Energy, 94, 672-679.  https://doi.org/10.1016/j.energy.2015.11.020
  12. Yang, G., Wang, Y., & Li, X. (2020). Prediction of the NOx emissions from thermal power plant using long-short term memory neural network, Energy, 192, 116597. 
  13. Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory, Neural Computation, 9(8), 1735-1780.  https://doi.org/10.1162/neco.1997.9.8.1735
  14. Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., Ye, Q., & Liu, T. Y. (2017). Lightgbm: A highly efficient gradient boosting decision tree, Advances in Neural Information Processing Systems, 30. 
  15. He, X., Zhao, K., & Chu, X. (2021). AutoML: A survey of the state-of-the-art, Knowledge-Based Systems, 212, 106622.