Hybrid Water Demand Forecasting Model Associating Artificial Neural Network with Fourier Series

Abstract

This paper addressed the problem of water-demand forecasting for real-time operation of water supply systems. The present study was conducted to identify the best fit model using hourly consumption data from the water supply system of Araraquara, São Paulo, Brazil. Artificial neural networks (ANNs) were used in view of their enhanced capability to match or even improve on the regression model forecasts. The ANNs used were the multilayer perceptron with the back-propagation algorithm (MLP-BP), the dynamic neural network (DAN2), and two hybrid ANNs. The hybrid models used the error produced by the Fourier series forecasting as input to the MLP-BP and DAN2, called ANN-H and DAN2-H, respectively. The tested inputs for the neural network were selected literature and correlation analysis. The results from the hybrid models were promising, DAN2 performing better than the tested MLP-BP models. DAN2-H, identified as the best model, produced a mean absolute error (MAE) of 3:3 L∕s and 2:8 L∕s for training and test set, respectively, for the prediction of the next hour, which represented about 12% of the average consumption. The best forecasting model for the next 24 hours was again DAN2-H, which outperformed other compared models, and produced a MAE of 3:1 L∕s and 3:0 L∕s for training and test set respectively, which represented about 12% of average consumption.

Improvement of mid- to long-term runoff forecasting based on physical causes: application in Nenjiang basin, China

“…For example, to improve flood forecasting accuracy, Chen (2010) used the decision group backpropagation network (DGBPN), while Yang and Chen (2009) integrated the linear transfer function (LTF) and self-organizing map (SOM) to efficiently determine the intervals of weights and biases. In addition, Odan and Reis (2012) used the multilayer perceptron with the back-propagation algorithm (MLP-BP) for water demand forecasting, while Anctil and Tape (2004) used an artificial neural network and wavelet hybrid model for rainfall-runoff forecasting. Although simulation methods are important, it is also necessary to distinguish the runoff formation mechanisms and main impact factors, then further abstract and generalize the physical processes of hydrological phenomena.…”

Gene Expression Programming Coupled with Unsupervised Learning: A Two-Stage Learning Process in Multi-Scale, Short-Term Water Demand Forecasts

“…One study combined ANN with the wavelet bootstrapping machine learning approach as a hybrid model to improve performance of the models by pre-processing the data [20]. In another study, performance of ANN was improved through a hybrid approach using the Fourier time series to model the water demand forecast [21]. Another recent study proposes the coupling of the kernel partial least squares-autoregressive moving average with wavelet transformation as a hybrid approach for modeling annual urban water demand [22].…”

Correlation Analysis of Water Demand and Predictive Variables for Short-Term Forecasting Models

“…In their turn, short-term water demand forecasting models can help to suitable define the operation and management of the water systems, with the aim of supplying water to costumers with maximum efficiency [1,3]. As a result, working with water predictive models is central to suitably establish the set of variables involved in the model.…”

“…Among the benefits associated with suitable water demand forecasting, leakage identification, optimal operation of pumps and valves, and the possibility of improving planning and design of network expansions must be highlighted. These engineering aspects represent a key step forward in the improvement of WDS operation efficiency, which ultimately will lead to the provision of quality water supply [1].…”