Accurate reference decisions are important to optimize irrigation scheduling in greenhouse environments and to ensure optimal plant growth and resource management.
The purpose of this study is to identify the most accurate method for predicting ET0 in naturally ventilated greenhouse conditions. Four machine learning (ML) models were analyzed: standalone adaptive neurofuzzy inference system (ANFI), decision tree (DTree), support vector machine (SVM), and ANFI with improved reptile search algorithm (IRSA). These models were evaluated for prediction accuracy using performance metrics including R-squared (𝑅2), route mean squared error (RMSE), mean absolute error (MAE), bias, and scatter index (SI) for training and validation datasets. The results showed variation in accuracy and effectiveness of different estimation equations, while the ANFIS IRSA model showed excellent performance across the metrics evaluated.
The success of this model highlights the possibility of advancement in precision agriculture by accurately predicting ET0 in greenhouse conditions, paving the way for further research in this field.
A comparative evaluation of machine learning models to predict daily evapotranspiration in naturally ventilated greenhouses. Sahoo Bibhuti Bhusan; Najafzadeh Mohammad; dt santosh; Jithendra Thandra; Panigurahibanamari; Mishra Shucchi; Gupta Sushindra Kumar; Bhushan Mani; 10/10/2025; T2-Journal of Irrigation and Drainage Engineering. VL -151; American Society of Civil Engineering; https://doi.org/10.1061/jidedh.ireng-10441
Source: ASCE Library
