인공지능 기반 질소산화물 배출량 예측을 위한 연구모형 개발 = Development of Prediction Model for Nitrogen Oxides Emission Using Artificial Intelligence

국문 초록 (Abstract)

지속적으로 강화되는 환경오염 물질 배출 규제로 인해, 질소 산화물(NO_x)의 배출량 예측 및 관리는 산업 현장에서 많은 관심을 받고 있다. 본 연구에서는 인공지능 기반 질소산화물 배출량 예측모델 개발을 위한 연구모형을 제안하였다. 제안된 연구모형은 데이터의 전처리 과정부터 인공지능 모델의 학습 및 평가까지 모두 포함하고 있으며, 시계열 특성을 가지는 NO_x 배출량을 예측하기 위하여 순환 신경망 중 하나인 Long Short-Term Memory (LSTM) 모델을 활용하였다. 또한 의사결정나무 기법을 활용하여 LSTM의 time window를 모델 학습 이전에 선정하는 방법을 채택하였다. 본 연구에서 제안된 연구모형의 NO_x 배출량 예측 모델은 가열로에서 확보한 조업 데이터로 학습되었으며, 최적 모델은 hyper-parameter를 조절하여 개발되었다. 개발된 LSTM 모델은 학습 데이터 및 평가 데이터에 대하여 모두 93% 이상의 NO_x 배출량 예측 정확도를 나타내었다. 본 연구에 제안된 연구모형은 시계열 특성을 가지는 다양한 대기오염 물질의 배출량 예측모델 개발에 응용될 수 있을 것으로 기대된다.

다국어 초록 (Multilingual Abstract)

Prediction and control of nitrogen oxides (NO_x) emission is of great interest in industry due to stricter environmental regulations. Herein, we propose an artificial intelligence (AI)-based framework for prediction of NO_x emission. The framework includes pre-processing of data for training of neural networks and evaluation of the AI-based models. In this work, Long-Short-Term Memory (LSTM), one of the recurrent neural networks, was adopted to reflect the time series characteristics of NO_x emissions. A decision tree was used to determine a time window of LSTM prior to training of the network. The neural network was trained with operational data from a heating furnace. The optimal model was obtained by optimizing hyper-parameters. The LSTM model provided a reliable prediction of NO_x emission for both training and test data, showing an accuracy of 93% or more. The application of the proposed AI-based framework will provide new opportunities for predicting the emission of various air pollutants with time series characteristics.

참고문헌 (Reference)

1 권순호, "인공신경망 및 다중회귀 모형을 이용한 대설피해 추정 함수 개발" 한국방재학회 17 (17): 315-325, 2017

2 박철홍, "데이터 마이닝을 이용한 시멘트 소성공정 질소산화물(NOx)배출 관리 방법에 관한 연구" 한국품질경영학회 46 (46): 739-752, 2018

3 Tabachnick, J., "Using Multivariate Statistics" Pearson Education 2007

4 Olah, C, "Understanding LSTM Networks"

5 Song, X., "Time-series Well Performance Prediction Based on Long Short-Term Memory(LSTM)Neural Network Model" 186 : 106682-, 2020

6 Selvin, S., "Stock Price Prediction Using Lstm, Rnn and CNN-sliding Window Model" 1643-1647, 2017

7 Lee, H., "Status and Improvement Measures for Total Air Pollution Load Management" 36 : 2019

8 Connor, J. T., "Recurrent Neural Networks and Robust Time Series Prediction" 5 : 240-254, 1994

9 Li, Y., "Prediction for Tourism Flow based on LSTM Neural Network" 129 : 277-283, 2018

10 Dong, C., "Predicting the Heating Value of MSW with a Feed Forward Neural Network" 23 : 103-106, 2003

11 Benesty, J., "Noise Reduction in Speech Processing" Springer-Verlag Berlin Heidelberg 2009

12 Joo, S., "NOx Emissions Characteristics of the Partially Premixed Combustion of H2/CO/CH4 Syngas Using Artificial Neural Networks" 80 : 436-444, 2015

13 Xu, M., "Modelling of the Combustion Process and NOx Emission in a Utility Boiler" 79 : 1611-1619, 2000

14 Junninen, H., "Methods for Imputation of Missing Values in Air Quality Data Sets" 38 : 2895-2907, 2004

15 Hochreiter, S., "Long Short-Term Memory" 1780 : 1735-1780, 1997

16 Reinbacher, F., "Influence of Smooth Temperature Variation on Hotspot Ignition" 22 : 110-130, 2018

17 Khashei, M., "Improvement of Auto-Regressive Integrated Moving Average models using Fuzzy logic and Artificial Neural Networks (ANNs)" 72 (72): 956-967, 2009

18 Sola, J., "Importance of Input Data Normalization for the Application of Neural Networks to Complex Industrial Problems" 44 : 1464-1468, 1997

19 Zhang, G., "Forecasting with Artificial Neural Networks-the State of Art" 14 (14): 35-62, 1998

20 Meyler, A., "Forecasting Irish Inflation Using ARIMA Models"

21 Freeman, B. S., "Forecasting Air Quality Time Series Using Deep Learning" 68 : 866-886, 2018

22 Chattopadhyay, S., "Feed forward Artificial Neural Network model to predict the average summer-monsoon rainfall in India" 55 : 369-382, 2007

23 Taghavifar, H., "Exhaust Emissions Prognostication for DI Diesel Group-hole Injectors Using a Supervised Artificial Neural Network Approach" 125 : 81-89, 2014

24 Ferretti, G., "Estimation of NOx Emissions in Thermal Power Plants Using Neural Networks" 123 : 465-471, 2001

25 Karakoyun, E. Ş., "Comparison of ARIMA Time Series Model and LSTM Deep Learning Algorithm for Bitcoin Price Forecasting" 171-179, 2018

26 Khoshhal, A., "CFD Study on Influence of Fuel Temperature on NOx Emission in a HiTAC Furnace" 38 : 1421-1427, 2011

27 Kuhn, M., "Building Predictive Models in R Using the Caret Package" 28 : 1-26, 2008

28 Karim, Z. A. A., "Attaining Simultaneous Reduction in Nox and Smoke by Using Water-in-biodiesel Emulsion Fuels For Diesel Engine" 3 : 1-21, 2019

29 Taghavifar, H., "Appraisal of Artificial Neural Networks to the Emission Analysis and Prediction of CO2, soot, and NOx of n-heptane Fueled Engine" 112 : 1729-1739, 2016

30 Safavian, S. R., "A Survey of Decision Tree Classifier Methodology" 21 : 660-674, 1991

31 Studzinski, W., "A Computational and Experimental Study of Combustion Chamber Deposit Effects on NOx Emissions" 93281 : 5-, 1993