Forecasting urban water demand using different hybrid-based metaheuristic algorithms' inspire for extracting...

Area of Science:

• Environmental Science
• Water Resource Management
• Computational Intelligence

Background:

• Accurate urban water demand forecasting is crucial for sustainable water resource management.
• Traditional methods often struggle to capture complex relationships between meteorological variables and water consumption.

Purpose of the Study:

• To develop and evaluate a novel methodology for quantifying urban water needs.
• To enhance the accuracy of water demand prediction by integrating advanced data processing and artificial intelligence.

Main Methods:

• Utilized a combination of data preprocessing techniques to improve input data quality.
• Employed an artificial neural network (ANN) optimized with a genetic algorithm enabled particle swarm optimisation (PSOGA) algorithm.
• Compared the PSOGA-ANN model against various metaheuristic algorithms, including modified PSO and standard PSO.

Main Results:

• All models adequately simulated monthly urban water demand based on meteorological variables.
• Statistical fitness measures confirmed that the PSOGA-ANN model significantly outperformed the benchmarking algorithms.
• Effective data processing enhances initial data quality and optimizes predictions for urban water demand.

Conclusions:

• The proposed PSOGA-ANN methodology offers a robust and accurate approach for urban water demand forecasting.
• Optimized data processing and advanced algorithms are key to improving the reliability of water resource management strategies.
• This research provides a valuable tool for water utilities to better manage supply and demand.