[논문]역전파 신경망 모델을 이용한 기준 작물 증발산량 산정

김민영; 최용훈; 수잔 오샤네시; 폴 콜레이지; 김영진; 전종길; 이상봉

doi:10.5389/ksae.2019.61.6.111

[국내논문] 역전파 신경망 모델을 이용한 기준 작물 증발산량 산정
Estimation of Reference Crop Evapotranspiration Using Backpropagation Neural Network Model 원문보기

한국농공학회논문집 = Journal of the Korean Society of Agricultural Engineers, v.61 no.6, 2019년, pp.111 - 121

김민영 (Department of Agricultural Engineering, National Institute of Agricultural Sciences (NAS), Rural Development Administration (RDA)) , 최용훈 (Department of Agricultural Engineering, National Institute of Agricultural Sciences (NAS), Rural Development Administration (RDA)) , 수잔 오샤네시 (Conservation and Production Research Laboratory, United States Department of Agriculture, Agricultural Research Service (USDA-ARS)) , 폴 콜레이지 (Conservation and Production Research Laboratory, United States Department of Agriculture, Agricultural Research Service (USDA-ARS)) , 김영진 (Department of Agricultural Engineering, National Institute of Agricultural Sciences (NAS), Rural Development Administration (RDA)) , 전종길 (Department of Agricultural Engineering, National Institute of Agricultural Sciences (NAS), Rural Development Administration (RDA)) , 이상봉 (Department of Agricultural Engineering, National Institute of Agricultural Sciences (NAS), Rural Development Administration (RDA))

초록
AI-Helper

작물 증발산량은 수자원 계획 및 관리, 물수지 분석, 작물 관개 계획 및 생산량 추정 등에 널리 활용되고 있으며, 특히 FAO에서 공인한 Penman-Monteith식 (FAO 56-PM)은 잠재 증발산량 산정을 위한 표준방법으로 많이 사용되고 있다. Penman-Monteith식을 이용한 잠재증발산량 산정은 최소온도, 평균온도, 최대온도, 상대습도, 풍속과 일사량인 6가지 항목에 대한 시계열 자료가 필요한데, 결측 또는 미계측된 경우에는 사용이 어려운 단점을 가지고 있다. 따라서, 본 연구에서는 역전파 신경망(BPNN) 모델을 이용해서 6개 미만의 기상항목으로도 잠재증발산량이 추정가능한지를 확인하였다. 여섯 가지 기상항목을 각각 1~6개의 조합으로 입력자료를 구성하고, BPNN 모델을 이용해서 학습, 검증 및 테스트를 한 결과, 입력 자료가 많아질수록 좋은 결과가 산출되었으며, 일사량, 최대온도와 상대습도만으로도 결정계수($R^2$)가 0.94정도로 비교적 높은 예측결과를 얻을 수 있었다. 또한 산정 오차를 줄이고, 항목간의 상관관계를 높이기 위해서는 역전파 신경망 구조의 적절한 선택이 중요한 것으로 확인되었다. 역전파 신경망 모델을 사용하면 요구되는 기상 항목과 데이터의 양에 대한 제약 없이 예측이 가능할 수 있기 때문에 기준 증발산량 산정에 유용하게 활용될 수 있을 것이며 향후 작물 재배를 위한 적정 관개계획 수립에도 유용하게 사용될 것이라 사료된다.

Abstract ▼ AI-Helper

Evapotranspiration (ET) of vegetation is one of the major components of the hydrologic cycle, and its accurate estimation is important for hydrologic water balance, irrigation management, crop yield simulation, and water resources planning and management. For agricultural crops, ET is often calculated in terms of a short or tall crop reference, such as well-watered, clipped grass (reference crop evapotranspiration, $ET_o$). The Penman-Monteith equation recommended by FAO (FAO 56-PM) has been accepted by researchers and practitioners, as the sole $ET_o$ method. However, its accuracy is contingent on high quality measurements of four meteorological variables, and its use has been limited by incomplete and/or inaccurate input data. Therefore, this study evaluated the applicability of Backpropagation Neural Network (BPNN) model for estimating $ET_o$ from less meteorological data than required by the FAO 56-PM. A total of six meteorological inputs, minimum temperature, average temperature, maximum temperature, relative humidity, wind speed and solar radiation, were divided into a series of input groups (a combination of one, two, three, four, five and six variables) and each combination of different meteorological dataset was evaluated for its level of accuracy in estimating $ET_o$. The overall findings of this study indicated that $ET_o$ could be reasonably estimated using less than all six meteorological data using BPNN. In addition, it was shown that the proper choice of neural network architecture could not only minimize the computational error, but also maximize the relationship between dependent and independent variables. The findings of this study would be of use in instances where data availability and/or accuracy are limited.

주제어

표/그림 (10)

그림 Fig. 1 Description of study area
그림 Fig. 2 A schematic structure of BPNN (one hidden layer)
표 Table 1 Different combinations of input variables and its correspondent output variable
그림 Fig. 3 Average meteorological conditions in Bushland
표 Table 2 Statistical parameters of daily meteorological variables in study area
표 Table 3 Correlative regression between inputs and output for matching input combinations
그림 Fig. 4 Scattering diagrams of ET_o estimated by BPNN and FAO 56-PM during the testing periods depending upon the combination of input
표 Table 4 Best architecture of BPNN and its computational parameters
그림 Fig. 5 Statistical comparison of model performance by BPNN and MLR
표 Table 5 Multiple Linear Regression equations per best combination

AI 본문요약
AI-Helper

* AI 자동 식별 결과로 적합하지 않은 문장이 있을 수 있으니, 이용에 유의하시기 바랍니다.

제안 방법

A PC-based neural network application software, NeuralWorks Professional II/Plus (Neuralworks®, Carnegie, Pennsylvania, USA) used in this study, allows users to adjust key network and training parameters in BPNN.
Except for one individual variable used as an input, the rest of input combinations (two, three, four, five and six variables) were prepared to match solar radiation, which has the highest correlation regression with ET_o, with the others, which has the relatively high correlations (Table 1). A total of 20 combinations (from C₁ to C₂₀) were prepared to determine their best network configuration and regression equation for BPNN and MLR. In further explanation, a combination 1 (C₁) to a combination 6 (C₆) has only one variable defined as input and a combination 7 (C₇) to a combination 11 (C₁₁) has two variables as inputs as per high correlation with ET_o.
All required data, average, minimum and maximum air temperature (Tavg, Tmin and Tmax , °C), relative humidity (RH, %), wind speed (WS, m/s) and solar radiation (SR, watt/m2), were measured at 6 second intervals, reported as 15 minutes mean values from a weather station, and then stored in a Datalogger (Campbell Scientific, CR3000, Logan, Utah, USA).
) plays a major role in the agricultural management of water resources and its accurate prediction would signify better planning and management of the resources. Due to limitations of the FAO 56-PM equation, ANN modeling with BP algorithm was proposed as an alternative in this study to calculate ET_o.
, 2008). First of all, this study begun with three layer learning network which consists of an input layer, a hidden layer and an output layer (Fig. 2), but during the training procedure, more than one hidden layers were evaluated to calculate the weighted inputs with activation functions to produce the better outputs. In designing a robust and accurate ANN model, the modeler must address a number of important factors, including the type and structure of the neural network, the input prediction variables used, and data pre-processing.
In this study, the same training and validation datasets with BPNN modeling were used in generating the MLR equations and computing the coefficient of determination at a significance level of 5% between the FAO 56-PM and MLR equations (Table 5). The smallest discrepancies between ET_o calculated by BPNN and ET_o calculated by the FAO 56-PM equation were achieved from C₃, C₉, C₁₂, C₁₆, C₁₉ and C₂₀ with 1, 2, 3, 4, 5 and 6 combination of input variables, respectively.
The objectives of this study were to adopt a Backpropagation neural network (BPNN) model to calculate daily ETo from different input combinations (minimum, average & maximum air temperature, relative humidity, wind speed and solar radiation) and to assess the computational performance of ETo values between BPNN and Multiple Linear Regression (MLR).
, 2014). Therefore, this study began with evaluating the number of hidden layer and PEs in the hidden layer which exhibit non-linear behavior between inputs and output. All other computational parameters (momentum, learning coefficient ratio, learning rule and transfer function) were also determined.
To consider ANN modeling valid without manipulating data and/or evidence of contradictory results, all data used in this study was divided into three sets, one for the training, the other for the validation of the trained results, and another for the testing of the trained-validated results. A total of 1,633 datasets from 2010, 2012, 2014 to 2017 years were used, and 62% of data for network training, 8% for validation and 30% for testing were assigned, respectively.

대상 데이터

To consider ANN modeling valid without manipulating data and/or evidence of contradictory results, all data used in this study was divided into three sets, one for the training, the other for the validation of the trained results, and another for the testing of the trained-validated results. A total of 1,633 datasets from 2010, 2012, 2014 to 2017 years were used, and 62% of data for network training, 8% for validation and 30% for testing were assigned, respectively. All meteorological input and output variables were standardized in the range of 0 to 1 using the Min-Max normalization method (Choi et al.
A total of six meteorological variables were collected from a weather station which was located at the experimental station in USDA Agricultural Research Service Conservation and Production Research Laboratory, Bushland, Texas, USA (35° 11’ N, 102° 6’ W, 1,170 m above MSL) (Fig. 1).
Sensors used were temperature & relative humidity sensor (HC2S3, Rotronic, Hauppauge, New York, USA), pyranometer (LI200-RX, LI-COR, Lincoln, Nebraska, USA), and wind sentry set (Model 03002-L, R.M. Young, Traverse City, Michigan, USA).

데이터처리

In MLR analysis, the values of ET_o were used as the dependent variable, while each combinations was used as independent variables to derive the coefficients in the MLR model. The regression equations were generated with the aid of Microsoft Excel in this study.

이론/모형

A total of 1,633 datasets from 2010, 2012, 2014 to 2017 years were used, and 62% of data for network training, 8% for validation and 30% for testing were assigned, respectively. All meteorological input and output variables were standardized in the range of 0 to 1 using the Min-Max normalization method (Choi et al., 2018) and then partitioned using K-fold cross validation.

성능/효과

5 produced the best results, which indicates that no single value of learning rate was optimal for all combinations. The number of iterations was also considered one of the influencing factors on model performance and the preliminary test showed that more than 50,000 iterations did not improve the model accuracy (data not known). Instead of listing all twenty combinations, only the best performing combination per each group (same number of input variables used) is shown in Table 4, for example, C₃ was the best performing combination among C₁ through C₆.
In this study, the same training and validation datasets with BPNN modeling were used in generating the MLR equations and computing the coefficient of determination at a significance level of 5% between the FAO 56-PM and MLR equations (Table 5). The smallest discrepancies between ET_o calculated by BPNN and ET_o calculated by the FAO 56-PM equation were achieved from C₃, C₉, C₁₂, C₁₆, C₁₉ and C₂₀ with 1, 2, 3, 4, 5 and 6 combination of input variables, respectively.
925, respectively. This result showed that BPNN has a better performance than MLR, and both models indicate that no more than three input variables (T_max , SR, RH) would improve the accuracy.

후속연구

Overall, this study showed that the possibility of estimating ET_o with lesser number of meteorological variables as inputs and could provide valuable information on irrigation scheduling in an easily accessible way.

참고문헌 (35)

Abedi-Koupai, J., M. J. Amiri, and S. Eslamian, 2009. Comparison of artificial neural network and physically-based models for estimating of reference evapotranspiration in greenhouse. Australian Journal of Basic and Applied Sciences 3(3): 2528-2535.
Allen, R. G., L. S. Pereira, D. Raes, and M. Smith, 1998. Crop evapotranspiration-Guidelines for computing crop water requirements. FAO Irrigation and Drainage Paper, No 56, FAO, Rome.
Allen, R. G., W. O. Pruitt, J. L. Wright, T. A. Howell, F. Ventura, R. Snyder, D. Itenfisu, P. Steduto, J. Berengena, J. B. Yrisarry, M. Smith, L. S. Pereira, D. Raes, A. Perrier, A. Alves, I. Walter, and R. Elliot, 2006. A recommendation on standardized surface resistance for hourly calculation of reference $ET_o$ by the FAO 56 Penman-Monteith method. Agricultural Water Management 81(1-2): 1-22.

상세보기
Allen, R. G., L. S. Pereira, T. A. Howell, and M. E. Jensen, 2011. Evapotranspiration information reporting: II. Recommended documentation. Agricultural Water Management 98: 921-929

상세보기
Basheer I. A., and M. Hajmeer, 2000. Artificial neural networks: Fundamentals, computing, design, and application. J. Microbiol. Methods 43(1): 3-31.

상세보기
Benzaghta, M. A., T. A. Mohammed, and A. I. Ekhmaj, 2012. Prediction of evaporation from Algardabiya reservoir. Libyan Agriculture Research Center Journal International 3: 120-128. doi:10.5829/idosi.larcji.2012.3.3.1205.
Choi, Y., M. Kim, S. O'Shaughnessy, J. Jeon, Y. Kim, and W. Song, 2018. Comparison of artificial neural network and empirical models to determine daily reference evapotranspiration. Journal of the Korean Society of Agricultural Engineers 60(6): 43-54. doi:10.5389/KSAE.2018.60.6.043.
Coulibaly, P., F. Anctil, R. Aravena, B. Bobee, 2001. Artificial neural network modeling of water table depth fluctuations. Water Resources Research 37(4): 885-896. doi:10.1029/2000WR900368.

상세보기
Dai, X., H. Shi, Y. Li, Z. Ouyang, and Z. Huo, 2009. Artificial neural network models for estimating regional reference evapotranspiration based on climate factors. Hydrological Processes 23: 442-450. doi:10.1002/hyp.7153.

상세보기
Djman, K., K. Lombard, K. Komlan, and S. Allen, 2018. Variability of the ratio of alfalfa to grass reference evapotranspiration under semiarid climate. Irrigation & Drainage Systems Engineering 7(204): 1-6. doi:10.4172/2168-9768.1000204.
Drexler, J. Z., R. L. Snyder, D. Spano, and U. K. T. Paw, 2004. A review of models and micrometeorological methods used to estimate wetland evapotranspiration. Hydrological Processes 18(11): 2071-2101. doi:10.1002/hyp.1462.

상세보기
George, B. A., B. R. S. Reddy, N. Raghuwanshi, and W. W. Wallender, 2002. Decision support system for estimating reference evapotranspiration. Journal of Irrigation and Drainage Engineering 128(1): 1-10. doi:10.106/ASCE.0733-9437.

상세보기
Goel, A., 2009. ANN based modeling for prediction of evaporation in reservoirs (Research Note). International Journal of Engineering, Transactions A: Basics 22(4): 351-358.
Haykin, S., 1998. Neural networks: A comprehensive foundation. Prentice-Hall, Englewood Cliffs.
Itenfisu, D., R. L. Elliott, R. G. Allen, and I. A. Walter, 2003. Comparison of reference evapotranspiration calculations as part of the ASCE standardization effort. Journal of Irrigation and Drainage Engineering 129(6): 440-448. doi:10.1061/ASCE. 0733-9437.

상세보기
Jain S. K., A Sarkar, and V. Garg, 2008. Impact of declining trend of flow on Harike Wetland, India. Water Resources Management 22(4): 409-421. doi:10.1007/s11269-007-9169-9.

상세보기
Jennifer, M. J., and R. S. Sudheer, 2001. Evaluation of reference evapotranspiration methodologies and AFSIRS crop water use simulation model. Final report, Division of Water Supply Management, St. Johns River Water Manag, Dist., Palatka, Florida.
Jensen, M. E., R. D. Burman, and R. G. Allen, 1990. Crop and irrigation water requirements. Manual and Reports on Engineering Practice No. 70, ASCE, New York.
Kecman, V., 2001. Learning and soft computing. London, England: MIT press.
Khoob, A. R., 2008. Comparative study of Hargreaves's and artificial neural network's methodologies in estimating reference evapotranspiration in a semiarid environment. Irrigation Sci. 26(3): 253-259. doi:10.1007/s00271-007-0090-z.

상세보기
Kim, M., C. Y. Choi, and C. P. Gerba, 2008. Source tracking of microbial intrusion in water system using artificial neural networks. Water Research 42(4-5): 1308-1314. doi:10.1016/j.watres.2007.09.032.

상세보기
Laaboudi A., B. Mouhouche, and B. Draoui, 2012. Conceptual reference evapotranspiration models for different time steps. Journal of Petroleum & Environmental Biotechnology 3(4): 1-8. doi:10.4172/2157-7463.1000123.
Landeras G., A. Ortiz-Barredo, and J. J. Lopez, 2008. Comparison of artificial neural network models and empirical and semi-empirical equations for daily reference evapotranspiration estimation in the Basque Country (Northern Spain). Agricultural Water Management 95: 553-565. doi:10.1016/k/agwat/2007.12.011.

상세보기
Lee, E. J., M. S. Kang, J. A. Park, J. Y. Choi, and S. W. Park, 2010, Estimation of future reference crop evapotranspiration using artificial neural networks. Journal of the Korean Society of Agricultural Engineers 52(5): 1-9. doi:10.5389/KSAE.2010.52.5.001.
Lu, Y., D. Ma, X. Chen, and J. Zhang, 2018, A simple method for estimating field crop evapotranspiration from Pot Experiments. Water 10: 1-19. doi:10.3390/210121823.

상세보기
Maier, H. R., and G. C. Dandy, 2000. Neural networks for the prediction and forecasting of water resources variables: A review of modeling issues and applications. Environmental Model. Software 15: 101-124. doi:10.1016/S1364-8152(99)00007-9.

상세보기
Mia, M. M., S. K. Biswas, and M. C. Urmi, 2015. An algorithm for training multilayer perceptron (MLP) For image reconstruction using neural network without overfitting. IJSTR 10: 271-275.
Palayasoot, P., 1965. Estimation of pan evaporation and potential evapotranspiration of rice in the central plain of Thailand by using various formulas based on climatological data. M. S. Thesis, College of Engineering, Utah State University, Logan.
Parisi, S., L. Mariani, G. Cola, and T. Maggiore, 2009. Mini-lysimeters evapotranspiration measurements on suburban environment. Italian Journal of Agrometeorologoy 3: 13-16.
Smith, M., R. G., Allen, J. L. Monteith, A. Perrier, L. Pereira, and A. Segeren, 1992. Report of the expert consultation on procedures for revision of FAO guidelines for prediction of crop water requirements. UN-FAO, Rome, Italy, 54p.
Sudheer, K. P., and S. K. Jain, 2003. Radial basis function neural networks for modeling stage discharge relationship. J. Hydrol. Eng. 8(3): 161-164.

상세보기
Traore, A., H. H. Tamboura, A. Kabore, L.J. Royo, I. Fernandez, I. Alvarez, M. Sangare, D. Bouchel, J. P. Poivey, L. Sawadogo, and F. Goyache, 2008. Multivariate analyses on morphological traits in Burkina Faso goat. Arch Anim Breed 51: 588-600. doi:10.1016/j.smallrumres.2008.09.011.

상세보기
Vyas, K. N., and R. Subbaiah, 2016. Application of artificial neural network approach for estimating reference evapotranspiration. Current World Environment 11(2): 637-647. doi:10.12944/CWE.11.2.36.

상세보기
Zanetti, S. S., E. F. Sousa, V. P. S. Oliveira, F. T. Almeida, and S. Bernardo, 2007. Estimating evapotranspiration using artificial neural network and minimum climatological data. Journal of Irrigation and Drainage Engineering 133(2): 83-89. doi:10.1061/(ASCE)0733-9437(2002)128:4(224).

상세보기
Wu, W., G. C. Dandy, and H. R. Maier, 2014. Protocol for developing ANN models and its application to the assessment of the quality of the ANN model development process in drinking water quality modeling. Environ. Model. Softw. 54: 108-127. doi:10.1016/j.envsoft.2013.12.016.

상세보기

저자의 다른 논문 :

LOADING...

표제어: PCR

동의어: Packet Collision Rate

용어 설명 출처 목록 (6)

용어 설명: PCR은 세균 특이성이 있는 primer를 이용하여 적은 수의 세균이 있을지라도 쉽게 검출할 수 있는 유용한 방법이며, 이를 이용하여 구강 내 치면세균막이나 타액에서 직접 세균을 검출할 수 있게 되었다[8].

내보내기 구분	파일저장 인쇄 메일전송
구성항목	기본정보 상세정보 관리번호, 논문명, 저널/프로시딩명, 저자 , 발행년, 권, 호, 시작페이지, 끝페이지, 발행기관 관리번호, 논문명, 대등논문명, 저자 , 저널/프로시딩명, 발행기관, 발행년, 발행언어, 권, 호, 시작페이지, 끝페이지, ISBN, ISSN, 주제분야, 키워드, 초록(한글), 초록(영문), 저자(소속기관)
저장형식	Text(ASCII format) Excel format RefWorks Direct Export RIS format (for Reference Manager, ProCite, EndNote), Scholar's Aids, Mendeley
메일정보	받는사람 (필수) @ 보내는사람 (선택) @ 제목 내용 KISTI 검색결과 이메일 서비스
안내	총 건의 자료가 검색되었습니다. 다운받으실 자료의 인덱스를 입력하세요. (1-10,000) 검색결과의 순서대로 최대 10,000건 까지 다운로드가 가능합니다. 데이타가 많을 경우 속도가 느려질 수 있습니다.(최대 2~3분 소요) 다운로드 파일은 UTF-8 형태로 저장됩니다. 파일의 내용이 제대로 보이지 않을실 때는 웹브라우저 상단의 보기 -> 인코딩 -> 자동선택 여부를 확인하십시오. ~ Text(ASCII format) Excel format

연합인증