Urban Residential Water Demand Prediction Based on Artificial Neural Networks and Time Series Models

2015 ◽  
Vol 29 (10) ◽  
pp. 3651-3662 ◽  
Author(s):  
Muhammad A. Al-Zahrani ◽  
Amin Abo-Monasar
Energy ◽  
2018 ◽  
Vol 151 ◽  
pp. 347-357 ◽  
Author(s):  
Henrique do Nascimento Camelo ◽  
Paulo Sérgio Lucio ◽  
João Bosco Verçosa Leal Junior ◽  
Paulo Cesar Marques de Carvalho ◽  
Daniel von Glehn dos Santos

Atmosphere ◽  
2018 ◽  
Vol 9 (2) ◽  
pp. 77 ◽  
Author(s):  
Henrique do Nascimento Camelo ◽  
Paulo Sérgio Lucio ◽  
João Verçosa Leal Junior ◽  
Daniel von Glehn dos Santos ◽  
Paulo Cesar Marques de Carvalho

Author(s):  
Fahd Alqasemi ◽  
Salah AL-Hagree ◽  
Muneer Alsurori ◽  
Mohammed Hadwan ◽  
Zakaria Aljaberi ◽  
...  

Author(s):  
João Vitor Alves Da Cruz ◽  
Bruno Alberto Soares Oliveira

<p>Técnicas de predição de demanda são utilizadas em inúmeros ramos da indústria, com o objetivo de agregar valor ao negócio das empresas, especialmente por meio da busca pelo dimensionamento ótimo dos recursos de produção. A predição de demanda em refeitórios, com o intuito de balancear a quantidade de alimento produzido, buscando um melhor aproveitamento dos ingredientes, é um desafio, pois fatores como a quantidade de usuários, o tempo de atendimento e o tipo de alimento utilizado podem ser bastante variáveis neste tipo de problema. O estudo das filas, neste contexto, é de primordial importância, dado que, conhecendo suas características, podem-se estimar, por meio de previsão, informações que podem melhorar a qualidade de atendimento. O presente trabalho teve por finalidade utilizar modelos baseados em Redes Neurais Artificiais (RNA) para realizar regressões em uma série temporal personalizada, gerada por meio de metodologia própria, mediantes os dados coletados in loco no restaurante do IFMG - Campus Bambuí. Teve-se por principal objetivo desenvolver um modelo computacional que fosse capaz de descrever o comportamento para os intervalos de tempo no atendimento dos usuários. Por meio deste recurso, pôde-se gerar informações importantes para a tomada de decisão, como os horários de maior e menor pico de atendimento.</p><p><strong>Palavras-chave</strong>: Redes neurais artificiais, regressão, séries temporais.</p><p>==================================================================================</p><p>Demand prediction techniques are used in numerous industry sectors with the aim of adding value to the business of the companies, especially through the search for optimal sizing of production resources. The prediction of demand in restaurants with the intention of balancing the quantity food produced looking for better use of ingredients is a challenge, since factors like the quantity of users, the time of service and the kind of food can be quite variable in this type of problem. The study of queue, in this context, is of paramount importance, given that, knowing its characteristics, it is possible to estimate, by means of prediction, information that can improve service quality. Present work had the purpose of using models based on Artificial Neural Networks (ANN) to perform regressions in a personalized time series, generated through its own methodology with data collected in the restaurant of the IFMG - Campus Bambuí. The main objective was to develop a computational model that would be able to describe the behavior for the time intervals in the restaurant customer service. Through this resource, it was possible to generate important information for decision making, such as the peak times of higher and lower demands.</p><p><strong>Keywords</strong>: Artificial neural networks, regression, time series.</p>


Author(s):  
G. Jenitha

<p> In this study, we propose a new hybrid approach for time series prediction based on the efficient capabilities of fuzzy cognitive maps (FCMs) with structure optimization algorithms and artificial neural networks (ANNs). The proposed structure optimization genetic algorithm (SOGA) for automatic construction of FCM is used for modeling complexity based on historical time series, and artificial neural networks (ANNs) which are used at the final process for making time series prediction. The suggested SOGA-FCM method is used for selecting the most important nodes (attributes) and interconnections among them which in the next stage are used as the input data to ANN used for time series prediction after training. The FCM with proficient learning calculations and ANN have been as of now demonstrated as adequate strategies for setting aside a few minutes arrangement anticipating. The execution of the proposed approach is exhibited through the examination of genuine information of every day water request and the comparing expectation. The multivariate examination of recorded information is held for nine factors, season, month, day or week, occasion, mean and high temperature, rain normal, touristic action and water request. The entire approach was actualized in a clever programming device at first sent for FCM forecast. Through the exploratory investigation, the value of the new mixture approach in water request forecast is illustrated, by computing the mean outright blunder (as one of the outstanding expectation measures). The outcomes are promising for future work to this bearing.</p>


Sign in / Sign up

Export Citation Format

Share Document