Forecasting weekly reference evapotranspiration using Auto Encoder Decoder Bidirectional LSTM model hybridized with a Boruta-CatBoost input optimizer

•A novel deep learning-based machine learning model is proposed for weekly evapotranspiration forecasting.•A new feature selection algorithm (Boruta-CatBoost) is applied to determine effective lags for time series forecasting.•Three different climatic conditions within Iranian region are investigate...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Computers and electronics in agriculture Jg. 198; S. 107121
Hauptverfasser: Karbasi, Masoud, Jamei, Mehdi, Ali, Mumtaz, Malik, Anurag, Yaseen, Zaher Mundher
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier B.V 01.07.2022
Schlagworte:
agriculture
algorithms
Auto Decoder Encoder
Bi-LSTM
Boruta-Catboost
electronics
evapotranspiration
FAO-56 Penman-Monteith
Forecasting
neural networks
Reference evapotranspiration
Reference evapotranspiration
Bi-LSTM
Auto Decoder Encoder
FAO-56 Penman-Monteith
Boruta-Catboost
Forecasting
ISSN:0168-1699
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:•A novel deep learning-based machine learning model is proposed for weekly evapotranspiration forecasting.•A new feature selection algorithm (Boruta-CatBoost) is applied to determine effective lags for time series forecasting.•Three different climatic conditions within Iranian region are investigated.•The employed model demonstrated an accurate forecasting performance for weekly evapotranspiration forecasting. Reference evapotranspiration (ETo) is one of the most important and influential components in optimizing agricultural water consumption and water resources management. In the present study, an efficient deep learning model, Auto Encode Decoder Bidirectional Long Short-Term Memory (AED-BiLSTM), was applied for the first time in forecasting 1–3 weeks ahead of weekly ETo. Three meteorological stations, including Kermanshah (semi-arid), Nowshahr (very humid), and Yazd (arid), which have different climates, were investigated and evaluated. For this purpose, a statistical period of 20 years (2000 to 2019), of which 15 years (2000–2014) for training and the final five years (2015–2015) for model testing were considered. The generalized regression neural network (GRNN) and extreme gradient boosting (XGBoost) machine learning models were used to compare the performance of the newly developed model. A novel CatBoost-Boruta algorithm was used to determine the critical lags in the forecasting process. The results showed that the newly developed model (AED-BiLSTM) in all three meteorological stations had higher ability and accuracy in forecasting weekly ETo than the GRNN and XGBoost models. The outcomes of testing phase in Kermanshah station ascertained that the AED-BiLSTM in terms of (ETo(t + 1): R (correlation coefficient) = 0.9835, RMSE (root mean square error, mm/week) = 3.4597; ETo(t + 2): R = 0.9805, RMSE = 3.7215; ETo(t + 3): R = 0.9795, RMSE = 3.8857) outperformed the other models. In Nowshahr station AED-BiLSTM model had higher accuracy in terms of (ETo(t + 1): R = 0.9057, RMSE = 4.6786; ETo(t + 2): R = 0.9036, RMSE = 4.7697; ETo(t + 3): R = 0.9024, RMSE = 4.8437). Also, in Yazd station AED-BiLSTM perform better than other models in terms of (ETo(t + 1): R = 0.9827, RMSE = 3.1319; ETo(t + 2): R = 0.9800, RMSE = 3.3998; ETo(t + 3): R = 0.9774, RMSE = 3.5679). The XGBoost and GRNN models performed better after the AED-BiLSTM model, respectively. The effect of climate type on the accuracy of the models showed that the models have higher efficiency and accuracy in arid and semi-arid climates.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0168-1699
DOI:10.1016/j.compag.2022.107121