scholarly journals Application of LSTM Networks for Water Demand Prediction in Optimal Pump Control

Water ◽  
2021 ◽  
Vol 13 (5) ◽  
pp. 644
Author(s):  
Christian Kühnert ◽  
Naga Mamatha Gonuguntla ◽  
Helene Krieg ◽  
Dimitri Nowak ◽  
Jorge A. Thomas
Keyword(s):  
Drinking Water ◽  
Water Demand ◽  
Water Distribution ◽  
Short Term Memory ◽  
Time Series Prediction ◽  
Real Life ◽  
Training Data ◽  
Demand Prediction ◽  
Additional Information ◽  
Water Demand Prediction

Every morning, water suppliers need to define their pump schedules for the next 24 h for drinking water production. Plans must be designed in such a way that drinking water is always available and the amount of unused drinking water pumped into the network is reduced. Therefore, operators must accurately estimate the next day’s water consumption profile. In real-life applications with standard consumption profiles, some expert system or vector autoregressive models are used. Still, in recent years, significant improvements for time series prediction have been achieved through special deep learning algorithms called long short-term memory (LSTM) networks. This paper investigates the applicability of LSTM models for water demand prediction and optimal pump control and compares LSTMs against other methods currently used by water suppliers. It is shown that LSTMs outperform other methods since they can easily integrate additional information like the day of the week or national holidays. Furthermore, the online- and transfer-learning capabilities of the LSTMs are investigated. It is shown that LSTMs only need a couple of days of training data to achieve reasonable results. As the focus of the paper is on the real-world application of LSTMs, data from two different water distribution plants are used for benchmarking. Finally, it is shown that the LSTMs significantly outperform the system currently in operation.

Adaptable urban water demand prediction system

2015 ◽  
Vol 15 (5) ◽  
pp. 958-964 ◽  
Author(s):  
G. Banjac ◽  
M. Vašak ◽  
M. Baotić
Keyword(s):  
Mutual Information ◽  
Water Demand ◽  
Distribution Systems ◽  
Water Distribution ◽  
Meteorological Data ◽  
Model Parameters ◽  
Prediction System ◽  
Prediction Ability ◽  
Demand Prediction ◽  
Water Demand Prediction

In this work, identification of 24-hours-ahead water demand prediction model based on historical water demand data is considered. As part of the identification procedure, the input variable selection algorithm based on partial mutual information is implemented. It is shown that meteorological data on a daily basis are not relevant for the water demand prediction in the sense of partial mutual information for the analysed water distribution systems of the cities of Tavira, Algarve, Portugal and Evanton East, Scotland, UK. Water demand prediction system is modelled using artificial neural networks, which offer a great potential for the identification of complex dynamic systems. The adaptive tuning procedure of model parameters is also developed in order to enable the model to adapt to changes in the system. A significant improvement of the prediction ability of such a model in relation to the model with fixed parameters is shown when a certain trend is present in the water demand profile.

Application of short-term water demand prediction model to Seoul

2002 ◽  
Vol 46 (6-7) ◽  
pp. 255-261 ◽  
Author(s):  
C.N. Joo ◽  
J.Y. Koo ◽  
M.J. Yu
Keyword(s):  
Water Demand ◽  
Model Building ◽  
Time Series Prediction ◽  
Social Factor ◽  
Maximum Temperature ◽  
Factors Affecting ◽  
Demand Prediction ◽  
Daily Water ◽  
Water Demand Prediction ◽  
Artificial Neural Network Ann

To predict daily water demand for Seoul, Korea, the artificial neural network (ANN) was used. For the cross correlation, the factors affecting water demand such as maximum temperature, humidity, and wind speed as natural factors, holidays as a social factor and daily demand 1 day before were used. From the results of learning using various hidden layers and units in order to establish the structure of optimal ANN, the case of 3 hidden layers and numbers of unit with the same number of input factors showed the best result and, therefore, it was applied to seasonal water demand prediction. The performance of ANN was compared with a multiple regression method. We discuss the representation ability of the model building process and the applicability of the ANN approach for the daily water demand prediction. ANN provided reasonable results for time series prediction.

scholarly journals Multi-Model Coupling Water Demand Prediction Optimization Method for Megacities Based on Time Series Decomposition

2021 ◽  
Author(s):  
Xin Liu ◽  
Xuefeng Sang ◽  
Jiaxuan Chang ◽  
Yang Zheng
Keyword(s):  
Water Supply ◽  
Water Demand ◽  
Short Term Memory ◽  
Prediction Interval ◽  
Virus Disease ◽  
Generalization Ability ◽  
Demand Prediction ◽  
Daily Data ◽  
Daily Water ◽  
Water Demand Prediction

AbstractThe water supply in megacities can be affected by the living habits and population mobility, so the fluctuation degree of daily water supply data is acute, which presents a great challenge to the water demand prediction. This is because that non-stationarity of daily data can have a large influence on the generalization ability of models. In this study, the Hodrick-Prescott (HP) and wavelet transform (WT) methods were used to carry out decomposition of daily data to solve the non-stationarity problem. The bidirectional long short term memory (BLSTM), seasonal autoregressive integrated moving average (SARIMA) and Gaussian radial basis function neural network (GRBFNN) were developed to carry out prediction of different subseries. The ensemble learning was introduced to improve the generalization ability of models, and prediction interval was generated based on student's t-test to cope with the variation of water supply laws. This study method was applied to the daily water demand prediction in Shenzhen and cross-validation was performed. The results show that WT is superior to HP decomposition method, but maximum decomposition level of WT should not be set too high, otherwise the trend characteristics of subseries will be weakened. Although the corona virus disease 2019 (COVID-19) outbreak caused a variation in water supply laws, this variation is still within the prediction interval. The WT and coupling models accurately predict water demand and provide the optimal mean square error (0.17%), Nash-Sutcliffe efficiency (97.21%), mean relative error (0.1), mean absolute error (3.32%), and correlation coefficient (0.99).

scholarly journals WATER DEMAND PREDICTION USING ARTIFICIAL NEURAL NETWORK FOR SUPERVISORY CONTROL

2016 ◽  
Vol 36 (1) ◽  
pp. 148-154
Author(s):  
BI Gwaivangmin ◽  
JD Jiya
Keyword(s):  
Neural Network ◽  
Supervisory Control ◽  
Water Demand ◽  
Water Distribution ◽  
Distribution Network ◽  
Water Distribution Network ◽  
Optimal Operation ◽  
Demand Prediction ◽  
The Neural Network ◽  
Water Demand Prediction

With increase in population growth, industrial development and economic activities over the years, water demand could not be met in a water distribution network.  Thus, water demand forecasting becomes necessary at the demand nodes.  This paper presents Hourly water demand prediction at the demand nodes of a water distribution network using NeuNet Pro 2.3 neural network software and the monitoring and control of water distribution using supervisory control.  The case study is the Laminga Water Treatment Plant and its water distribution network, Jos.  The proposed model will be developed based on historic records of water demand in the 15 selected demand nodes for 60 days, 24 hours run. The data set is categorized into two set, one for training the neural network and the other for testing, with a learning rate of 50 and hidden nodes of 10 of the neural network model.  The prediction results revealed a satisfactory performance of the neural network prediction of the water demand. The predictions are then used for supervisory control to remotely control and monitor the hydraulic parameters of the water demand nodes. The practical application in the plant will cut down the cost of water production and even to a large extend provide optimal operation of the distribution networks solving the perennial problem of water scarcity in Jos. http://dx.doi.org/10.4314/njt.v36i1.19

scholarly journals Inequality in Access to Drinking Water and Subsidies between Low and High Income Households in Mexico City

Water ◽  
2018 ◽  
Vol 10 (8) ◽  
pp. 1023 ◽  
Author(s):  
Jorge Morales-Novelo ◽  
Lilia Rodríguez-Tapia ◽  
Daniel Revollo-Fernández
Keyword(s):  
Drinking Water ◽  
Mexico City ◽  
Water Demand ◽  
Water Distribution ◽  
Water System ◽  
Well Being ◽  
High Income ◽  
Baseline Scenario ◽  
Continuous Variables ◽  
Public Network

Economic and population growth in Mexico City (CDMX) is the main cause of an increase in water demand against a naturally limited endowment, which increases the gap between water demand and supply. In a water scarcity environment, households are facing pressure to maintain their involvement in the city’s only operating body, the Water System of Mexico City (SACMEX) total supply. The objective of this work is to measure the inequality in the distribution of drinking water and water subsidies between households connected to the public network of CDMX in order to generate objective indicators of the phenomenon. Having such information provides a baseline scenario of the problem and allows for the delineation of a policy covering the minimum levels of well-being in the supply of drinking water that is appropriate for the most important city in the country. The method consists of measuring inequality through continuous variables estimating the Lorenz curve, the Gini coefficient, the targeting coefficient and elasticity in water consumption and in water subsidies among households in CDMX. Data comes from a household survey carried out in 2011, Consumption Habits, Service and Quality of Water by Household in Mexico City (EHCSCA). Results show that drinking water and subsidies present a regressive distribution, benefit high-income households and, to a lesser degree, the poorest households in the city and highlight the urgency and importance for SACMEX to redefine its policy on water distribution, fees and subsidies. The present study’s scope can contribute to the monitoring of the distribution of drinking water and of subsidies among household groups. The study justifies that the indicators employed in this work can be used and are recommended as a valuable tool in water management, especially in a dynamic environment.

Sign in / Sign up

Export Citation Format

Share Document