Daily Reference Evapotranspiration Estimation using Linear Regression and ANN Models
| Autor: | K. C. Sekhar Reddy, P. Mallikarjuna, S. A. Jyothy |
|---|---|
| Rok vydání: | 2012 |
| Předmět: |
Estimation
Artificial neural network Mechanical Engineering Regression analysis Building and Construction Agricultural and Biological Sciences (miscellaneous) Wind speed Improved performance Evapotranspiration Architecture Linear regression Sunshine duration Statistics Civil and Structural Engineering Mathematics |
| Zdroj: | Journal of The Institution of Engineers (India): Series A. 93:215-221 |
| ISSN: | 2250-2157 2250-2149 |
| DOI: | 10.1007/s40030-013-0030-2 |
| Popis: | The present study investigates the applicability of linear regression and ANN models for estimating daily reference evapotranspiration (ET0) at Tirupati, Nellore, Rajahmundry, Anakapalli and Rajendranagar regions of Andhra Pradesh. The climatic parameters influencing daily ET0 were identified through multiple and partial correlation analysis. The daily temperature, wind velocity, relative humidity and sunshine hours mostly influenced the study area in the daily ET0 estimation. Linear regression models in terms of the climatic parameters influencing the region and, optimal neural network architectures considering these influencing climatic parameters as input parameters were developed. The models’ performance in the estimation of ET0 was evaluated with that estimated by FAO-56 Penman–Montieth method. The regression models showed a satisfactory performance in the daily ET0 estimation for the regions selected for the present study. The optimal ANN (4,4,1) models, however, consistently showed an improved performance over regression models. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink