Long term planning for the water supply system of the Metropolitan Area of Amsterdam under new water saving regimes of the retail customers: Understanding the system and addressing uncertainties

The centralized drinking water supply system of the Metropolitan Area of Amsterdam, operated and owned mainly by Waternet, has high performance indicators and satisfies the full demand for water in the urban grid and surrounding areas on a full cost recovery basis. Social and technological developments, stemming mainly from (i) the increasing...

Advanced control of a water supply system: A case study

Conventional automatic production flow control and pump pressure control of water supply systems are robust and simple: production flow is controlled based on the level in the clear water reservoir and pump pressure is controlled on a static set-point. Recently, more advanced computer-based control methods were developed in which production flow...

An ensemble stacked model with bias correction for improved water demand forecasting

Water demand forecasting is an essential task for water utilities, with increasing importance due to future societal and environmental changes. This paper suggests a new methodology for water demand forecasting, based on model stacking and bias correction that predicts daily demands for groups of ~120 properties. This methodology is compared...

Water Demand Forecasting Accuracy and Influencing Factors at Different Spatial Scales Using a Gradient Boosting Machine

Understanding, comparing, and accurately predicting water demand at different spatial scales is an important goal that will allow effective targeting of the appropriate operational and conservation efforts under an uncertain future. This study uses data relating to water consumption available at the household level, as well as postcode...

Short-Term Forecasting of Household Water Demand in the UK Using an Interpretable Machine-Learning Approach

This study utilizes a rich UK data set of smart demand metering data, household characteristics, and weather data to develop a demand forecasting methodology that combines the high accuracy of machine learning models with the interpretability of statistical methods. For this reason, a random forest model is used to predict daily demands 1 day...