Graphic Operation Terminal-incorporated experimental prescheduled auto-configurable centralized rational water distribution system Open Access

Sustainable water supply management is a key challenge in the water–energy nexus. Traditional water distribution operates on fixed time durations, often leading to inadequate supply for end users. To address these challenges, the proposed work develops a testbed for a realistic water distribution model. An efficient and user-friendly automated system has been designed by integrating a Programmable Logic Controller (PLC), Graphic Operation Terminal (GOT), sensors, actuators, and the Internet of Things (IoT). A rational algorithm optimizes water distribution based on the availability and demand of water. The system dynamically configures water supply quantities for end users based on economic divisions and resource availability, validated through extensive case studies comparing traditional and automated modes. The statistical analysis of the proposed automation system demonstrated a distribution uniformity (DU) of 99.978% and a coefficient of uniformity (CU) of 99.984% when the tank's water level was high (3 m head), and a DU of 99.982% with a CU of 99.986% when the tank's water level was medium (2.55 m head). Additionally, the controller achieves 14–17% power savings by maintaining a higher overhead tank level.

By the 2050s water-scarce sub-basins may be increased due to rise in nitrogen pollution. In 2010, 984 sub-basins were water-scarce, and the number may shoot up to 3,061 in a worst-case scenario (Wang et al. 2024). Changes in industrial structure, climate, global residents' growth, and water pollution report substantial challenges to the ecological development of drinking water resources (Sun & Snyder 2024). Wastewater treatment is one of the significant solutions to minimize water scarcity. A bacterial consortium effectively removes chemical pollutants in domestic wastewater by using sand biofilters, offering a cost-effective, eco-friendly solution with minimal multistage filtration (Ibrahim et al. 2020).

Incorporating an industrial controller – Programmable Logic Controller (PLC) and Graphic Operation Terminal (GOT) technologies – simplifies and enhances the water distribution system by ensuring an adequate water supply while conserving both energy and water, making the system more cohesive and adaptable (Lu et al. 2021; Creaco et al. 2019). Water distribution networks often face low nodal pressure, causing insufficient supply. A significant water supply discrepancy was observed in Mekelle City, where only 50% of the population has access to drinking water from the system (Beker & Kansal 2024). Experimental verification of irrigation showed that the smart irrigation controllers consistently cut down water demand by 15% among common users and more than 40% among extravagant users (Lunstad & Sowby 2024).

Automation and control in water distribution reduces use of pump energy. Tests showed a 15.1–17.8% drop in specific energy consumption by maintaining pressure setpoints (Salvino et al. 2022). Excessive water use has caused numerous issues impacting its availability (Filho et al. 2024). Water allocation follows three key principles: (1) equity, ensuring fair distribution; (2) sustainability, balancing ecology and future needs; and (3) efficiency, promoting optimal use (Zhou & Han 2019). Trapped air in pipes poses a serious issue, leading to diminished performance or complete flow blockage in water systems. Studies assess the critical velocity needed to clear air pockets in various pipe sizes (Ramezani et al. 2016; May et al. 2018).

Residential water consumption is influenced by socio-demographic factors and user potential behavior pattern. Findings highlight that income, education, housing type, age, household size, and nationality are key determinants of household water consumption (Almulhim & Abubakar 2024). The analysis of samples collected from 789 houses in Dhani Mohabbatpur village revealed variability in per capita daily water consumption across different socio-economic groups (Singh & Turkiya 2013). A study of 394 households in Joinville, Brazil, found an average water use was 143.67 L/person/day. Key factors influencing consumption included income, household size, children, and housing features like building age, bathtubs, bathrooms, and pools (Grespan et al. 2022). A survey of 276 respondents across five quarters collected primary data. Twelve factors accounted for 85.79% variations in household water use (Ogunbode et al. 2023).

Digital technologies enhance utility operations, promoting a sustainable water cycle. An online survey of 64 utilities across 28 countries found water distribution as the key entry point for digital adoption (Daniel et al. 2023). Water distribution automation is a part of Industry 4.0, it enables precise measurement, optimizes operations, and cuts costs (Javaid et al. 2022; Malik et al. 2022). To tackle the water distribution challenges, the rational and optimal allocation of regional water resources has become a vital strategy (Yang et al. 2024).

Therefore, after comprehensive study, we have chosen the latest compact model FX5U PLC for automation of water distribution; these features include high-speed input/output (I/O), easily expandable (supports Max. 512 I/O), and high execution speed (Mitsubishi Electric-MELSEC iQ-F 2023). This I/O range can be increased for the benefit of majority people in villages by adopting the model iQ-R PLC (Mitsubishi Electric-MELSEC iQ-R 2023).

Considering real-time challenges involved in water distribution systems, we proposed an experimental prescheduled, auto-configurable, PLC-incorporated rational water distribution system integrated with a GOT. In traditional water supply systems, water is typically delivered from a tower. In this proposed work, we developed an experimental water distribution system that integrates the PLC, sensors, pump, and actuators (Solenoid Valves) to replicate the actual water distribution conditions. This system ensures that consumers receive sufficient water supply compared to traditional water distribution and saves energy consumption of the pump. In this study, the amount of water supply to consumers is determined by their specific needs, availability of resources, the number of individuals in the household, and economic classification. The proposed automation system will continue water distribution until the consumer receives enough water, regardless of the presence of air pockets. This system is designed to execute distribution operations rationally by meeting the water principles. The proposed water distribution system incorporates fundamental IoT technology, enhancing transparency and enabling remote operations.

The experimental efforts detailed in this proposed system address various challenges in the water distribution system. This research aims to answer various research questions (RQs) through implementing automation tools and proposed technologies. RQ1: How can ample water be supplied to houses with different altitudes within an existing water pipeline system? RQ2: What is the technique for creating a prescheduled and auto-configurable user-friendly graphical supervisory system for water distribution? RQ3: How can water be sustainably distributed to end users in rational mode when the liquid availability is below the required level? This study introduces diverse techniques to address the requirements outlined in the RQs. A prescheduled rational algorithm was formulated to ensure sustainable water distribution to each household. Additionally, various user-friendly graphical operating screens were created in the GOT for supervisory operations in water distribution control.

The structure of this work is outlined as follows. Section 1 describes the preambles of the water distribution system. Section 2 elaborates on the design of an experimental test bed for a real water distribution model. Section 3 details the hardware wiring of field devices for PLC. Section 4 discusses the automation tools. Section 5 describes experimental procedures and traditional mode supply responses. Section 6 deliberates the statistical methods for auto-rational water supply and their responses. Section 7 delves into the discussion of result analysis of water acquired for traditional and auto-rational mode, energy efficiency and sustainability aspects. Finally, Section 8 presents the paper's conclusion, scalability and future works followed by acknowledgments and references.

The nomenclature and abbreviations of the phrases in the content and figures of this paper are presented in Table 1.

Figure 1 illustrates the construction of a water distribution system designed to supply water to eight houses, divided into two stations: station-1 (serving four houses) and station-2 (serving four houses). During the pipeline construction, the height of the overhead tank and the lengths of the pipelines were randomly adjusted and tested to simulate irregular flow rates, replicating real-world water distribution conditions. The anti-corrosion level sensors (float type)-high-level sensor (HLS), medium-level sensor (MLS), and low-level sensor (LLS) were incorporated in the overhead tank to detect the water levels.

In this test bed construction, irregular flow rates were observed for a 500-L (1,000 mm height) overhead tank positioned at a tower height of 2,133 mm and connected to a 25 mm diameter main pipeline spanning 5,689 mm in length. The water supply extends to two substations, with each substation pipeline having a vertical length of 431 mm. Substation-1 connects four households through horizontal pipeline T connectors featuring various lengths (249 and 719 mm for left-side houses, and 249 and 736 mm for right-side dwellings). Similarly, substation-2 connects four households with varying lengths from its pipeline (249 and 747 mm for left-side houses, and 228 and 726 mm for right-side dwellings). Substation-1 and substation-2 are located at distances of 2,568 and 4,716 mm, respectively, from the overhead tank.

To measure flow rates at each house, flow sensors (flow sensors at station-1 houses: FS-S1H1, FS-S1H2, FS-S1H3, FS-S1H4 and station-2 houses: FS-S2H1, FS-S2H2, FS-S2H3, FS-S2H4) were installed (Figure 1). Each flow sensor used in the system is of type YFS301, which operates on a 5-V supply and has a maximum flow rate capacity of 600 LPH. These sensors generate electric pulse outputs corresponding to the flow rate of water and a pressure gauge was installed on the main pipeline to measure the pressure head relative to the tank level.

The block representation of the automated water distribution system is illustrated in Figure 3. In this system, flow sensors (FS-S1H1, FS-S1H2, etc.) and solenoid valves (SV-S1H1, SV-S1H2, etc.) from each house are connected to the FX5U PLC controller. The PLC is responsible for storing data of water flow rates, water consumption by end users and controlling the solenoid valves, automatically opening or closing them as needed. Since the solenoid valves require a driver circuit, a relay module is used to interface them with the PLC. A rational control algorithm is deployed in the PLC, which executes automated decision-making to ensure sustainable water distribution. Additionally, overhead tank level sensors (high, medium, and low) are connected to the PLC, allowing it to regulate pump operations based on real-time feedback from the sensors.

The water distribution status is visually represented on the GOT using graphical operations. To ensure optimal data transmission of sensors and solenoids data, Ethernet communication is established between the PLC and GOT modules. As shown in Figure 3, the communication infrastructure includes a network switch, enabling seamless connectivity among PLC, GOT, PC and router, where a personal computer (PC) is used for control algorithm implementation, PLC and GOT configuration, program deployment, online supervision and diagnostics for efficient troubleshooting. Furthermore, the built-in web-enabled functions of the PLC (Mitsubishi Electric-MELSEC iQ-F Web Function 2023) facilitate wireless communication with controllers via IoT devices. This functionality allows operators to perform supervisory operations remotely, including real-time monitoring and control via mobile devices.

In this communication scenario, sensors and switches are interlinked to the input module in a sourcing mode (Kohpeisansukwattana et al. 2024). The PLC provides a power supply capacity of 24 V, 400 mA for the PLC's input and output modules circuitry. Three water level sensors (HLS, MLS, and LLS) are connected from the overhead tank to the PLC (X11, X12, X13 terminals, respectively) for diagnosing water availability at three levels (High-Level, Medium-Level, and Low-Level), as illustrated in Figure 4. The master start and stop buttons are connected to X14 and X15 terminals. Input terminals are constructed with an opto-isolator circuit, safeguarding the PLC CPU from malfunctions or power fluctuations.

The Hall effect type flow sensor-generating 450 pulses per litre, was installed at each end user to determine the quantity of water supplied. The input terminals (X0–X7) of the FX5U PLC operate with a high-speed response (Mitsubishi Electric-MELSEC iQ-F Programming 2023). The eight flow sensors (FS-S1H1, FS-S1H2, FS-S1H3, FS-S1H4, FS-S2H1, FS-S2H2, FS-S2H3, and FS-S2H4) under the two substation houses are integrated into the PLC input terminals X0, X1, X2, X3, X4, X5, X6, X7, respectively (shown in Figure 4).

All the solenoid valves operate at 230 V AC. The FX5U PLC connects relay devices in transistor-source output mode, ranging from 5 to 30 V DC. The photo-coupler insulation circuit of output terminals safeguards the CPU from malfunction caused by power supply issues (Mitsubishi Electric-MELSEC iQ-F 2023). The eleven solenoid valves (main station-SV, substation-1 SV, substation-2 SV, SV-S1H1, SV-S1H2, SV-S1H3, SV-S1H4, SV-S2H1, SV-S2H2, SV-S2H3, and SV-S2H4) are connected to terminals Y13, Y12, Y11, Y10, Y7, Y6, Y5, Y4, Y3, Y2, Y1 of the PLC (illustrated Figure 4), respectively. The output terminals of the controller are integrated with the solenoid valves through the driver circuit (relay module), as shown in Figures 4 and 5.

Table 2 details the input and output terminal addresses along with their respective connected devices. The nomenclature X0, X1, …, X15 represents the PLC input terminal addresses, while Y0, Y1, Y2, …, Y13 represents the output terminal addresses.

The status of field elements, including sensors and solenoid valves, is monitored in the GOT at a centralized location, specifically in the control room. Technicians working in the field, particularly those outside control rooms and involved in pipeline systems, need access to the status of field elements. To address this, we have implemented an indicator panel (pointed out in Figure 5) to observe field elements' status easily. This panel proves invaluable in troubleshooting the pipeline system.

A single FX5U PLC (Mitsubishi Electric-MELSEC iQ-F 2023) is well suited for controlling water distribution in small to medium-sized community/villages. GX Works3, the latest generation of programming tool and maintenance software, is designed explicitly for controller's operations. It incorporates numerous features to ensure a trouble-free engineering environment solution for water distribution.

The home screen (Figure 9(a)) provides the operator with options to switch between status monitoring, station-1 control, station-2 control, entire station control, and utilization screens in a user-friendly manner. The home screen allows the operator to enable/disable auto (automation mode) and manual mode (traditional mode) operation. The status monitoring screen (Figure 9(b)) displays the current ON/OFF status of solenoid valves, overhead tank water levels, and water flow rates at individual houses.

The specified time duration for supplying water to station-1, station-2, and individual houses is configured through a numerical input for set time, as depicted in Figure 9(c) and 9(d). The GOT system empowers the operator to execute control operations simultaneously for the entire station-1 and station-2 (Figure 9(e)). The water utilization screen, illustrated in (Figure 9(f)), allows monitoring of the quantity of water utilized by end users across different stations. The screen-switching function facilitates the operator in transitioning seamlessly between different screens in a user-friendly manner.

In the proposed water distribution system, graphical monitoring and control operations can be performed using the GOT (touch panel for the human–machine interface). The experimental study examines water supply operations in two distinct modes. The first mode involves a traditional approach, where water is provided for a predetermined time without considering end users' specific needs. The second mode entails a rational automation system for water distribution considering end users' specific needs.

Step 1: As part of the experimental procedure for testing in traditional mode, the operator must first enable the manual (man mode) test operation mode on the GOT home screen (Figure 9(a)). When the mode is enabled, the graphical switch will show green indication. If it is disabled, the switch will appear red indication.

Step 3: To test the system in traditional mode of operations, graphical switches (Figure 11) are provided to manually switch ON/OFF the solenoid valves of main station pipeline, substation-1 and -2, and individual house's solenoid valve. Dedicated graphical screens have been created for each station to facilitate manual testing of water distribution in traditional mode. Figure 9(c) and Figure 9(d) represent the graphical control interfaces for station-1 and station-2, respectively. In addition to individual house control, a graphical switch located at the top right corner of Figure 9(c) and 9(d),) enables simultaneous control of an entire station. When the switch is activated, it opens the main solenoid valve, substation-1 valve, and the solenoid valves for houses 1, 2, 3, and 4 simultaneously. Furthermore, another graphical screen (Figure 9(e)) has been designed to allow the ON/OFF operation of all solenoid valves across station-1 and station-2 simultaneously for predefined time durations, ensuring flexible testing of the manual control system.

Consider the test operation of supplying water to all houses under station-1 under condition-1. The supply time duration for station-1 (Figure 11) must be set by entering the desired time (in minutes like 120) in the numeric input window. After specifying the time duration, clicking the ON/OFF graphical switch will activate all solenoid valves under station-1, allowing water supply for the selected period.

The status is color-coded, where GREEN indicates ON and RED indicates OFF. In case of supplying water to all the houses under station-1, even though all solenoid valves are ON, no flow rate is detected (Figure 12) at house-1 and house-4 due to insufficient water pressure. This scenario reflects real-world water distribution challenges, where houses located farther from the distribution station may receive little or no water due to low-pressure heads, while others may receive excess water due to high pressure heads.

TASK 1: The analysis was conducted by supplying water to the selected houses at station-1 and station-2, considering the tank water level conditions (high/medium).

TASK 1: Water was supplied in the traditional mode to station-1 and station-2 for 120 min.

TASK 2: The analysis was conducted by supplying liquid to the selected houses at station-1 only, considering the tank water level conditions (high/medium).

TASK 2: Water was supplied in the traditional mode to only station-1 for 120 min.

TASK 3: The analysis was conducted by supplying water exclusively to selected houses at station-2, considering the tank water level conditions (high/medium).

TASK 3: Water was supplied in the traditional mode to only station-2 for 120 min.

TASK 4: The analysis was conducted by supplying water to randomly selected houses for additional case studies, considering the tank water level conditions (high/medium).

TASK 4: Water was supplied in the traditional mode to a randomly selected houses for 120 min.

The graphs represented for case-1 to case-30 (indicated in Figures 14–17) illustrate the water obtained by individual houses in the traditional mode is relative to their demand. Analyzing the above case studies revealed that supplying water to houses in the traditional mode led to inadequate or excessive water distribution to end users. Some end users received zero water in case-1, case-5, case-7, case-10, case-11, case-14, case-15, case-19, and case-22 due to the pipeline's airlocks and inadequate pressure heads. In case-2, S1H4 received 116.4 L less than the demanded 923 L, while S2H1 received an excess of 719 L compared to the demanded 254 L. In case-3, S1H4 obtained 205.9 L out of the 923 L, in case-26, S1H2 attained 478 L out of the 292.5 L demanded and found many more irregularities in various case studies. Highlighting these variations transpired in water distribution due to the liquid supply for a fixed time, air pockets (May et al. 2018), and the lack of adequate gravity heads. It was observed that there were various time lags in receiving the water when it started supplying the water (Figures 14–17). The asymmetrical water distribution to different end users across multiple case studies was consistently observed from case-1 to case-30.

This approach automatically allocates demanded amount of water to end users. Under this water supply mode, the required water supply for each house is calculated based on the water consumption survey results of semi-arid village of Dhani Mohabbatpur (Singh & Turkiya 2013).

A survey on household domestic water consumption patterns was conducted for 763 households in the semi-arid village of Dhani Mohabbatpur, located in the Hisar district of Haryana, India. The study considered various activities, including washing clothes, drinking, house cleaning, bathing, flushing, cooking, washing utensils, livestock domestication, and other uses, to evaluate per capita water consumption. The volume of containers used for collecting household water was measured, and the number of containers used for various activities was recorded. In cases where running tap water was used, the duration of tap usage was determined, and the flow rate of water per minute was measured. The survey instrument also included both open- and closed-ended questions. It covered various aspects such as daily and activity-wise water consumption, water quality, sources, duration and regularity of supply, and the accessibility of different water sources.

By studying further references, it was noted that occupation, education, and other parameters can also be used as the basis for classifying socio-economic categories. However, household annual income remains the sole indicator for classifying various socio-economic groups, as it is easy to operate and understand. The identity-collected annual income data of households was categorized in five socio-economic classes based on annual income of the households in Dhani Mohabbatpur village. These five classes are categorized as SEC-A (>Rs. 50,000), SEC-B (Rs. 50,001–100,000), SEC-C (Rs. 100,001–200,000), SEC-D (Rs. 200,001–400,000), and SEC-E (>Rs. 400,000). According to the socio-economic class-based water demand survey, per capita water demands are as follows: SEC-A requires 97.5 L, SEC-B requires 111.4 L, SEC-C requires 127 L, SEC-D requires 131.9 L, and SEC-E requires 124.2 L of water.

A rational calculation-based algorithm is formulated to ensure sufficient water supply to end users. In the proposed system, the PLC automatically supplies the required amount of water to end users. The PLC's data registers (Table 3) store each household's demanded water, the acquired water quantity (AWQ) used by end users, and the set water quantity (SWQ) allocated by the PLC. The demanded water, based on the economic classification of all households (S1H1, S1H2, S1H3, S1H4, S2H1, S2H2, S2H3, and S2H4), is stored in the PLC registers (D50, D52, D54, D56, D58, D60, D62, and D64), respectively.

If the available water is greater than the required water for ‘n’ days and distribution is executed in automation mode, the PLC distributes the demanded water to households based on the data stored in its registers. During water supply to the eight houses, the AWQ, measured by flow sensors, is recorded in the PLC registers (D200, D202, D204, D206, D208, D210, D212, and D214), as the PLC is directly connected to the flow sensors.

If the available water is less than the required amount for ‘n’ days and distribution is executed in automation mode, the PLC runs a rational algorithm to determine the water allocation based on scarcity. The calculated SWQ is stored in the SWQ registers (D100, D102, D103, D104, D105). The PLC then distributes water according to the values stored in these registers.

Total required daily water for two station's (m) end users: 4226.1 L (Table 3). In this proposed model, the availability of liquid in reservoir (ALR) is calculated for ‘n’ days, where n is a variable depending on different seasons of the year, scheduling of supplying of liquid from primary water sources, and others.

• The ALR for the coming ‘n’ days is ‘f.’

• The required amount of water for ‘n’ days (r) = m*n.

The required amount of water (r) for ‘n’ days, along with the water demand of each end user (e.g., stored in register D50 for S1H1 house), is pre-programmed into the PLC controller. The PLC also receives real-time data on the available water (f) in the reservoir via sensors. It then compares the available water (f) with the required amount (r). If the available water (f) is greater than the required amount (r) the PLC automatically activates the solenoid valve for the respective household. Once the amount of water delivered to the household (e.g., stored in register D200) matches the requested demand (D50), the PLC switches off the solenoid valve, thereby completing the supply process. If the condition ‘f>r’ is false, meaning the available water is insufficient to meet the full demand, the PLC cannot supply the requested amount. However, the available water must still be distributed rationally. In this case, the PLC performs a rational division calculation to determine the proportion of water to be supplied (e.g., stored in register D100), based on the available volume, individual demand, and the duration of ‘n’ days, as outlined in formula D100. The PLC then activates the solenoid valve for each household and continues the supply until the acquired water (e.g., register D200) matches the calculated rational allocation (D100).

As represented in Figure 24, the comparative analysis reveals significant discrepancies in water distribution under the traditional mode. Households S1H1, S1H4, S2H1, and S2H4 did not receive any water, while S1H2 received an excessive amount. Additionally, S1H3, S2H2, and S2H3 received less water than required, highlighting a high degree of non-uniformity in distribution. As illustrated in Figure 25, all households received water; however, the distribution remained inconsistent. S1H1 received an excessive amount, while S1H4 obtained only 205.9 L out of its 923-liter demand. Similarly, S2H2 received 282 L out of the required 557 L. Despite all households being supplied with water, the distribution lacked uniformity, highlighting inefficiencies in allocation. In contrast, the proposed automation system effectively delivered the required amount of water to all households, ensuring equitable and sustainable water distribution. Similarly, a comparative assessment of the remaining case studies under traditional and automated modes highlights that the proposed automation system guarantees optimized, uniform, and efficient water distribution.

Figure 26 demonstrates the analysis of water allocation when the available supply for eight households over five days is 19,214 L, which falls short of the required 21,130.50 L. In such scenarios, the available water must be distributed equitably based on demand over five days. The PLC system efficiently performs rational calculations and proportionally allocates water among end users. Under the auto-rational mode, S1H1 received 461 L out of the required 508 L, meeting 90.07% of its demand. Similarly, S1H2 received 265.75 L out of 292.2 L (90.09%), S2H1 received 231 L out of 254 L (90.09%), and S2H3 received 692.86 L out of 762 L (90.09%). This approach ensures a balanced and systematic distribution of water, even in conditions of scarcity, thereby promoting sustainable resource management. This system adheres to the principles of water distribution, thereby retaining the feasibility benefits for operators and remote operations.

The environmental footprint of automation infrastructure of water distribution depends on multiple factors, including resource efficiency, energy utilization, material consumption, and emissions. Here's a breakdown of key attributes: In the traditional mode of water distribution, the pump is switched ‘ON’ continuously for fixed interval of time irrespective of overflow of water from overhead tank and requirement of water supply to the end users. This leads to more power consumption. In the proposed automation system, three sensors are installed to detect the high, medium and low level of water in the overhead tank. The control algorithm developed in this study ensures that the pump is switched off during overflow while delivering the required amount of water to end users.

Energy consumption of water supply pump was tested under two dimensions ((1) overhead tank with high water level (3 m head) and (2) overhead tank with medium water level (2.5 m). It was noticed that power consumption is 14–17% more when the water level of overhead tank is medium. In this proposed model, the controller always maintains high water levels in overhead tank with the help of water level sensor, in such a way that the proposed system minimizes the power consumption by the pump while also maintaining sustainability by reducing CO₂ emissions.

PLCs are with high end technology with a long lifespan, durability, cost effectiveness and eco-friendly mechanisms with focus energy conservation, but initial installation cost automation system for water distribution should be afforded (Pérez-Martínez et al. 2021). The FX5U PLC and GOT product manufacturers follow ISO 14001 environmental standards and restriction of hazardous substances (RoHS) compliance policies to minimize hazardous material use.

The statistical analysis of distribution uniformity (DU) and coefficient of uniformity (CU) (Hui et al. 2022) are essential for minimizing losses, optimizing water use, and ensuring equitable water distribution across different users.

DU and CU are effective tools for analyzing the efficiency and equality of the proposed automatic rational water distribution system. In this framework, DU determines how evenly water is supplied to the least-served houses, ensuring that even those at the low-pressure zones receive an adequate and fair supply. CU, on the other hand, evaluates the whole consistency of water supply across all households. High values of DU and CU denote a well-balanced system where every home receives a near-demanded share of water. These metrics help to identify system inefficiencies. By using DU and CU, we can optimize system design.

Traditional Mode:

When the water is supplied in the traditional mode, various deviations were observed in the amount of water received by end users out of their demand under various case studies. The following collected samples represent the percentage of the amount of water received by the end user out of their demand.

Samples collected (53-Samples) from traditional mode of water Supply from case-1, 2,3,4,7,8,9,10,15,16,17, 18, 23, 24, 25, and 26 under condtion-1:

Case-1: 0, 148.1, 59.7, 0, 0, 85.7, 36.6, 0, case-2: 101.1, 12.6, 283, case-3: 117.8, 22.3, 161.4, 50.6, case-4: 126.5, 115.3, 111.3, case-7: 0, 193.7, 89.7, 0, case-8: 216, case-9: 133.6, 47.8, case-10: 118.1, 95.7, case-15: 0, 163.9, 84.9, 0, case-16: 467.3, case-17: 113.3, 222, case-18: 215.3, 84.9, 5, case-23: 149.9, 60.5, 86.3, 37.1, case-24: 64.7, 79.1, 107.9, 41.3, case-25: 30.9, 81.2, 39.1, 49.6, case-26: 163.4, 28.62, 36.4, 99.8.

Automation Mode:

When the water was supplied in auto mode, the end users received the adequate amount of water under condition-1. The below collected samples represent the percentage of the amount of water received by the end user out of their demand.

Samples collected (53-samples) from automation mode of water supply from case-1, 2,3,4,7,8,9,10,15,16,17, 18, 23, 24, 25, and 26 under condtion-1:

Case-1:100.02, 100.02, 100.01, 100.04, 100.04, 100.01, 100,100.01, case-2: 100.03, 100, 100.05, case-3: 100.02, 100.04, 100, 100, case-4: 100.05, 100.04, 100.01,case-7: 100.01, 100.04, 100, 100.03, case-8: 100.05, case-9: 100.01, 100.01, case-10: 100.02, 100.02, case-15:100.06, 100.01, 100, 100.03, case-16: 100.04, case-17: 100.02, 100.03, case-18: 100, 100.06, 100.03, case-23: 100.04, 100.04, 100, 100.03, case-24: 100.05, 100.02, 100.02, 100, case-25: 100.06, 100.03, 100.03, 100.01, case-26: 100.02, 100.02, 100.03, 100.05.

Traditional mode:

Samples collected (45-samples) from traditional mode of water supply from case-5,6,11,12,13,14,19,20,21, 22, 27, 28, 29, and 30 under condtion-2:

Case-5: 0, 111.8, 46.9, 0,0, 105.3, 34.3, 0, case-6: 90.2, 21.2, 257, case-11: 0, 159.6, 71.8, 0, case-12: 183.7, case-13: 111.4, 28.8, case-14: 91.8, 91.8, case-19: 0, 125.2, 48, case-20: 241.4, case-21: 118.1, 71.8, case-22: 183.2, 61, 0, case-27: 110.7, 47.5, 67.9, 30.4, case-28: 51, 60.7, 87.7, 34.4, case-29: 24.6, 63.7, 33.1, 41.4, case-30: 124.7, 23.4, 30.7, 82.1.

Automation mode:

Samples collected from automation mode of water supply from cases 5, 6, 11, 12, 13, 14, 19, 20, 21, 22, 27, 28, 29, and 30 under condtion-2:

Case-5: 100.02, 100.01, 100.01, 100.04, 100.03, 100.01, 100, 100.01, case-6:100.02, 100, 100.04, case-11: 100, 100.03, 100, 100.03, case-12: 100.01, case-13: 100.01, 100.02, case-14: 100.04, 100.01, : case-19: 100.05, 100, 100.03, case-20: 100.03, case-21: 100.01, 100.04, case-22: 100, 100.05, 100.03, case-27: 100.03, 100.04, 100, 100.02, case-28: 100.04, 100.01, 100.02, 100, case-29: 100.05, 100.02, 100.02, 100, case-30: 100, 100.02, 100.02, 100.04.

The statistical comparative analysis of DU and CU for traditional and automated modes under condition-1 and condition-2 are presented in Table 4. It was observed that the DU value for the traditional mode was 10.625% under condition-1 and 14.028% under condition-2. A low DU indicates an uneven water distribution, where some end users received zero water supply. Under condition-2, water was supplied for 120 min at a lower pressure compared to condition-1. This resulted in a reduction in excess water received by end users, thereby lowering the average value of all samples and increasing the DU value compared to condition-1 in the traditional mode.

In contrast, the DU values for the automation mode were recorded as 99.978% under condition-1 and 99.982% under condition-2, indicating highly uniform water distribution and efficient system performance. The CU values for the traditional mode were recorded as 33.219% under condition-1 and 31.725% under condition-2, indicating inconsistent water distribution, high variability due to irregular pressures, and reduced efficiency. In contrast, the proposed automation system achieved a high CU of 99.984 under condition-1 and 99.986 under condition-2, demonstrating efficient resource use and optimized system performance.

Analysis of data samples from various case studies revealed that, in the traditional supply mode, some end users received excess amount of water. This occurs because the supply pump operates for a fixed duration, regardless of actual water demand. As a result, there is excess consumption of water. The statistical analysis presented in the Table 5 demonstrates that the proposed automation of water distribution approach reduces water wastage.

Table 5 presents an analysis of the excess water received by end users in both traditional and automated distribution modes, based on various case studies (Figures 14–17) under condition-1 and condition-2. It was observed that, in the traditional mode, end user S2H1 under case-16 (Figure 16) received a maximum of 933 L of water. Similarly, in different case studies, end users received varying amounts of excess water due to differences in pressure heads at their respective locations during water distribution.

In contrast, under the automated water supply mode, the maximum excess water received by an end user was recorded as 0.25 L. In the proposed system, the PLC ensures that only the demanded amount of water is distributed, regardless of pressure variations. Once an end user receives the allocated amount, the PLC automatically closes the corresponding solenoid valve. This approach significantly reduces water wastage compared to traditional water distribution methods.

Limitations of the proposed system:

• The selected flow sensor output pulse should be less than 200kHz for the selected FX5U PLC.

• The use of poor-quality solenoid valves can result in leakage in the water pipeline system, even when the solenoid valve is not activated.

• In the proposed automation system, a continuous power supply is essential for water distribution. A backup power source is necessary to ensure smooth operation of the water distribution system during power outages.

• Water distribution systems have a scope for physical and cyber-physical attacks. Hence, implementing security functions is crucial to protect the system from unauthorized threats.

Potential areas for future research:

As part of future research, there is significant potential to develop an algorithm capable of detecting attacks during both the scheduling and non-scheduling phases of automated water distribution, along with evaluating the severity of such attacks. Additionally, implementing multi-directional security measures can further strengthen the protection of the water distribution controller against cyber-physical attacks originating from various threat vectors.

This paper investigated the water obtained by end users in traditional and auto-rational modes of operation. It specifically highlights that the oversight of the economic sectional division of end users and the number of consumers in a household results in inefficient water distribution. It also identified factors like air pockets, pressure variations, and elevation differences as key contributors to flow inconsistencies. An experimental test bed for an automatic water distribution system has been developed, integrating a rational algorithm to optimize supply based on resource availability and economic divisions while addressing airlocks and pressure variations.

The proposed automation system achieved a high DU of 99.978 and 99.982% and a CU of 99.984 and 99.986% under water supply conditions at overhead tank's water level is high (3 m head) and medium (2.5 m head), respectively. The automated water distribution system saves 14–17% power by maintaining tank level is high and preventing unnecessary pump operation. Experimental analysis of the developed prototype showed a maximum excess water supply of 933 L in traditional mode, compared to just 0.25 L in automated mode. The rational algorithm confirmed a balanced and systematic water distribution, even when available water was insufficient, thereby supporting sustainable resource management. The IoT-enabled PLC ensures remote monitoring and control via PCs and smartphones.

The features of the proposed automation system are more flexible for scalability to large regions. One iQ-R series PLC can integrate 4096 I/O devices for the end users of water resources. Maximum ‘120’ PLC's can be integrated for large population regions on a single network platform, which can communicate nearly 5 Lakh I/O devices. The network capabilities of PLC allow us to communicate with multiple PLCs with the data transmission speed of 1Gbps. Various regions can be integrated to share data of availability of water resources in those regions and PLC allows to share the resources to the water scarcity regions, which leads to increase sustainability.

The future scope of this model is to promote classified security functions for physical and cyber-physical attacks on water distribution systems to save the water. The water distribution system makes the path easy to generate energy by installing small water turbines at each end user's house, this effort creates relatively low environmental footprint. Adding PLCs with variable frequency drives helps reduce energy consumption in water supply pump-driven applications.

The authors wish to thank Mitsubishi Electric India-Factory Automation ATC-CVRCE, Hyderabad, for their generosity in providing equipment and software support.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.