Water scarcity is a global problem. Deficit irrigation (DI) reduces evapotranspiration, improving water efficiency in agriculture. Reference evapotranspiration ${(ET}_0)$ is an important factor in determining DI. ${ET}_0$ forecasting predicts field water consumption and enables proactive irrigation decisions, offering guidance for water resource management. However, implementation of ${ET}_0$ forecasting faces challenges due to complex calculations and extensive meteorological data requirements. This project aims to develop a machine learning system for ${ET}_0$ forecasting. The project involves studying ${ET}_0$ methods and identifying required meteorological parameters. Historical meteorological data and weather forecasts were obtained from meteorological websites and analyzed for accuracy after preprocessing. A machine learning-based model was created to forecast reference crop evapotranspiration. The model’s input parameters were selected through path analysis before it was optimized using Bayesian optimization to reduce overfitting and improve accuracy. Three forecasting models were developed: one based on historical meteorological data, one based on weather forecasts, and one that corrects the weather forecasts. All three models achieved good accuracy, with root mean square errors ranging from 0.52 to 0.81 mm/day. Among them, the model based on weather forecast had the highest accuracy; the RMSE six days before the forecast period was between 0.52 and 0.75 mm/day, and the RMSE on the seventh day of the forecast period was 1.12 mm/day. In summary, this project has established a mathematical model of ${ET}_0$ prediction based on machine learning, which can achieve more accurate predictions for within a few days.

The maturity grading of walnuts during harvesting relies on experience. In this paper, walnut images in a natural environment were collected to construct a dataset, and deep learning algorithms were utilized to combine walnut internal physical and chemical indicators to carry out research on walnut maturity detection methods and further research on walnut oil content prediction by combining walnut images with walnut oil content indicators. The main contents of this paper include collecting walnut images in a natural environment, constructing datasets, and using deep learning algorithms combined with internal physical and chemical indexes of walnuts to study walnut maturity detection and oil content prediction methods. First, two walnut image acquisition schemes were designed, and a total of 9504 images were collected from 23 August to 21 September 2021. The dataset was expanded to 18,504 images through data preprocessing and image enhancement. A self-supervised Gaussian attention network (GATCluster) walnut ripeness detection method based on image clustering is proposed to develop ripeness criteria through unsupervised clustering, and the accuracy of the criteria is verified by analysis of variance (ANOVA). The maturity detection accuracy of the test set of 1500 images is 88.33%. Secondly, a walnut oil content prediction method based on improved ResNet34 is proposed. The feature extraction capability is improved by introducing the Squeeze-and-Excitation Networks (SENet) channel attention mechanism and the convolutional self-attention module. The prediction results on 50 images show that the root mean square error, average absolute percentage error, and regression coefficient are 2.96, 0.103, and 0.8822, respectively. The experiments show that the method performs well in predicting the oil content of walnuts at different maturity levels.