Content uploaded by Immanuel Savio
Author content
All content in this area was uploaded by Immanuel Savio on Jul 05, 2021
Content may be subject to copyright.
An IoT-Based Water Management
System for Smart Cities
Immanuel Savio Donbosco and Udit Kr. Chakraborty
Abstract Water conservation is one of the prime concerns in the current scenario
where environmental conditions are deteriorating at an alarming rate. Smart cities,
unlike the conventional system of living, are in the frontline of innovation in terms
of both connectivity and technological advancement. The main idea is to use the
available technology to make life easy with lesser harm to the environment. An
Internet of Things (IoT) and data analytics (DA) based water management system
will be a basic ground for implementation and future research on how data and IoT
can be used to make this happen. This paper proposes an Internet of Things (IoT)
and data analytics (DA) based water distribution cum management system that would
help in optimal distribution of water based on user consumption at the plot holding
level. The proposed system would not only save water misuse but also help in storing
usage data for analysis and town planning at a macro-level.
Keywords Water management ·Data analytics ·Internet of Things ·Smart city
1 Introduction
Smart cities combine technology and innovation in day-to-day living. The idea behind
every smart city is that the application of IoT and intelligent techniques in regular
everyday activities can increase the ease of living and efficiently perform these activ-
ities with less effort [1]. In a smart city, the basic idea of IoT is implemented which
is “anything that can be automated will be automated”. One very important task in
a smart city or any city for that matter is water distribution and conservation. In
traditional cities and housing communities, water is distributed manually with the
use of traditional pipe and motor systems [2]. This being very inefficient in terms
I. S. Donbosco (B)·U. Kr. Chakraborty
Department of Computer Science and Engineering, Sikkim Manipal Institute of Technology,
Majitar, Sikkim, India
e-mail: immanuelsavio@gmail.com
U. Kr. Chakraborty
e-mail: udit.c@smit.smu.edu.in
© Springer Nature Singapore Pte Ltd. 2021
C. Bhuiyan et al. (eds.), Water Security and Sustainability,
Lecture Notes in Civil Engineering 115,
https://doi.org/10.1007/978-981- 15-9805- 0_21
247
248 I. S. Donbosco and U. K. Chakraborty
of availability and also conservative management of this valuable resource, planned
and efficient water management is required. A water management system with the
implementation of IoT and data analytics is expected to decrease water wastage and
also be more efficient in terms of water availability and also from a conservation point
of view [3]. This paper presents a plot holding level plan for smart water manage-
ment. Applicable in upcoming smart cities [4], this IoT- and DA-based system uses
user consumption behavior to optimize the water distribution. Using smart sensors,
the proposed system also presents a plan to harvest rainwater at the plot level. The
following sections of the paper will give detailed explanation on the basic terminolo-
gies used in this paper. It gives the specifications of the sensors and their usage. The
next section will give the architecture of the water supply block. Following, it is the
structure of the water tank and how its structure will help the rainwater harvesting
mechanism and how the sensors are placed on the tank. The final part is the descrip-
tion of the water distribution of the water in the smart city and how it is optimized
with the use of DA. In the end, the conclusions and future works summarize the paper
and the methodologies proposed and present the possible research and advancement
in the proposed methodology.
2 Existing Distribution Method
Urban water distribution methods, or the conventional method of water distribution
which is being followed in cities all around the world, work but are inefficient and
have a lot of opportunities for development. It has a one-way cycle, consisting of the
reservoir/any water body, treatment plants, water storage, homes/consumers, finally
to the waste management/treatment facility and send back to nature as groundwater
or natural water (Fig. 1). This method of water distribution can have certain downfalls
like inefficient water supply or unavailability of the resource at times due to unwanted
Fig. 1 Conventional water distribution architecture
An IoT-Based Water Management System for Smart Cities 249
usage or an architectural failure in the system. This paper has been targeted to over-
come some of these basic failures that might occur in the conventional method of
water distribution.
The aim of this paper is to optimize the distribution method, to achieve better,
faster distribution and also better use of technology to conserve water, which is the
need of the hour.
3 Basic Terminology
This section presents in brief the terminologies and technologies used in this paper.
This includes IoT, data science, the different sensors and its purpose.
3.1 Internet of Things (IoT)
The growth in computing power, miniaturization of components and higher band-
width for faster communication has helped in the development of Internet technolo-
gies. What was initially envisaged as only a digital mailing system has now grown
to be the World Wide Web. Taking full advantage of such technologies, an exciting
new field has emerged which allows unprecedented control over devices spread
over distant locations and enables data gathering for analysis and intelligent control.
Termed as Internet of Things and better known as IoT, this is the systematic interrela-
tion or interconnection of devices for different machines, objects or even humans each
with a unique identifier and all interconnected with each other all through the Internet
[5]. The main aim of Internet of Things is automation, human–computer interac-
tion and data collection and analysis. Concepts like embedded systems, automation,
wireless sensor networks and control systems, etc. contribute to Internet of Things.
IoT has found widespread applications in home automation, security systems,
elder and child care, personal health care, telemedicine, transportation and traffic
control, livestock tracking and farming management, wild life and defense to name
a few. Applications are growing by the day and are limited only by imagination and
sensor availability. One such application on water management is presented herein
below.
3.2 Smart City
A smart city is an urban area where sensor-based Internet of Things is used to aid
in management of resources and civil life. The sensors are used to collect data from
various sources to be analyzed and used for better utilization of assets and services.
This data may include citizen count, traffic data, electricity usage data, rainfall data,
250 I. S. Donbosco and U. K. Chakraborty
etc. which can later be used in analytical models to draw inferences to significantly
change the quality of life for the citizens. “Smart city” or a “digital city” is the use of
modernized techniques in communication, sensing, analysis and integration to run
everyday living conditions [6].
Smart implementation of technology can give intelligent and prompt responses to
different needs including but not limited to commute and traffic management, public
safety, resource distribution and management and commercial trade and activities.
In layman terms, a smart city is a way of living with and aided by technology and
data. Unlike conventional cities, smart cities integrate technology with governance
and that is what makes smart cities different. Automation is made from the smallest
entities such as a simple traffic light to more complex infrastructure such as water
supply, energy transfer, governance and emergency situation handling. A smart city is
therefore portrayed as being better equipped to face growth-related changes through
a simple transnational relationship of governance with citizenry and resource usage
patterns.
3.3 Data Analytics (DA)
Increasing environmental awareness among people and the desire to work fast and
effectively through the reduction of time spent on unproductive activities is forcing
the realization on governments toward building smart cities. While one important
aspect of this lies in connectivity through sensors, the other is data analytics.
Data analytics (DA) is the basic science of analyzing trends and features in data.
As the name suggests, it deals with analyzing patterns in data to derive informa-
tion on which decisions can be based. Data analytics is a sub-section of machine
learning, as it uses several machine learning techniques to analyze the data including
regression, classification techniques and also bagging and boosting techniques such
as XGBoost and AdaBoost [6]. Use of analytical techniques in IoT and smart city-
based applications can improve their performance [7]. Analysis of traffic patterns
can help in deciding traffic flow regulation. Weather data analysis can help in street
light control, which may additionally also depend on the traffic density saving power.
Analysis of carbon emission data may help in devising means of reduction of emis-
sion. A host of such application can be found through intelligent sensor usage and
data analytics. Data analytics in IoT is almost always clubbed with cloud techniques
to improve data retrieval management so that the data is safely stored and accessible
for use. The current paper proposes an IoT-based water management technique,
which augmented with data analytics can be used in smart cities for effective water
management and sustainable town planning.
An IoT-Based Water Management System for Smart Cities 251
3.4 Sensors and Transmitters
The data for the analysis obtained from the water supply mechanism and the water
reservoir and tanks present in the city is collected continuously using sensors and
transmitted to the cloud [8]. This following section has the basic description of the
technology used. This project uses the Arduino UNO is an open-source microcon-
troller to connect and coordinate the different sensors. Arduino UNO is an open-
source microcontroller developed by the Arduino group [9]. This board has 14 I/O
pins to connect different sensor and transmitter to collect the data and transmit the
data to the cloud. It is programmed using the Arduino IDE and a simple USB cable.
This board is powered by a 9volt power supply. Different sensors are used to measure
the water conditions and levels for data collection [8]. The microcontroller, sensors
and transmitters used are:
•Arduino UNO
•SRF-05 sensor (ultrasonic sensor)
•YF-S201 Hall effect sensor (Water flow sensor)
•LM393 chip-based sensor (Rain sensor)
•ESP-8266 Wi-Fi transmitter
•Stepper motor
1. RF-05 sensor (ultrasonic sensor): The SRF-05 ultrasonic sensor (Fig. 2)isa
wide range distance sensor which uses the SONAR technology [10]. The SRF-05
sensor has a range of 4 m in total.
It is used in the top of the water tank in this project to measure the amount of
water present in the tank. Let x be the height measured from the top of the tank
to the surface of water level and h, r be the height and radius respectively of the
cylindrical tank. So, the volume of water in the tank can be calculated using the
following formula.
πr2(h−r)(1)
2. YF-S201 Hall effect sensor: Accurate flow measurement is an essential step in
terms of both qualitative and economic points of view. Flow meters have proven
excellent devices for measuring water flow, and now it is very easy to build a
water management system using the renowned water flow sensor YF-S201 [11].
This sensor sits in line with the waterline and contains a pinwheel sensor to
measure how much water has moved through it. There is an integrated magnetic
Hall effect sensor that outputs an electrical pulse with every revolution.
3. LM393 chip-based rain sensor: The LM393 chip-based rain sensor is a sensor
used to sense whether there is rain or not. This sensor uses the principle of
resistance. It is basically connected to the Arduino board to sense the presence
of moisture on the rain board. If there is moisture, then a signal is sent back to
the microcontroller.
252 I. S. Donbosco and U. K. Chakraborty
Fig. 2 SRF-05 ultrasonic sensor with Arduino UNO
4. ESP-82866 Wi-Fi transmitter: The ESP-8266 [12] (Fig. 3) Wi-Fi transmitter is
the low-cost wireless microchip with a fully integrated TCP/IP stack ready for
deployment with any microcontroller such as Arduino for experimental purposes.
Fig. 3 ESP-8266 Wi-Fi transmitter connected to Arduino
An IoT-Based Water Management System for Smart Cities 253
The ESP8285 is an ESP8266 with 1 MiB of built-in flash, allowing for single-chip
devices capable of connecting to Wi-Fi.
4 Blocks of Water Supply Architecture
This section gives a brief overview of the architecture of the smart city and how the
basic blocks of the city are laid out for the water transmission.
4.1 Block Structure of Smart City
The water management system in the smart city is solely based on the fact that the city
is divided into several blocks, and each block consists of several houses to which the
water is supplied. The paper assumes this block structure to ease the explanation. The
entire city is seen as a collection of blocks which are again recursively organized as
sub-blocks. The recursion can go up to a suitable level as would be required for effi-
cient and effective management. Individual houses would be the atomic constituents
of such blocks. As shown in Fig. 4, initially the water is stored in the reservoir, from
where the water for the smart city is obtained. From the reservoir, it is directly sent to
the water treatment plant as it is treated to make it ready for use. Every outlet in the
water supply architecture is monitored using motorized valve for ease of activation.
Fig. 4 Microcontroller setup
254 I. S. Donbosco and U. K. Chakraborty
4.2 Server and Data Control Room
One of the main components of this architecture is the server and data control room
where the data from every sensor in the block is collected and sent to the cloud and
also a main hub for the maintenance of the valve and sensor mechanisms. The sent
data is then used in the cloud to perform different extrapolation and feature modeling
methods to find the water consumption and usage rate and predict the values for the
future use.
4.3 Underwater Storage Facility
Water from the treatment plant is taken to the underwater storage facility [1]. This
facility is present near every block of the city, and the water in this storage unit is
used for immediate consumption on demand [11]. The water from this facility is sent
to different blocks and houses using the sensors and the motor-controlled valve and
smart water pumping systems.
4.4 Water Tanks and Pressure Control Systems
The water from the underwater facility is sent to every house in a block using pumps
and sensors [5]. The specially designed water tanks are used to measure parameters
like current water level and amount of water used to make the water supply more
efficient [13], and conservation and management of the resource become easier. The
structure and organization of the tank are explained in the further section.
5 Structure of the Water Tank
The water tank present in every house is of a specific structure. Firstly, the water tank
is fully customized to suit our requirements of the smart city [14]. It mainly takes
care of three functions:
1. Senses the shortage of water to release water into the tank.
2. Records the amount of water used by the person/house on a daily basis.
3. Uses a rainwater harvesting method to smartly open the valve to capture falling
rainwater.
An IoT-Based Water Management System for Smart Cities 255
5.1 Build of the Water Tank
As mentioned above, the water is stored and supplied to each house through a smart
water tank fitted on top of each house [6]. The smart water tank has several features
as mentioned below.
5.1.1 Microcontroller Setup
The water tank is fitted with the Arduino UNO microcontroller, and it is powered
with a battery which is recharged with a solar panel. The board is kept in a waterproof
container to keep it from moisture which might cause short-circuiting. This micro-
controller will control all the sensors connected in and around the tank. The UNO
is connected with an ESP-8266 Wi-Fi transmitter which is used to connect with the
nearest access point which in turn is connected to the server room to transmit data
[9].
The basic architecture is mentioned in Fig. 5. The second sensor is the SRF-05
ultrasonic sensor. This sensor is used to measure the amount of water in the tank.
The sensor is fitted, face down inside the tank. The sensor emits ultra-sonic waves
and sends it in the tank. The waves are reflected back to the sensor, and it senses
the height h meters. Assuming that the radius of the base of the cylindrical tank is r
meters, the volume of water in the tank volume would be:
vol =πr2h(2)
Fig. 5 Smart city water distribution architecture
256 I. S. Donbosco and U. K. Chakraborty
This measure would help find the amount of water in the tank. If the measured
amount is lesser than the minimum threshold, the automated motorized valve opens
filling the tank with water. This also helps in supplying the house with the required
amount of water. If the tank is having enough water for the home’s usage, the valve
is kept closed until required later. The amount of water used by the house (data) is
collected using the YF-S201 Hall effect sensor. The tank has the outlet into the house
on its base. The Hall effect sensor is placed on the outlet. The amount of water used
up from the tank is calculated using Eq. (2), using the flow rate, average velocity
of the water and the cross-sectional area of the pipe. This gives the essential data to
analyze the data to identify and find out the rate of usage of water with the historic
data.
5.1.2 Rainwater Collection
The water tank is designed in a way not only to make the water distribution efficient
but also to collect rainwater during a shower in order to maximize the water conser-
vation. To facilitate this, a funnel-like structure is placed on top of the tank’s opening
with a door like structure which can be flipped and a water filter on its base. The door
of the funnel is connected to a stepper motor. This stepper motor is used to force
open and close the opening of the funnel. This is placed as keeping the funnel always
open might lead to evaporation of the water from the tank. The top of the tank is
fitted with the LM393 chip-based rain sensor connected to the Arduino board. When
there is rain, the sensor senses it and sends a signal to Arduino. Once it receives the
signal from the sensor, it sends a signal to the stepper motor to open the door of
the funnel to capture the oncoming rainwater and passes through the filter into the
tank for storage. This in turn increases the water ready for use in the tank. Once the
rain subsides, and the rain sensor dries out, the Arduino sends another signal to the
stepper motor to close the door of the funnel.
6 Water Distribution
The water distribution is also fully monitored and controlled with the help of sensors
and fully connected valve systems. The water from the reservoir is sent to the treat-
ment plant from which it reaches the underwater storage. From the underwater
storage, the water is sent to each block in a specific manner. All the monitoring
from when the valve opens, to how much water is sent, everything is controlled from
the server and data control center. The underwater storage facility sends the required
amount of water to separate distribution tanks in each block.
From the analysis of the data received and the predicted amounts of water, the
water is sent through pipelines with motorized valves to each house. The water sent
to each house depends on two things:
An IoT-Based Water Management System for Smart Cities 257
1. The signal from the sensors of the tank.
2. The requirement calculated from the data.
6.1 Signal from the Sensors
As mentioned in section IV, the water tank placed on top of every house is fitted with
different sensors. These sensors keep sending the data to the control center. This data
is used to determine when the water should be supplied and when not. When signals
are received from the sensor that the water level is low, the motorized valve at the
outlet of the pipe is opened and water is sent in. In case the sensor shows that enough
water is present in the tank adequate for the consumption of the household (obtained
from the historic data collected), then the water is not sent to the house. This helps
to divert that water to another area or block where it is actually required. Following
this method will reduce the wastage or unwanted supply of water to the houses in
the city.
6.2 Supply on Demand
The water supply to the houses as mentioned above also depends on another factor,
i.e., supply on demand. This means that the water is supplied only to an extent
which the household requires. This requirement is calculated by the analysis of the
historic data [15]. Extrapolating from usage patterns already recorded YF-S201 Hall
effect sensor, future requirement can be predicted. This would allow regulated water
supply in exact quantity as required by a household and prevent wastage. In case
of excess requirement, special requests can be arranged for individual household
delivery. The obtained value and the actual value are kept and later the error is used
to tune the model better. Storage of such information and excess supply needed by
individual households would help in tuning the predictive models for better analysis
including error margins in prediction. The same information can also be utilized for
taxing individuals for excess use beyond set limits. This application of data analysis
differentiates this method of supply from the regular distribution methods. Regular
tuning of the model can keep the prediction in range and can help optimize the
distribution better. Even seasonal variations in usage patterns can be tracked and
included in the model. Administrative decisions taken at times of water scarcity
caused by lack of rain can also be handled through such models.
258 I. S. Donbosco and U. K. Chakraborty
7 Conclusion and Future Works
The water distribution can be optimized, and the water can be conserved better by
the application of the above method in a real-life scenario. The proposed method-
ology can be improved with the application of better sensors and industry grade
materials which would make it efficient. Motorized valves can be replaced with a
better alternative and the Arduino with the ESP-8266 upgraded with a NodeMCU
or even better ones. This can make the working faster and better. Improvements in
the analytics can be done by using advanced ML algorithms with neural networks
or similar techniques for a better outcome.
In the current paper, as already mentioned, simplifying assumptions have been
made about the layout and plan of the city where it can be implemented. This has
been done purely for ease of identification of the individual households, as analysis
of the data gathered plays a pivotal role in the proposed model. It becomes essential
therefore to pinpoint every holding and regulate the supply accordingly. The scala-
bility of the proposed model in its current form is limited only by the organization of
the cities household. The same may however be implemented on any scale, small or
big, provided a proper distinction can be made between individual houses to enable
the analysis to be based on that identity.
Another novelty of the proposed model is its flexibility. The same model can be
used for farmland water management, wherein the identification of every household
would be replaced by individual farm plots. The analytics can be appropriately tuned
without major changes.
References
1. Mohanasundaram SV, Joyce A, Naresh SK, Gokulakrishnan G, Kale A, Dwarakanath T,
Haribabu P (2018) Smart water distribution network solution for smart cities: Indian scenario.
2018 Global Internet of Things Summit (GIoT). https://doi.org/10.1109/GIOTS.2018.8534524
2. Agarwal N, Agarwal G Role of cloud computing in development of smart city. Int J Sci Technol
Eng (IJSTE), (online) ISSN: 2349-784X
3. Merchanta A, Mohan Kumar MS, Ravindra PN, Vyas P, Manohar U (2013) Analytics driven
water management system for Bangalore city. In: 12th International conference on computing
and control for the water industry, CCWI2013, Procedia Engineering, vol 70, pp 1137–1146
4. Karwot J, Ka´zmierczak J, Wyczolkowski R, Paszkowski W, Przystałka P (2016) Smart water
in smart city: a case study, conference: SGEM 16th international scientific conference on earth
and Geosciencesat: Albena, Bulgaria vol 3/I. https://doi.org/10.5593/sgem2016B31.
5. Kaur MJ, Maheshwari P (2016) Building smart cities applications using IoT and cloud-based
architectures. In: Proceedings of 2016 international conference on industrial informatics and
computer systems (CIICS). https://doi.org/10.1109/GIOTS.2018.8534524.
6. Su K, Li J, Fu H (2011) Smart city and the applications. In: Proceedings of the 2011 inter-
national conference on electronics, communications and control (ICECC). https://doi.org/10.
1109/ICECC.2011.6066743
7. Koo D, Piratla K, John Matthews C (2015) Towards sustainable water supply: Schematic
development of big data collection using internet of things (IoT). Procedia Eng 118:489–497
An IoT-Based Water Management System for Smart Cities 259
8. Karwati K, Kustija J (2018) Prototype of water level control system. In: International sympo-
sium on materials and electrical engineering (ISMEE) 2017, materials science and engineering
384:012032. https://doi.org/10.1088/1757-899X/384/1/012032
9. Priya R, Rameshkumar GP (2017) A novel method to smart city’s water management system
with sensor devices and arduino. Int J Comput Intell Res 13(10):2391–2406 ISSN 0973-1873
10. Varun KS, Kumar KA, Chowdary VR, Raju CSK (2018) Water level management using
ultrasonic sensor (automation). Int J Comput Sci Eng 6(6):799–804. E-ISSN: 2347-2693
11. Paska D (2018) Digitalized water and smart cities—how can telecommunication networks
be used for environmental resilience? International Telecommunication Union, 2018, ICT
Discoveries, Special Issue No. 2
12. https://www.hackster.io/jeffpar0721/add-wifi-to-arduino-uno-663b9e
13. Mulyana Y, Hakim DL (2017) Prototype of water turbidity monitoring system. In: Proceedings
of the international symposium on materials and electrical engineering (ISMEE). https://doi.
org/10.1088/1757-899X/384/1/0120512
14. Karray F,Triki M, Jmal MW, Abid M, Obeid AM (2018) WiRoTip: an IoT-based wireless sensor
network for water pipeline monitoring. Int J Electr Comput Eng (IJECE) 8(5):3250–3258,
ISSN: 2088-8708. https://doi.org/10.11591/ijece.v8i5.pp3250-3258
15. Immanuel SD, Chakraborty (2019) UK Genetic algorithm: An approach on optimization.
In: 2019 International Conference on Communication and Electronics Systems (ICCES),
Coimbatore, India, pp 701–708. https://doi.org/10.1109/ICCES45898.2019.9002372
