Author(s): Khoi Anh Nguyen; Rodney Anthony Stewart; Hong Zhang
Linked Author(s): Rodney Stewart, Hong Zhang
Keywords: Water end use low resolution data machine learning neural network smart metering
Abstract: This study investigates residential water end use data, captured at 15 to 60-minute intervals, to inform water management and conservation strategies. Data from Australian metropolitan cities were used to develop an intelligent model that combines machine learning techniques, including Random Forest, Extreme Gradient Boosting, and Neural Networks. By incorporating features such as time flow rate, the model achieved high accuracy in predicting various water use categories. This research highlights the critical role of feature and machine learning model selection, providing a robust foundation for future studies aimed at enhancing model performance and developing category-specific models.
Year: 2025