Forecasting Daily Water Consumption: a Case Study in Town, Poland

被引：20

作者：

Piasecki, Adam ^{[1
]}

Jurasz, Jakub ^{[2
]}

Kazmierczak, Bartosz ^{[3
]}

机构：

[1] AGH Univ Sci & Technol, Fac Management, Dept Econ Finance & Environm Management, 30 Adama Mickiewicza Ave, PL-30059 Krakow, Poland

[2] AGH Univ Sci & Technol, Fac Management, Dept Engn Management, 30 Adama Mickiewicza Ave, PL-30059 Krakow, Poland

[3] Wroclaw Univ Sci & Technol, Fac Environm Engn, 27 Wybrzeze Wyspianskiego St, PL-50370 Wroclaw, Poland

来源：

PERIODICA POLYTECHNICA-CIVIL ENGINEERING | 2018年 / 62卷 / 03期

关键词：

artificial neural networks; multiple-linear regression; water consumption; SURFACE; NETWORKS; MODELS; FUZZY;

D O I：

10.3311/PPci.11930

中图分类号：

TU [建筑科学];

学科分类号：

0813 ;

摘要：

This paper presents Artificial Neural Network (ANN) and Multiple Linear Regression (MLR) methods for predicting future daily water consumption values based on three antecedent records of water consumption and humidity forecast for a given day, which are considered as independent variables. Mean Absolute Percentage Error (MOPE) is obtained for different configurations of the input sets and of the ANN model structure. Additionally, sets of explanatory variables are enhanced with dummy variables indicating typical days: working day, Saturday, Sunday/public holidays. The results indicated the superiority of the ANN approach over MLR, although the observed difference in performance was very limited.

引用

页码：818 / 824

页数：7

共 31 条

[1] Comparison of multiple linear and nonlinear regression, autoregressive integrated moving average, artificial neural network, and wavelet artificial neural network methods for urban water demand forecasting in Montreal, Canada [J].