Modelling evapotranspiration using discrete wavelet transform and neural networks

Partal, Turgay

doi:10.1002/hyp.7448

Cited by 62 publications

(26 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this study, the significant wavelets (approximation and detail series) were summed together once the insignificant coefficients were excluded, similar to what was done by Partal [58] and Kisi and Cimen [32]. In this study, the summed sub-series provided better results than using the individual wavelet coefficients as inputs.…”

Section: Wn Modelsmentioning

confidence: 89%

Standard Precipitation Index Drought Forecasting Using Neural Networks, Wavelet Neural Networks, and Support Vector Regression

Belayneh

Adamowski

2012

Applied Computational Intelligence and Soft Computing

128

View full text Add to dashboard Cite

Drought forecasts can be an effective tool for mitigating some of the more adverse consequences of drought. Data-driven models are suitable forecasting tools due to their rapid development times, as well as minimal information requirements compared to the information required for physically based models. This study compares the effectiveness of three data-driven models for forecasting drought conditions in the Awash River Basin of Ethiopia. The Standard Precipitation Index (SPI) is forecast and compared using artificial neural networks (ANNs), support vector regression (SVR), and wavelet neural networks (WN). SPI 3 and SPI 12 were the SPI values that were forecasted. These SPI values were forecast over lead times of 1 and 6 months. The performance of all the models was compared using RMSE, MAE, and R 2 . The forecast results indicate that the coupled wavelet neural network (WN) models were the best models for forecasting SPI values over multiple lead times in the Awash River Basin in Ethiopia.

show abstract

Section: Wn Modelsmentioning

confidence: 89%

Standard Precipitation Index Drought Forecasting Using Neural Networks, Wavelet Neural Networks, and Support Vector Regression

Belayneh

Adamowski

2012

Applied Computational Intelligence and Soft Computing

128

View full text Add to dashboard Cite

show abstract

“…The WD-ARIMA model exhibited very good performance for forecasting the A ET , surplus, and deficit, whereas the classical ARIMA model exhibited poor performance or was unable to forecast the WBCs. Moreover, studies (Chou, 2011;Kisi, 2008;Partal, 2009;Santos and da Silva, 2014;Rahman and Hasan, 2014;Nury et al, 2016;Adamowski and Chan, 2011;Khalek and Ali, 2016) have indicated that the performance of wavelet-aided models is better than that of the classical ARIMA and ANN models for forecasting nonstationary hydrometeorological variables. Because traditional methods such as Wiener filtering, Kalman filtering, and Fourier transform are unsuitable for nonstationary hydrological time series data (Adamowski and Chan, 2011;Sang, 2013), wavelet denoising can be used to improve the performance of the classical ARIMA model for forecasting hydrological variables.…”

Section: Discussionmentioning

confidence: 99%

“…. D4 + D3 + A) of the time series data were calculated and the obtained results were compared (Partal and Küçük, 2006;Partal, 2009). …”

Section: Wavelet Transform (Wt) and Periodicitymentioning

confidence: 99%

“…The most common statistical methods for hydrological forecasting are the ARIMA model and multiple linear regression (Young, 1999; Adamowski,icity using wavelet-transformed details and using the approximation components of the hydrometeorological time series data can provide insight regarding the effects of the time period on the data trend (Nalley et al, 2013;Araghi et al, 2015;Pathak et al, 2016). As a result, detecting the periodicity through the wavelet transformation of hydrometeorological time series data has gained popularity in recent years (Partal and Küçük, 2006;Partal, 2009;Nalley et al, 2013;Araghi et al, 2015;Pathak et al, 2016). Studies have been conducted on the spatiotemporal characteristics of hydrometeorological variables, such as rainfall (Shahid and Khairulmaini, 2009;Ahasan et al, 2010;Kamruzzaman et al, 2016a;Rahman and Lateh, 2016;Syed and Al Amin, 2016), temperature (Shahid, 2010;Nasher and Uddin, 2013;Syed and Al Amin, 2016;Kamruzzaman et al, 2016a), and P ET Acharjee, 2017), in Bangladesh.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Modeling the changes in water balance components of the highly irrigated western part of Bangladesh

Rahman

Ahmed

Adnan

et al. 2018

Hydrol. Earth Syst. Sci.

View full text Add to dashboard Cite

Abstract. The objectives of the present study were to explore the changes in the water balance components (WBCs) by co-utilizing the discrete wavelet transform (DWT) and different forms of the Mann-Kendall (MK) test and develop a wavelet denoise autoregressive integrated moving average (WD-ARIMA) model for forecasting the WBCs. The results revealed that most of the potential evapotranspiration (P ET ) trends (approximately 73 %) had a decreasing tendency from 1981-1982 to 2012-2013 in the western part of Bangladesh. However, most of the trends (approximately 82 %) were not statistically significant at a 5 % significance level. The actual evapotranspiration (A ET ), annual deficit, and annual surplus also exhibited a similar tendency. The rainfall and temperature exhibited increasing trends. However, the WBCs exhibited an inverse trend, which suggested that the P ET changes associated with temperature changes could not explain the change in the WBCs. Moreover, the 8-year (D3) and 16-year (D4) periodic components were generally responsible for the trends found in the original WBC data for the study area. The actual data was affected by noise, which resulted in the ARIMA model exhibiting an unsatisfactory performance. Therefore, wavelet denoising of the WBC time series was conducted to improve the performance of the ARIMA model. The quality of the denoising time series data was ensured using relevant statistical analysis. The performance of the WD-ARIMA model was assessed using the Nash-Sutcliffe efficiency (NSE) coefficient and coefficient of determination (R 2 ). The WD-ARIMA model exhibited very good performance, which clearly demonstrated the advantages of denoising the time series data for forecasting the WBCs. The validation results of the model revealed that the forecasted values were very close to actual values, with an acceptable mean percentage error. The residuals also followed a normal distribution. The performance and validation results indicated that models can be used for the short-term forecasting of WBCs. Further studies on different combinations of wavelet analysis are required to develop a superior model for the hydrological forecasting in the context of climate change. The findings of this study can be used to improve water resource management in the highly irrigated western part of Bangladesh.

show abstract

“…Nourani et al [15] combined WA and ANN to predict the runoff in the Ligvanchai valley of Tabriz, Iran. Partal [38] conducted a reference evapotranspiration estimation using the wavelet transform and the feedforward neural network methods to evaluate climate data (temperature, solar radiation, wind speed, relative humidity) at two stations in the United States.…”

Section: Introductionmentioning

confidence: 99%

Hybrid Decomposition-Reconfiguration Models for Long-Term Solar Radiation Prediction Only Using Historical Radiation Records

Wang

Qiu

2018

Energies

View full text Add to dashboard Cite

Solar radiation prediction is significant for solar energy utilization. This paper presents hybrid methods following the decomposition-prediction-reconfiguration paradigm using only historical radiation records with different combination of decomposition methods, Ensemble Empirical Mode Decomposition (EEMD) and Wavelet Analysis (WA), and the reconfiguration methods, regression model (RE) and Artificial Neural Network (ANN). The application in west China indicates that these hybrid decomposition-reconfiguration models perform well for monthly prediction, while the comparisons of the daily prediction show that the hybrid EEMD-RE model has a higher degree of fitting and a better prediction effect in long-term prediction of solar radiation intensity, which verifies (1) decomposition of original solar radiation data results in components with regular characteristics; (2) the relationship between the original solar radiation sequence and the derived intrinsic mode functions (IMFs) is linear; and (3) EEMD has strong adaptivity for non-linear and non-stationary series. The proposed hybrid decomposition-reconfiguration models have great application prospect for monthly long-term prediction of solar radiation intensity, especially in the areas where complex climate data is difficult to obtain, and the EEMD-RE model is recommended for the daily long-term prediction.

show abstract

Modelling evapotranspiration using discrete wavelet transform and neural networks

Cited by 62 publications

References 56 publications

Standard Precipitation Index Drought Forecasting Using Neural Networks, Wavelet Neural Networks, and Support Vector Regression

Standard Precipitation Index Drought Forecasting Using Neural Networks, Wavelet Neural Networks, and Support Vector Regression

Modeling the changes in water balance components of the highly irrigated western part of Bangladesh

Hybrid Decomposition-Reconfiguration Models for Long-Term Solar Radiation Prediction Only Using Historical Radiation Records

Contact Info

Product

Resources

About