Exportar Publicação

Ana Borges, Clara Cordeiro, Fernando Sebastião, M. Filomena Teodoro, Andrade, M. A. P., Rogério Duarte...Telma Guerra (2018). “Evaluation Water meters performance. 140th European Study Group with Industry .

A. Borges et al.,  "“Evaluation Water meters performance", in 140th European Study Group with Industry , Barreiro, 2018

@misc{borges2018_1775633557933,
	author = "Ana Borges and Clara Cordeiro and Fernando Sebastião and M. Filomena Teodoro and Andrade, M. A. P. and Rogério Duarte and Sandra Aleixo and Sérgio Fernandes and Telma Guerra",
	title = "“Evaluation Water meters performance",
	year = "2018",
	howpublished = "Impresso",
	url = "http://www.esgi140.ips.pt/"
}

TY  - CPAPER
TI  - “Evaluation Water meters performance
T2  - 140th European Study Group with Industry 
AU  - Ana Borges
AU  - Clara Cordeiro
AU  - Fernando Sebastião
AU  - M. Filomena Teodoro
AU  - Andrade, M. A. P.
AU  - Rogério Duarte
AU  - Sandra Aleixo
AU  - Sérgio Fernandes
AU  - Telma Guerra
PY  - 2018
CY  - Barreiro
UR  - http://www.esgi140.ips.pt/
AB  - This report presents a solution the case study-Evaluating Water Meters Performance-as proposed by Infraquinta for the 140th European Study Group with Industry (ESGI140), which took place in Barreiro School of Technology, Polytechnic Institute of Setu´bal (ESTBarreiro/IPS). Related to the problematic of detecting the water meter breaking points, Infraquinta proposed the challenge of extracting the trend component, from billing information (water consumption trough time), which is linked to water meter performance. More precisely, an algorithm for evaluating water meters performance by using historical data (monthly and hourly time series of water consumption). The main purpose of the algorithm should be to find out the meter performance breakpoint and hence where should water meters be replaced. The main challenge of this task is the decomposition of monthly time series into seasonality, trend and irregular components. The trend component is the one related to meter performance. For that we propose a methodology based on the decomposition of the time series–the SeasonalTrend Decomposition of time series based on Loess (STL), the values of water volume measured trough time. The decomposition of the time series is a way of identifying the influence of the seasonal and trend component and through their analysis one can explain the behaviour of the data. A second methodology inspired by the linear regression model is used to detect structural changes (breakpoints) within the trend component. Results show that the combination of these two methodologies was able to detect the water meter breaking points and, if incorporated in a system that automatically analyses the data and updates the information by systematically detecting the different breaking points could be the solution sought by Infraquinta.
ER  -