Journal of Korea Water Resources Association (한국수자원학회논문집)

Korea Water Resources Association (한국수자원학회)
권호 표지
DOI QR코드

DOI QR Code

Precipitation forecasting by fuzzy Theory : I - Applications of Neuro-fuzzy System and Markov Chain

퍼지론에 의한 강수예측 : I. 뉴로-퍼지 시스템과 마코프 연쇄의 적용

  • Published : 2002.10.01
Download PDF
⟨ Previous Next ⟩

Abstract

Water in the atmosphere is circulated by reciprocal action of various factors in the climate system. Otherwise, any climate phenomenon could not occur of itself. Thus, we have tried to understand the climate change by analysis of the factors. In this study, the fuzzy theory which is useful to express inaccurate and approximate nature in the real world is used for forecasting precipitation influenced by the factors. Forecasting models used in this study are neuro-fuzzy system and a Markov chain and those are applied to precipitation forecasting of illinois. Various atmosphere circulation factors(like soil moisture and temperature) influencing the climate change are considered to forecast precipitation. As a forecasting result, it can be found that the considerations of the factors are helpful to increase the forecastibility of the models and the neuro-fuzzy system gives us relatively more accurate forecasts.

대기에서의 물순환은 기후시스템이라는 커다란 공간 안에서 다양한 인자들의 상호작용을 통하여 이루어진다. 즉, 어떠한 기후 현상도 그 자체적으로 발생할 수는 없다. 따라서, 많은 연구자들은 영향인자들의 분석을 통하여 기후 변화를 이해하고자 노력하여 왔다. 본 연구에서는 다양한 인자에 의하여 영향을 받아 발생하는 강수량의 예측을 위하여 실제 세계의 근사적이고 부정확한 성질을 표현하는데 효과적인 퍼지 개념을 이용하였다. 예측을 위하여 적용한 모형은 크게 뉴로-퍼지 시스템과 마코프 연쇄이며, 일리노이주의 강수량 예측을 위하여 적용하였다. 예측은 강수량에 영향을 끼치는 다양한 대기순환 인자(예: 토양수분과 기온)를 고려하여 수행하였다. 예측 결과, 강수량 예측에 대기순환 인자들을 고려함으로써 모형의 예측능력을 향상시킬 수 있었고, 상대적으로 뉴로-퍼지 시스템의 예측이 보다 정확한 결과를 주었다.

Keywords

References

  1. 안중배, 류정희, 조익현, 박주영, 류상범 (1997) '한반도 기온 및 강수량과 주변 해역 해면온도와의 상관관계에 관한 연구'., 한국기상학회 논문집, Vol. 33, No. 2, pp. 128-141
  2. 오재호 (1999). 기후와 대기순환, 아르케
  3. 오재호 (1999). 변화하는 기후, 아르케
  4. 윤용남 (1998). 공업수문학, 청문각
  5. Bodri, L. and Cermak, V. )2000). 'Prediction of extreme precipitation using a neural network: application to summer flood occurrence in Moravia', Advances in Engineering Software, Vol. 31, p. 312-321 https://doi.org/10.1016/S0965-9978(99)00063-0
  6. Chang, F.J. and Chen, Y.C. (2001). 'A counterpropagation fuzzy-neural network modeling approach to real time streamflow prediction', Journal of Hydrology, Vol. 245, p. 153-164 https://doi.org/10.1016/S0022-1694(01)00350-X
  7. Eltahir, E.A.B. (1998a). 'A Soil moisture-rainfall feedback mechanics : Theory and observation', Water Resources Research, Vol. 34, No. 4, p. 756-776
  8. Eltahir, E.A.B. (1998b). 'A Soil moisture-rainfall feedback mechanics : Numerical experiment', Water Resources Research, Vol. 34, No. 4, p. 777-785 https://doi.org/10.1029/97WR03497
  9. Franks, S.W., Gineste, P., Beven, K.J., and Merot, P. (1998) 'On constraining the predictions of a distributed model : The incorporation of fuzzy estimates of saturated areas into the calibration process', Water Resources Research, Vol. 34, No. 4, p. 787-797 https://doi.org/10.1029/97WR03041
  10. Furundzic, D. (1998). 'Application example of neural networks for time series analysis', Signal Process, Vol. 64, p. 383-396 https://doi.org/10.1016/S0165-1684(97)00203-X
  11. Gautam, D.K. and Holz, K.P. (2001). 'Rainfall-runoff modelling using adaptive neuro-fuzzy systems', Jounal of Hydroinfomatics, March, p. 3-10
  12. Jang, J.S.R., Sun, C.T., and Mizutani, E. (1996). Neuro-fuzzy and soft computing : A Computational Approach to Learning and Machin Intelligence, Prentice Hall
  13. Kang, In-Sik, Hee-jung Baek (1993). 'Long range prediction of winter monthly-mean temperature in Korea', J. Kor. Meteo. Soo., Vol. 30, p. 247-260
  14. Lange, N.T. (1998). 'New Mathmatical Approach Hydrological modeling-An Application of Aricficial Neural Network', Phys. Chem. Earth, Vol. 24, No. 1-2, p. 31-35
  15. Lin, C.T. and C.S.G. Lee (1999). Neural Fuzzy Systems : A Neuro-Fuzzy Synergism to Intelligent Systems, Prentice Hall
  16. Luk, K.C., Ball, J.E., Sharma, A. (2001). 'An application of artificial neural networks for rainfall forecasting', Mathmatical and Computer Modeling, Vol. 33, p. 683-693 https://doi.org/10.1016/S0895-7177(00)00272-7
  17. Ouenes, A. (2000). 'Practical application of fuzzy logic and neural networks to fractured reservoir characterization', Computers Geosciences, Vol. 26, p. 953-962 https://doi.org/10.1016/S0098-3004(00)00031-5
  18. Ozelkan, E.C. (1996). 'Relationship between monthly atmospheric circulation patterns and precipitation : Fuzzy logic and regression approaches', Water Resources Research, Vol. 32, No. 7, p. 2097-2103 https://doi.org/10.1029/96WR00289
  19. Sajikumar, N. and Thandaveswara, B.S. (1999). 'A non-linear model using an artificial neural network', Journal of Hydrology, Vol. 216, p. 32-55 https://doi.org/10.1016/S0022-1694(98)00273-X
  20. Valdes, J. B., Entekhabi, D., and Bartolini, P. (1994). 'Long term predictability of river stages under ENSO influence, ASCE Hydraulic Engineering', Vol. 8, No. 1, p. 366-370
  21. Waylen, P. R. and Caviedes, C. N. (1990). 'Annual and seasonal fluctuations of precipitation and streamflow in the Aconcagua River Basin', Journal of Hydrology, Vol. 120, p. 79-102 https://doi.org/10.1016/0022-1694(90)90143-L
  22. Woolhiser, D. A. and Keefer, T. O. (1993). 'Southern Oscillation effects on daily precipitation in the southwestern United States', Water Resources Research, Vol. 29, No. 4, p. 1287-1295 https://doi.org/10.1029/92WR02536