Development of a Deep Learning-Based Prediction Model for Water Consumption at the Household Level

Abstract

The importance of efficient water resource supply has been acknowledged, and it is essential to predict short-term water consumption in the future. Recently, it has become possible to obtain data on water consumption at the household level through smart water meters. The pattern of these data is nonlinear due to various factors related to human activities, such as holidays and weather. However, it is difficult to accurately predict household water consumption with a nonlinear pattern with the autoregressive integrated moving average (ARIMA) model, a traditional time series prediction model. Thus, this study used a deep learning-based long short-term memory (LSTM) approach to develop a water consumption prediction model for each customer. The proposed model considers several variables to learn nonlinear water consumption patterns. We developed an ARIMA model and an LSTM model in the training dataset for customers with four different water-use types (detached houses, apartment, restaurant, and elementary school). The performances of the two models were evaluated using a test dataset that was not used for model learning. The LSTM model outperformed the ARIMA model in all households (correlation coefficient: mean 89% and root mean square error: mean 5.60 m3). Therefore, it is expected that the proposed model can predict customer-specific water consumption at the household level depending on the type of use.

Enhancing Water Management in Shale Gas Extraction through Rectangular Pulse Hydraulic Fracturing

“…This method has played a crucial role in boosting the production of domestic energy resources and transforming the landscape of energy production. The process involves the following steps: well preparation, fluid, and proppant selection, the fracturing process, and water flow-back [11].…”

“…This pulsating approach generates a dynamic stress field within the shale formation, inducing a complex fracture network extending outward from the wellbore [10]. RPHF has been shown to reduce water consumption significantly [11,12] and enable the implementation of efficient techniques for water flow-back management [13]. The controlled fracture growth in RPHF can minimize the risk of fluid migration and groundwater contamination and lessen the environmental risks associated with conventional hydraulic fracturing [14].…”

“…Through this comparative analysis, we aim to demonstrate the potential advantages of RPHF as a more environmentally friendly and efficient alternative to conventional hydraulic fracturing, highlighting its potential role in shaping the future of shale gas extraction. The following Figure 1 illustrates the cycle of water consumption during a hydraulic fracturing process [11].…”

Baseflow Separation for Improving Dam Inflow Prediction using Data-Driven Models

“…With the rapid development of data systems and computer hardware, the scope of data-driven models has expanded, and their use in the hydrological sector has increased over the past decade (Mosavi et al, 2018). For example, various types of data-driven models have been used for the spatiotemporal prediction of essential factors in water circulation, such as rainfall, flow rate, and soil moisture (Ahmad et al, 2010;Chen and Wang, 2022;Kim et al, 2022;Hao and Bai, 2023).…”

Prediction of Minimum Night Flow for Enhancing Leakage Detection Capabilities in Water Distribution Networks

“…As the leakage continues, the losses such as supply and pressure also increase. Recently, Neuro Fuzzy has been used to predict leaks in water pipe networks [4][5][6] and Long Short-Term Memory (LSTM) has been used to predict water flood, water quality, and consumption [7][8][9][10]. In this paper, using the flow data acquired from the pipe network monitoring system based on Information and Communication Technology (ICT), the predicted values calculated by LSTM and Multi-Layer Perceptron (MLP) are compared with the actual minimum night flow.…”