Demonstration of real-time monitoring in smart graded-water supply grid: an institutional case study

An adequate water supply is requisite for the sustainable development of both urban and rural areas. Rapid climate change, population growth, urbanization, unsustainable water extraction, and improper wastewater handling have threatened the clean water supply across the globe (Li et al. 2020). It is projected that over 50% of the world's population will be under extreme water scarcity by 2050 (Koncagül et al. 2020). India is one of the water-scarce countries with only 4% of world water resources, therefore, at a high risk of a water crisis (He et al. 2021). Increasing demand and limited supply of water have raised big challenges for both the government and consumers to manage water resources and supply. Most conventional water distribution systems (WDSs) are either not fully equipped to measure, control, and reduce the loss of water or aging. The disparity between the amount of water supplied into the distribution system and the amount of water utilized by the end user is another key concern plaguing water utilities in the country. A significant fraction of the water supplied to a region is lost owing to water leakage during distribution in network pipes, metering mistakes, or malpractices in consumption. Manual handling and lack of real-time analysis of water network monitoring data further exacerbate water losses in the water distribution grids. To mitigate these water-related issues, the United Nations is promoting the digital transformation of WDSs to ensure the availability and sustainable management of water and sanitation for all primarily in developing countries under Sustainable Development Goal (SDG) no. 6 (United Nations 2023). Advanced technologies like artificial intelligence (AI), the Internet of things (IoT) (Mohanty et al. 2016), and new-generation information communication technology (ICT) (Ahad et al. 2020) are playing important roles in resolving urgent water-related problems. Some recent reviews (Zaman et al. 2020; Oberascher et al. 2022a) provide a comprehensive analysis of potential applications of ICT in urban water infrastructures to enhance sustainability and increase the quality of life for citizens. Therefore, adopting advanced technologies is paramount in wiser decision-making and achieving SDG-related metrics as society transitions to a more sustainable future. In addition, many countries around the world are adopting integrated water resource management to address area-specific water challenges with varying degrees of success (Cacal & Taboada 2022; Tayyab et al. 2022). Understanding the interactions between ground and surface water resources, agriculture, environment, and society helps access the water needs, policies and required action plans for long-term water management.

A smart water grid (SWG) is an emerging paradigm for long-term sustainable water sources and services (Fabbiano et al. 2020; Baanu & Babu 2022). The concept of SWG combines advanced tools with ICT as an integral part of the solution for water distribution and management challenges (Lalle et al. 2021). It improves the accessibility and quality of water while simultaneously alleviating water-related issues such as leaks, pressure, flow, and demand in a sustainable manner. Access to robust and actionable data is required for water supply management, planning water resources, designing water infrastructure, and conducting scientific research. Therefore, gathering precise and useful water data is crucial. The deployment of smart IoT sensors to generate real-time data is one of the novel technologies that should become particularly crucial for data collection (Oberascher et al. 2022a). Monitoring and analyzing real-time water consumption facilitate systematic and intelligent decision-making in managing water distribution and treatment with minimum use of energy and waste of water in the entire water distribution grid. Although technological information on the smart water grids is accessible, but the path to their deployment is dubious. The key impediments include insufficient integrated and open solutions, difficulties in meeting user's needs, lack of case studies with verified solutions, general awareness, and a lack of regulatory aid. Case studies are useful in learning and adapting solutions to problems associated with real-world scenarios.

During the last decade, several case studies and projects have been reported in the literature to demonstrate the impact of advanced technologies on water distribution (Savić et al. 2014; Kouroupetroglou et al. 2015; Abbas et al. 2017; Rizzoli et al. 2018; Antzoulatos et al. 2020; Koo et al. 2021). However, demonstration studies on smart water distribution networks (WDNs) in universities and academic campuses, particularly in arid or semi-arid climates are quite scarce. Universities are considered to be major water consumers with a large number of users and diverse activities (EPA 2012; Almeida et al. 2021). Studies have reported that water consumption per capita is typically higher in university campuses (Yagoub et al. 2019; Adams & Jokonya 2022). Management of water supply in large university campuses requires real-time access and visualization of water quantity and quality data. Information on heterogeneous water use by students, faculties, staff, and building types in different time domains allows operators to distribute water with reduced cost. For example, the United Arab Emirates University carried out a study on indoor water use to an understanding of the broad pattern of water consumption on the campus (Yagoub et al. 2019). The findings revealed that 47.5% of water use occurred in residential buildings compared to academic buildings based on activity-driven consumption. Among the most recent, the campus of the University of Innsbruck is reported as a Smart Water Campus with a range of sensors including 12 water meters and 6 pressure sensors for monitoring the WDN, and weather stations installed on the ground (Oberascher et al. 2022b). The infrastructure offers an innovative testbed for smart and data-driven applications in the field of urban water infrastructure.

In India, the state of Rajasthan is a physically and economically water-scarce region thus under a significant water risk (Niti Ayog 2018). Water contamination due to suspended solids, turbidity, fluoride, and salinity is very common in rural and remote regions (Swami et al. 2018). The western region of Rajasthan is mainly occupied by the Thar Desert and, therefore, has a number of drought-prone districts. In addition, the availability of potable water per person is declining steadily due to depleting water sources, population growth, and urbanization. Jodhpur is one of the second largest cities in the state that experiences severe water-related issues such as contamination of water resources by industry (namely, textile and steel industries) wastes. Being in the western part of the state, it also suffers from geogenic climate effects, surface water inaccessibility, water loss, and over-exploitation. Since water is subjected to treatment before it can be used for many useful purposes the overhead in terms of energy consumption varies across different treatment procedures depending on the available sources of water. Over the past two decades, the city has expanded its suburban areas to the northern and southern regions including the settlement of several governmental institutions and organizations. High water demand with limited infrastructure potential for an augmented supply of water has become a common scenario in the region including public and educational institutions. Due to the expansion of higher education in the past few decades, it has become imperative to plan and manage water efficiently at higher educational institutions and other university campuses in an efficient manner. Yet, many institutions, such as the Indian Institute of Technology Jodhpur (IIT Jodhpur), currently operate their WDSs manually. The campuses require water for variety of purposes such as drinking, cleaning, cooling, landscaping, and more. Typical operation is based on water supply from the treated source and emerging conditions to start and stop the pumps and gather data. The water flow and quality information are disruptive, imprecise, uncoordinated, and sometimes come too late to be useful. Moreover, single-source water and original water infrastructure do not take into account the high demands for water posed by the current scale of use. Furthermore, the Government of India is encouraging higher educational institutions in India to transform their campus infrastructure to green and carbon-neutral by 2030 (World Economic Forum 2021). It is, therefore, necessary to upgrade the water infrastructure with advanced monitoring and management intelligence to meet the future demand of increasing student-centric needs with reliable supply and real-time response toward water-related issues. Moreover, the treatment of wastewater and greywater is important in closing the loop by water reuse and recycling to cope with the varying demand and limited supply of water.

This paper reports a case study of the demonstration of the ‘smart graded-water supply grid’ to sustainably cater to the growing water supply needs of an academic institution in India. A smart graded-water supply grid of capacity 1 MLD (million litres per day) is developed to maximize water use by sustainable management of graded-water distribution inside the institute. The system can distribute bulk water based on graded-water requirements using smart sensors and an ICT network in combination with IoT. Water quantity and quality are continuously monitored in real-time to monitor water consumption, leakages, and quality parameters. Water supply and consumption data are crucial for the system operators to implement water and energy management plans for achieving net-zero sustainability goals within the campus. Real-time data is useful to optimize the water pressure in the distribution network according to the demand pattern. Demand-driven operation cuts down the pumping energy, while still meeting minimum pressure and water quality requirements. The demonstration of a smart graded-water supply system will raise awareness regarding the potential of similar innovative projects at university campuses or municipality scale to reduce water demand while ensuring quality standards. The aim of this study is to share knowledge and disseminate meaningful data essential for developing an SWG for an Indian community. The focus of the work stands on operational as well as future-ready water distribution solutions in India, which need an integration of the water grid to create mutual benefits with respect to reliability and energy savings.

The structure of the paper is organized as follows: Section 2 describes the methodology followed during the implementation of the smart graded-water supply grid and its components. It includes a detailed description of the site, and water supply system under study; the deployment of sensors, and monitoring system is also explained in this section. Section 3 presents the data set collected from the sensors to understand the water use patterns and demand over diurnal and weekly time periods. Section 4 discloses the lessons learned and future work direction. Finally, the paper is summarized and concluded in Section 5.

The existing pipeline system has the capacity to supply approximately 1 MLD daily. The primary source of water supply to the institute is operated by the Public Health Engineering Department (PHED), Jodhpur. The source water is fetched through the Indira Canal from the Manaklao pumping station through the trunk main, which is approximately 10 km from the campus. The pre-treated water is first collected in the reservoirs for further treatment at the water treatment plant (WTP) inside the campus. The water is graded and treated before distributing it on the campus from the pumping station (as seen in the inset on the right in Figure 2) for graded water services. The domestic water is treated at the treatment plant through a process consisting of sedimentation, filtration, and final disinfection with liquid chlorine. In addition, responsible and sustainable strategies are adopted by the campus to save local water, and energy. The greywater produced on the campus is treated through four decentralized anaerobic wastewater treatment systems and recycled for non-potable uses like irrigation using pop-up irrigation systems and drip irrigation. It ensures efficient use of water and reduces the institute's water demand. The campus also has rainwater harvesting systems where the rainwater is collected on building roofs, and is then channeled through swales to the rainwater retention pool/pond. However, in the current scenario, the recycled water and the harvested rainwater are not fed to the distribution grid. The campus also has artificial ponds to accommodate reserve water for wildlife on the campus during the dry season.

The WDS of the campus is operated by the Office of Infrastructure Engineering of the institute and supplies graded water services within the campus. The campus has 17 student hostels, 3 faculty and staff residential complexes (Types B, C, and transit), 12 academic units (departments, centers, and schools), 1 administration building, 1 lecture hall building, and a library, 1 club, 1 sports complex, 1 health center, 2 shopping and food complexes, and other academic and non-academic units. Several other academic, research, sports, and cultural activities over the year are also catered by the same supply grid. Approximately 5,000 users, including students, faculties, staff, and visitors, are currently meeting their domestic water requirements from the grid. It was estimated that students' hostel water demand is over 30–35% of the total water consumption on the campus. Therefore, the focus of this paper will be on the domestic water supply grid. However, the system is functioning remarkably close to its maximum capacity, and often due to limited supply from the source unable to deliver water with sufficient pressure heads. The Smart Graded-Water Supply project aims to fulfill the rising demand for domestic water supply is growing fast due to the development of the institute's infrastructure and population. It will enable efficient management of the existing water supply by reclaiming water for potable and non-potable uses. The students of IIT Jodhpur are the true beneficiaries of the project.

Sensor nodes allow continuous monitoring of the distribution network and record pressure, flow rate, and important quality parameters in real-time or near real-time, which are crucial in early warning systems. Suitable monitoring location helps detect leaks (Zaman et al. 2020) and contaminants (Adedoja et al. 2018), therefore, permitting better diagnosis of the supply system. However, owing to the provision and the maintenance cost of the sensor nodes, limited yet strategic nodes were prudently deployed.

The corresponding pressure variations can be seen in Figure 10(b). The pressure variation at each node is similar with most values close to the nominal network pressure of 1.5 bar. The figure demonstrates spiked patterns consisting of pressure transients that are caused by rapid water consumption changes and recurring pump operation. Similar patterns are detected in the time-series pressure data of a sensor in the WDS of a Dutch drinking water company, Vitens (Geelen et al. 2019). It is critical to monitor pressure in the supply networks to avoid degradation of the pipes or valves caused by high water pressure.

It is worth mentioning that the institute practices various ways possible to secure the continuous supply of water and to meet the demands of the residents. Pump scheduling minimizes the night flow and ensures adequate supply in the campus during limited supply days. After midnight, the pump is shut down until early morning to conserve water otherwise wasted as night flow. The event is characterized by a gradual decrease and sloped pattern in pressure and zero flow rate during the period. For example, the weekdays data set except the one collected on Thursday displays no flow in the pipeline from midnight until the early morning hour (as seen in Figure 11). In contrast, a minimum night flow of approximately 5 m³/h can be observed on Thursday when a continuous water supply is maintained. In addition, an event of sudden spike (framed in a rectangle in Figure 11) in water flow rate occurs during early morning hours, which demonstrates pump activation causing the pressure to rise sharply before resuming the normal supply pressure. Besides oscillations, spikes, and slopes, a valley pattern is also seen (encircled in Figure 11). It is characterized by a brief but considerable drop in pressure displaying a sudden and momentary increase in water consumption. The features in the pressure data are consistent with the recurring events found in Viten's WDN data (Geelen et al. 2019) useful in locating abnormal events from the usual.

Regulating the pH of the water in the distribution system is also necessary. For effective disinfection with chlorine, water pH should preferably be less than 8.0. Failure to do so can result in the contamination of drinking water. The pH of water monitored during the study period is displayed in Figure 13(b). The data show that the pH variation is within a permissible range of WHO, i.e., between 6.5 and 8.5. The lowest and highest pH values exhibited are 8.27 and 7.24, respectively. The relationship of pH with water temperature can also be simultaneously observed where the latter has fluctuated between 26.5 and 47.5 °C. The high water temperature may be a result of water pipes being exposed to the direct sun at the water quality sampling location. In the second phase implementation, the amount of unwanted biological and chemical processes will be measured at various nodes in the grid.

Real-time monitoring of graded-water is essential for a reliable and sustainable water supply. Yet a number of practical implications during the first phase of the project were realized and are discussed in this section. Although no one solution fits all the lessons learnt might be useful in the deployment of monitoring systems or similar solutions in smart water grids. Time series flow and pressure data offer detailed insights and conclusions with respect to the water consumption pattern by different inhabitants. However, in the present study, only a simple analysis of a selected sample of raw water distribution data is presented. In order to enhance the data interpretation, other techniques should be utilized that are suitable for monitored variables or validation of not only the raw data but metadata. In addition, factors causing redundancies in the data must be eluded in future implementation for better robustness and performance of the supply system. In the first phase of the implementation, network coverage problems were occasionally experienced, which resulted in delayed installation, interrupted sensing, and data logging. One potential future work direction may tackle coverage problems while considering several metrics and issues such as connectivity, energy management, topology, and realistic sensing. The inclusion of stakeholders covering technology partners, regulators, financial agencies, local governance, and line-ministries is critical to help ease the development as well as rapid adoption, hence must be guaranteed. Therefore, as a scope for future development, the framework demonstrated in the present work will consider the following:

• i.

  Steady scaled up by including more robust sensor nodes to improve the efficiency and security of the entire supply system to understand water users' behavior and how hydraulics impacts the quality of the grid.

• ii.

  Installation of more water quality nodes to monitor and regulate the graded-water quality standards and to provide feedback on the treatment process and its performance thereby improving the reclaimed and reuse water quality.

• iii.

  Data modeling to validate, evaluate, and analyze data from each district metered area with a capability of determining anomalies.

• iv.

  Hydraulic modeling of the network to predict systems performance in various scenarios and for developing the digital twin of the supply grid for the future implementation of smart graded-water management.

• v.

  Cybersecurity, data privacy, and governance requirements for the security and success of the future-ready smart graded-water supply grid at all application levels.

Continuous real-time monitoring of water supply grids is essential for the efficient management of water resources and supply by enabling analysis of data collected. This paper reports insights from a case study, where a framework for the smart graded-water supply grid is developed and demonstrated in an academic institution in the semi-arid region of India. The framework takes graded-water supply as well as quality assurance into consideration, which can be replicated in other university innovation projects or smart cities. The primary focus of the paper implicates the deployment of sensory nodes, and real-time monitoring framework, which provides valuable data on water flow and network pressure. Results present the time-series data monitored and interpreted to understand the student-centric water consumption pattern, status of domestic water supply, leaks and anomalies and to disseminate effective strategies for water conservation and energy in the system. Near real-time monitoring of pH and chlorine concentration in the water supply ensures satisfactory disinfection to the standardized level. The high-resolution hydraulic data provided by sensors has potential not only in terms of its usefulness to assist timely decisions but also in developing and calibrating the hydraulic network model and comparison with the real-time network simulations. Moreover, the accessibility of data informs users about their water use and its effects on present and future generations. The developed framework of the smart graded-water supply grid will serve as a testbed network allowing researchers and practitioners to test and validate core technologies for future Smart Water Grids in Smart Cities with great societal and economic impact.

The authors like to acknowledge the Office of Infrastructure Engineering, and the Jodhpur City Knowledge and Innovation Foundation (JCKIF), and IIT Jodhpur for providing the information and resources to execute the work.

Data cannot be made publicly available; readers should contact the corresponding author for details.

The authors declare there is no conflict.