| Home | E-Submission | Sitemap | Contact Us |  
top_img

The Korean text of this paper can be translated into multiple languages on the website of http://jksee.or.kr through Google Translator.

J Korean Soc Environ Eng > Volume 43(3); 2021 > Article
J Korean Soc Environ Eng 2021;43(3): 206-217. doi: https://doi.org/10.4491/KSEE.2021.43.3.206
경안천 용존 산소 예측을 위한 입력 인자 선정 및 기계 학습 모델 비교
김민지1 , 변선정1 , 김경민2 , 안종화1,2
1강원대학교 환경공학과
2강원대학교 신산업개발 T-EMS 융합학과
Selection of Input Factors and Comparison of Machine Learning Models for Prediction of Dissolved Oxygen in Gyeongan Stream
Min Ji Kim1 , Seon Jeong Byeon1 , Kyung Min Kim2 , Johng-Hwa Ahn1,2
1Department of Environmental Engineering, Kangwon National University
2Department of Integrated Energy and Infra System, Kangwon National University
Corresponding author  Johng-Hwa Ahn ,Tel: 033-250-6357, Fax: 033-259-5550, Email: johnghwa@kangwon.ac.kr
Received: December 28, 2020;  Revised: March 9, 2021;  Accepted: March 9, 2021.  Published online: March 31, 2021.
ABSTRACT
Objectives
In this study, we select input factors for machine learning models to predict dissolved oxygen (DO) in Gyeongan Stream and compare results of performance evaluation indicators to find the optimal model.
Methods
The water quality data from the specific points of Gyeongan Stream were collected between January 15, 1998 and December 30, 2019. The pretreatment data were divided into train and test data with the ratio of 7:3. We used random forest (RF), artificial neural network (ANN), convolutional neural network (CNN), and gated recurrent unit (GRU) among machine learning. RF and ANN were tested by both random split and time series data, while CNN and GRU conducted the experiment using only time series data. Performance evaluation indicators such as square of the correlation coefficient (R2), root mean square error (RMSE), and mean absolute error (MAE) were used to compare the optimal results for the models.
Results and Discussion
Based on the RF variable importance results and references, water temperature, pH, electrical conductivity, PO4-P, NH4-N, total phosphorus, suspended solids, and NO3-N were used as input factors. Both RF and ANN performed better with time series data than random split. The model performance was good in order of RF > CNN > GRU > ANN.
Conclusions
The eight input factors (water temperature, pH, electrical conductivity, PO4-P, NH4-N, total phosphorus, suspended solids, and NO3-N) were selected for machine learning models to predict DO in Gyeongan Stream. The best model for DO prediction was the RF model with time series data. Therefore, we suggest that the RF with the eight input factors could be used to predict the DO in streams.
Key Words: Artificial Neural Network, Convolutional Neural Network, Gated Recurrent Unit, Gyeongan Stream, Random Forest
TOOLS
PDF Links  PDF Links
Full text via DOI  Full text via DOI
Download Citation  Download Citation
  E-Mail
Share:      
METRICS
0
Crossref
37
View
2
Download
Related article
Editorial Office
464 Cheongpa-ro, #726, Jung-gu, Seoul 04510, Republic of Korea
TEL : +82-2-383-9653   FAX : +82-2-383-9654   E-mail : ksee@kosenv.or.kr
About |  Browse Articles |  Current Issue |  For Authors and Reviewers
Copyright © Korean Society of Environmental Engineers. All rights reserved.                 Developed in M2PI