Short-term water demand predictions coupling an artificial neural network model and a genetic algorithm

Abstract

The application of artificial neural network (ANN) models for short-term (15 min) urban water demand predictions is evaluated. Optimization of the ANN model's hyperparameters with a Genetic Algorithm (GA) and use of a growing window approach for training the model are also evaluated. The results are compared to those of commonly used time series models, namely the Autoregressive Integrated Moving Average (ARIMA) model and a pattern-based model. The evaluations are based on data sets from two Canadian cities, providing 15 minute water consumption records over respectively 5 years and 23 months, with a respective mean water demand of 14,560 and 887 m3/d. The GA optimized ANN model performed better than the other models, with Nash-Sutcliffe Efficiencies of 0.91 and 0.83, and Relative Root Mean Square Errors of 6 and 16% for City 1 and City 2, respectively. The results of this study indicate that the optimization of the hyperparameters of an ANN model can lead to better 15 min urban water demand predictions, which are useful for many real time control applications, such as dynamic pressure control.

Assessing the Benefits of Nature-Inspired Algorithms for the Parameterization of ANN in the Prediction of Water Demand

“…ANN is currently the most common machine learning technique applied in the hydrological area, particularly learning using a feedforward back-propagation (FFBP) structure The FFBP was used in precisely simulating municipal water needed across various spatiotemporal scales because of its ability to map the nonlinear (i.e., trend and seasonal) behavior of water data (Shirkoohi et al 2021;Zounemat-Kermani et al 2020).…”

Forecasting water demand for Istanbul by applying different machine learning algorithms

“…Thus, it tried to minimize the deviations in the water consumption estimation. Shirkoohi et al (2021) used artificial neural networks to estimate the short-term water consumption of a city, such as 15 minutes and optimized the parameters of this network with a genetic algorithm. Tian and Xue (2017) estimated the annual water demand by the Bayesian method with the feedback mechanism of artificial neural networks.…”

Improving short-term water demand forecasting using evolutionary algorithms

“…The aim of long-term analyses, considering the horizon of 20–30 years, is to support making decisions related to designing and developing water supply systems. Short-term simulations, usually hourly, daily, or weekly 3 , 4 , are used to optimize the work and energy costs of pump stations (Pump Scheduling Optimization, PSO) 5 and to solve current operational problems 6 . Tiwari and Adamowski 7 as well as Candelieri et al 8 also distinguish medium-term predicting with respect to weekly time ranges.…”