Journal of the Korean Society of Marine Environment & Safety (해양환경안전학회지)

The Korean Society of Marine Environment and safety (해양환경안전학회)
권호 표지
DOI QR코드

DOI QR Code

Prediction of Dissolved Oxygen in Jindong Bay Using Time Series Analysis

시계열 분석을 이용한 진동만의 용존산소량 예측

  • Received : 2020.04.10
  • Accepted : 2020.06.26
  • Published : 2020.06.30
Download PDF
⟨ Previous Next ⟩

Abstract

In this study, we used artificial intelligence algorithms for the prediction of dissolved oxygen in Jindong Bay. To determine missing values in the observational data, we used the Bidirectional Recurrent Imputation for Time Series (BRITS) deep learning algorithm, Auto-Regressive Integrated Moving Average (ARIMA), a widely used time series analysis method, and the Long Short-Term Memory (LSTM) deep learning method were used to predict the dissolved oxygen. We also compared accuracy of ARIMA and LSTM. The missing values were determined with high accuracy by BRITS in the surface layer; however, the accuracy was low in the lower layers. The accuracy of BRITS was unstable due to the experimental conditions in the middle layer. In the middle and bottom layers, the LSTM model showed higher accuracy than the ARIMA model, whereas the ARIMA model showed superior performance in the surface layer.

본 연구에서는 인공지능기법을 이용하여 진동만의 용존산소량 예측을 하였다. 관측자료에 존재하는 결측 구간을 보간하기 위해 양방향재귀신경망(BRITS, Bidirectional Recurrent Imputation for Time Series) 딥러닝 알고리즘을 이용하였고, 대표적 시계열 예측 선형모델인 ARIMA(Auto-Regressive Integrated Moving Average)과 비선형모델 중 가장 많이 이용되고 있는 LSTM(Long Short-Term Memory) 모델을 이용하여 진동만의 용존산소량을 예측하고 그 성능을 평가했다. 결측 구간 보정 실험은 표층에서 높은 정확도로 보정이 가능했으나, 저층에서는 그 정확도가 낮았으며, 중층에서는 실험조건에 따라 정확도가 불안정하게 나타났다. 실험조건에 따라 정확도가 불안정하게 나타났다. 결과로부터 LSTM 모델이 중층과 저층에서 ARIMA 모델보다 우세한 정확도를 보였으나, 표층에서는 ARIMA모델의 정확도가 약간 높은 것으로 나타났다.

Keywords

References

  1. Box, G. E. P. and G. M. Jenkins(1976), Time series analysis: forecasting and control, 1976, ISBN: 0-8162-1104-3.
  2. Cao, W., D. Wang, J. Li, H. Zhou, L. Li, and Y. Li(2018), BRITS: bidirectional recurrent imputation for time series. In Advances in Neural Information Processing Systems, pp. 6775-6785.
  3. Cho, C. H.(1991), Mariculture and eutrophication in Jinhae Bay, Korea. Marine Pollution Bulletin, Vol. 23, pp. 275-279. https://doi.org/10.1016/0025-326X(91)90687-N
  4. Choi, W. J., C. K. Park, and S. M. Lee(1994), Numerical Simulation of the Formation of Oxygen Deficient Water-masses in Jinhae Bay, Vol. 27, No. 4, pp. 413-433.
  5. Diaz, R. J. and R. Rosenberg(2008), Spreading dead zones and consequences for marine ecosystems, Science, Vol. 321, pp. 926-929. https://doi.org/10.1126/science.1156401
  6. Diaz, R. J. and R. Rosenberg(1995), Marine benthic hypoxia: a review of its ecological effects and behavioural responses of benthic macrofauna. Oceanography and Marine Biology Annual Review, Vol. 33, pp. 303-345.
  7. Frain, J.(1999), Lecture Notes on Univariate Time Series Analysis and Box Jenkins Forecasting, Economic Analysis Research and Publications, p. 26.
  8. Gers, F., N. Schraudolph, and J. Schmidhuber(2002), Learning precise timing with LSTM recurrent networks, Journal of Machine Learning Research, Vol. 3, pp. 115-143.
  9. Glorot, X. and Y. Bengio(2010), Understanding the difficulty of training deep feedforward neural networks. In Aistats, Vol. 9, pp. 249-256.
  10. Hochreiter, S. and J. Schmidhuber(1997), Long short-term memory. Neural computation, Vol. 9, No. 8, pp. 1735-1780. https://doi.org/10.1162/neco.1997.9.8.1735
  11. https://en.wikipedia.org/wiki/Long_short-term_memory(2019).
  12. Hussian, M. A., S Abbas, M. R. K. Ansari, and A. Zaffar (2013), Perturbations of modeling and forecast of Karachi coastal region seawater, Proceedings of the Pakistan Academy of Science, Vol. 50, No. 3, pp. 235-245.
  13. Jung, W. S., W. C. Lee, S. J. Hong, J. L. Kim, and D. M. Kim(2014), Hypoxia Extimation of Coastal Bay through Extimation of Stratification Degree, Journal of the Korean Society of Marine Environment & Safety, Vol. 20, No. 5, pp. 511-525. https://doi.org/10.7837/kosomes.2014.20.5.511
  14. Kang, H.(2019), A Study on prediction of hypoxia occurrence from inner bay, Kunsan National University.
  15. Karim, R.(2013), Season ARIMA for forecasting sea surface temperature of the north zone of the Bay of Bengal, Research&Reviews, Journal of Statistics, Vol. 2, No. 2, pp. 2278-2273.
  16. Kim, H. G.(1990), Characteristics of flagellate red tide and environmental conditions in Masan Bay. Bulletin of National Fisheries Research and Development Agency, Vol. 43, pp. 1-40.
  17. Kim, S. C.(2007), The corresponding problem of the prospect of the apiculture industry using the ARIMA model and National Agricultural Cooperative Association, Korea cooperative association research, Vol. 25, No. 1, pp. 183-210.
  18. Ku, S. K.(2013), Forecast of transportation demand for wide-area railways in 2014, Korea Railroad Corporation, p. 60.
  19. Levin, L. A.(2003), Oxygen minimum zone benthos: adaptation and community response to hypoxia. Marine Biology Annual Review, Vol. 41, pp. 1-45.
  20. Li, S., W. Li, C. Cook, C. Zhu, and Y. Gao(2018), Independently recurrent neural network (indrnn): Building a longer and deeper rnn. In Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 5457-5466.
  21. Liu, F. C., J. T. Liu, W. Su, and Y. Y. Guo(2009), Time series of coastal sea surface temperature: Simulation and prediction based on seasonal ARIMA model, Journal of Huaihai Institute of Technology (Natural Sciences Edition), Vol. 15, pp. 3709-318.
  22. Monteiro, P. M. S. and A. K. van der Plas(2006), Low oxygen water variability in the Benguela system. In: V. Shannon, G. Hempel, P. Malanotte-Rizzoli, C. Moloney, J. Woods (Eds.), The Benguela: Predicting A Large Marine Ecosystem, Elsevier, Vol. 14, pp. 71-90.
  23. National Institute of Fisheries Science(2009), Hypoxia in the coast of Korea, SP-2009-ME-021.
  24. Rabalais, N. N. and D. Gilbert(2009), Distribution and consequences of hypoxia, in: Watersheds, Bays and Bounded Seas, edited by: E. Urban, B. Sundby, P. Malanotte-Rizzoli, and J. M. Melillo, Island Press, Washington, DC, pp. 209-226.
  25. Seong, K. T., Y. H. Choi, J. H. Koo, and S. B. Jean(2014), Fluctuations and Time Series Forecasting of Sea Surface Temperature at Yeosu Coast in Korea, Journal of the Korean Society for Marine Environment and Energy, Vol. 17, No. 2, pp. 122-130. https://doi.org/10.7846/JKOSMEE.2014.17.2.122
  26. Slutzky, E.(1937), The summation of random causes as the source of cyclic processes, Econometrica: Journal of the Econometric Society, pp. 105-146.
  27. Srivastava, N., G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov(2014), Dropout: a simple way to prevent neural networks from overfitting, The journal of machine learning research, Vol. 15, No. 1, pp. 1929-1958.
  28. Yang, Y., J. Dong, X. Sun, E. Lima, Q. Mu, and X. Wang(2017), A CFCC-LSTM model for sea surface temperature prediction, IEEE Geoscience and Remote Sensing Letters, Vol. 15, No. 2, pp. 207-211. https://doi.org/10.1109/lgrs.2017.2780843
  29. Ye, M. J.(2015), A study on the development mode of hypoxia event in the Jinhae Bay during 2011-2013, and predictability of hypoxia using weather parameters, Pusan National University.
  30. Yule, G. U.(1926), Why do we sometimes get nonsense -correlations between Time-Series?--a study in sampling and the nature of time-series, Journal of the royal statistical society, Vol. 89, No. 1, pp. 1-63. https://doi.org/10.2307/2341482
  31. Zhang, Q., H. Wang, J. Dong, G. Zhong, and X. Sun(2017), Prediction of sea surface temperature using long short-term memory, IEEE Geoscience and Remote Sensing Letters 14.10, pp. 1745-1749. https://doi.org/10.1109/LGRS.2017.2733548
  32. Zhou, J., Y. Wang, F. Xiao, Y. Wang, and L. Sun(2018), Water Quality Prediction Method Based on IGRA and LSTM. Water, Vol. 10, No. 9, p. 1148. https://doi.org/10.3390/w10091148