Medium and long-term regional water demand prediction using Harris hawks optimisation-backpropagation neural network...

Area of Science:

• Environmental Science
• Water Resource Management
• Artificial Intelligence in Environmental Modeling

Background:

• Efficient regional water resource allocation necessitates precise medium and long-term water demand prediction.
• Traditional forecasting methods often struggle with the complexity and dimensionality of water demand factors.

Purpose of the Study:

• To introduce and evaluate a novel coupled Harris Hawks Optimisation-Backpropagation Neural Network (HHO-BPNN) model for medium and long-term water demand forecasting.
• To assess the predictive performance of HHO-BPNN against established forecasting techniques.
• To identify key drivers influencing regional water demand.

Main Methods:

• Developed a novel HHO-BPNN coupled model for water demand prediction.
• Applied Principal Component Analysis (PCA) for dimensionality reduction of water demand influencing factors.
• Utilized the sliding window method for 1, 3, and 5-year water demand forecasts in five cities.
• Evaluated model performance using Mean Square Error (MSE), Mean Absolute Percentage Error (MAPE), Mean Absolute Error (MAE), and Coefficient of Determination (R²).

Main Results:

• The HHO-BPNN model demonstrated superior performance compared to BPNN, Support Vector Machines, and Grey Prediction models.
• Achieved high prediction accuracy across agricultural, industrial, domestic, and ecological water demand categories, with an overall accuracy of 97%.
• Forecasts demonstrated strong alignment with local urban development plans, indicating practical utility.

Conclusions:

• The HHO-BPNN model offers an effective tool for precise regional water demand forecasting.
• The study provides insights into key drivers of water demand, supporting informed water management and decision-making.
• The model's high accuracy and alignment with development plans highlight its potential for practical application in urban planning and water resource management.