Applied Water Science (2018) 8:159

https://doi.org/10.1007/s13201-018-0821-8

ORIGINAL ARTICLE

Design and development of reverse osmosis (RO) plant status

monitoring system for early fault prediction and predictive
maintenance
Satyam Srivastava1 · Saikrishna Vaddadi1 · Pankaj Kumar2 · Shashikant Sadistap1

Received: 29 January 2018 / Accepted: 12 September 2018

© The Author(s) 2018

Abstract
Automation and reliability are the crucial elements of any advance reverse osmosis plant to meet the environmental and
economic demands. Early fault indication, diagnosis and regular maintenance are the key challenges with most of the reverse
osmosis plants in the Indian scenario. The present work introduces a modern reverse osmosis (RO) plant status monitoring
unit to monitor different plant parameters in real time and early prediction for faults and maintenance. Developed RO plant
status monitoring unit consists of a touch screen-based embedded monitoring unit, water quality sensors (pH, TDS), sampling
chamber for controlled water flow, flow sensors, pressure and level sensors. The present system has been developed in a
modular fashion so that it could be integrated with any capacity of RO plant units. Developed embedded system monitors vari-
ous parameters of the plant such as input power, efficiency of the plant, level of input and output water tank and also guides
operator with instructions for plant operation. Other than this, a dedicated smartphone app interface has been developed for
the operator to acquire data from status monitoring unit, storage on smartphone, and transfer it to the cloud. The developed
smartphone-based app also provides facility to integrate plant data with Google map with location information for easy
understanding and quick action. The system has also a backup facility to transfer data to the server using 2G GSM module
during the unavailability of the operator. A dedicated centralized Web server has been developed for real-time visualization
of all installed RO plant status monitoring units. Different machine learning techniques have been implemented on acquired
sensors data to predict early warnings related to power failure, membrane fouling and scaling, input water shortage, pipe,
tank leakage, water quality sensors damage, non-operation or wrong operation of the plant along with different maintenance
actions such as membrane water and chemical wash. Developed RO status monitoring unit has been tested with various RO
plants having capacity from 500 LPH to 2000 LPH and deployed at various nearby villages of Rajasthan.

Keywords  Water quality sensors · Reverse osmosis · Embedded system · Early warning · Predictive maintenance

Introduction

* Satyam Srivastava Increasing demand for fresh potable water leads toward
[email protected] the requirement of different advance automatic water treat-
Saikrishna Vaddadi ment solutions. Various water treatment solutions such as
[email protected] reverse osmosis (RO), microfiltration (MF), ultra-filtration
Pankaj Kumar (UF), nano-filtration (NF) have been explored by differ-
[email protected] ent researchers in the past, but RO is one of the leading
Shashikant Sadistap water treatment solutions due to its higher water purifica-
[email protected] tion capacity, energy efficiency along with the capability
1 to remove almost all toxic metals and bacteria from input
Academy of Scientific and Innovative Research (AcSiR),
CSIR-CEERI, CEERI Colony, Pilani, Jhunjhunu, water supply. The performance of RO membrane is quite
Rajasthan 333031, India sensitive and highly dependent on the quality of feed and
2
CSIR-CEERI, CEERI Colony, Pilani, Jhunjhunu, overall plant conditions. So, RO plant needs highly accurate
Rajasthan 333031, India control and monitoring system to monitor the feed water

13
Vol.:(0123456789)
 159   Page 2 of 10 Applied Water Science (2018) 8:159

and other plant parameters to maintain its operation close to to predict different maintenance actions such as membrane
optimum conditions. Regular maintenance of membrane and water wash, membrane chemical wash, sensors calibra-
monitoring of feed water and plant parameters increase the tion and cleaning, operation of RO plant. Other than this,
productivity of RO plant and also reduce the chances of early a smartphone-based app interface also has been developed
membrane fouling which prolongs the life of the RO mem- to acquire data from RO plant status monitoring unit using
brane and overall plant (Sobana and Rames 2011). Several classical Bluetooth module and further transfer it to Google
control techniques such as PID (Esfahani et al. 2016), fuzzy firebase cloud (Taherkordi and Frank 2016). An alternative
(Zilouchian and Jafar 2001), ANN (Lee et al. 2009), opti- data transfer scheme also has been implemented using 2G
mal control (Gambier et al. 2006), predictive control (Abbas GSM module during the unavailability of RO plant opera-
2006) have been explored to control various RO plant param- tor. Day by day RO plant parameters data also store at the
eters and maintain the plant in optimum conditions. Central- local memory of the embedded system and can be collected
ized PI controllers scheme for interfacing multivariable pro- using any USB drive. The rest of the manuscript is organized
cess has been also tried by different researchers for RO plant as follows. Section II provides a detailed description of the
parameters control using synthesis method (Kumar et al. proposed RO plant status monitoring scheme, and section III
2012). Selection of different techniques to tune the control- provides details of the developed embedded system for RO
ler parameters is one of the important and challenging steps plant status monitoring. Section IV consists of the detailed
and needs to be selected based on the specific application. description of the development procedure of smartphone-
One of the famous techniques to tune controller parameters based android application along with its features and util-
is the usage of compensators. Recycle compensator is one ity. Section V discusses the algorithms to predict different
of the famous techniques to tune the controller parameters maintenance actions and events. Section VI presents testing
(Armbrust and Sbarbaro 2010). Minimization of RO plant and field trial results of developed system installed at differ-
energy consumption using optimal-based control (Bartman ent nearby villages of Rajasthan, and Section VII discusses
et al. 2010) along with alternative energy sources (Chaabene discussion and conclusion.
and Sellami 2013) for RO plants also has been introduced by
different researchers. In Indian scenario, intelligent monitor-
ing along with event prediction for RO plants is one of the Proposed RO plant status monitoring
essential needs to increase the life of RO membrane and scheme
also to avoid different problems such as over-/undervoltage
fluctuations, dry run of high pressure and booster pumps, Overall idea is to monitor the critical parameters of RO
input water pipeline leakage, empty input water tank, over- plants installed at nearby villages of Rajasthan. Proposed
flow of output water tank. Different monitoring control tools monitoring scheme (Fig. 1) consists of different layers such
along with efficient control strategy have been introduced for as at local level, supervisory level and coordinator level and
real-time monitoring and control of RO plants, but most of finally at the central level. Local level monitoring of various
them are costly and bulky and predict only limited param- RO plants can be performed using developed embedded RO
eters such as water quality and flow rate (Torky et al. 2009). plant status monitoring unit and will be useful for different
Few of the efforts also have been seen in the direction of RO plant operators and villagers. Central level of monitor-
real-time membrane fouling monitoring and early prediction ing has been performed using cloud services and could be
using different analytical techniques (Hwang et al. 2010). useful for monitoring and visualization of data at any parts
Several efforts have been observed in the direction of RO of the world.
plant parameters monitoring and control, but still, there are Proposed scheme consists of parameters monitoring, data
various challenges such as early fault prediction of different storage and computational algorithms at different levels
events such as overvoltage and undervoltage fluctuations, and has fewer chances of information loss. Other than this,
membrane fouling, membrane scaling, plant efficiency, water proposed scheme also divides overall information, models
quality sensors damage which need to be addressed. Other and algorithms complexity at different layers. Information
than early fault prediction, predictive maintenance facility collected at the local level using developed embedded RO
related to the membrane, sensors, filters maintenance for plant status monitoring unit transfers to operator’s smart-
Indian RO plant operators is one of the essential needs to phone using Bluetooth communication. Further, collected
be tackled to increase the life of the overall plant and also data with plant location information are transferred to cloud
maintain optimum operating conditions. The present work level for centralized monitoring using the developed custom-
introduces an embedded RO plant status monitoring unit to ized android app. Google firebase has been used as a cloud
monitor various plant parameters such as water quality (pH, platform for the initial use and also to demonstrate the proof
TDS), input and output flow rate, input and output tank level, of concept. 2G GSM module is also used as a backup option
under- and overvoltage fluctuations along with the facility for data communication from monitoring unit to the Web

13
Applied Water Science (2018) 8:159 Page 3 of 10  159

Fig. 1  Implemented RO plant status monitoring scheme for fault prediction and regular maintenance

server. Few of the layers such as supervisory and coordinator app interface to store data and further transfer it to cloud
monitoring layer also can be introduced as the complexity with location information. Developed RO plant status moni-
increases with increase in the number of local RO monitor- toring unit for early fault prediction and predictive main-
ing nodes. tenance is a step toward the development of an advanced,
low-cost and easy to use modular embedded for RO plant
status monitoring. Detailed information about firmware and
Embedded system design and development software package development is as follows.
for RO plant status monitoring
Firmware development
An embedded system (Fig. 3b) has been designed and devel-
oped to acquire real-time data of off-the-shelf sensors inter- RO plant status monitoring unit (Fig. 3a) has been developed
faced with RO plant to measure various plant parameters in modular fashion (Fig. 4) and consists various submodules
such as feed and product flow rate, input power, water level, such as embedded data acquisition, monitoring and control
feed and product pH, TDS. A customized hardware platform unit along with commercial sensors with 4–20-mA current
(Fig. 2) has been designed to read and store data from vari- transmitters for pH and TDS sensors installed in sampling
ous sensors and also to drive different actuators used in RO chamber for real-time feed and product water quality data
plant. Developed hardware platform has been interfaced with collection, commercial uninterrupted power supply module
quad-core ARMv8 processor-based microcontroller unit for (U.P.S) for power backup. The developed portable embed-
data acquisition and processing. Developed touch screen- ded system acts as a heart of the overall RO plant status
based embedded system provides the facility for a user- monitoring unit. Embedded system consists three major
friendly graphical user interface for RO plant operator to sections such as dedicated designed hardware platform for
monitor real-time plant data. Further, a classical Bluetooth sensors and actuator interfacing, an ARMv8 processor-
module has been integrated with the developed embedded based microcontroller unit (Raspberry pi 3) for data acqui-
system to communicate with dedicated smartphone-based sition and processing and switch mode power supply for

13
 159   Page 4 of 10 Applied Water Science (2018) 8:159

Opto-
Frequency Level Port Darlington
Coupler
Measurement Translator Expender Transistor
based
Unit Unit Unit Array
Isolation

Flow Sensor
HDMI Touch Touch Analog to
Flow Switch Screen with Storage I2C/SPI Digital
Screen
GUI Device Ports Converter
Driver (ADC)
Level Sensor

Pressure
Sensor ARMv8 based Wireless
Multi Port Comm.
USB Hub
Microcontroller Current to
pH Sensor 2G GSM Port Voltage
Module Unit Converter
TDS Sensor

Serial Port Ethernet GPIO

DC Solenoid (RS-232) Port Ports Current
Valves Bluetooth Transmitter
Module Unit for pH
& TDS
Sensors
Regulated Phase Failure
Power Supply Monitoring
Unit Unit

Embedded Monitoring & Control Unit

Fig. 2  Functional block diagram of developed embedded RO plant status monitoring unit

Fig. 3  a Deployed RO plant status monitoring unit and b developed embedded processing and control unit

regulated power. Developed dedicated hardware module resources used to develop dedicated hardware modules are
design composed of embedded circuitry, wherein PCB mod- Darlington array (ULN2003) to drive relay array, external
ules are indigenously designed, fabricated and interfaced ­E2PROM (24LC1025) for data storage at local level, port
with ARMv8 processor-based microcontroller unit. The expander (MCP23017) to generate additional digital input

13
Applied Water Science (2018) 8:159 Page 5 of 10  159

Level Sensor (L1)

High
Flow Switch
Valve-1 (f1)
Mid Booster Pump 45.0 PSI

Low
Pressure
Flow Sensor (F1) Sensor
Raw Water Tank
(P1)

Valves Sand Filter

Control Valve-2

Water Drain Valve-5

Sensor Sensor
Data Data

10 Micron Filter

Carbon Filter

5 Micron Filter
Needle Valve
300.0 Pressure High
PSI Sensor Pressure
Reject Water Drain (P3) Pump
Level Sensor (L2)

High
Valve-4
Mid
RO Membrane
20.0 PSI
Low

Product Water Tank Pressure

Sensor Flow Sensor (F2)
(P2)

Fig. 4  Line diagram of reverse osmosis plant status monitoring unit integrated with reverse osmosis plant

output pins, analog-to-digital converter (ADS7828) to con- transfer and regulators (LM7805, LM1117) of different volt-
vert sensors analog response into digital form, buffer unit age levels (+ 5 v and + 3.3 v). Developed hardware platform
for isolation (74LS244), real-time clock (DS1307) to access has been further interfaced with an ARMv8 processor-based
the real time with data, Xbee pro module for wireless data microcontroller unit (Raspberry pi 3) for data acquisition,

13
 159   Page 6 of 10 Applied Water Science (2018) 8:159

processing and control. A 7-inch touch display along with in the local language. All developed modular programs are
its local controller has been integrated with the developed linked to the main menu graphical user interface-based pro-
embedded system to display the user-friendly graphical user gram to operate RO plant status monitoring unit for fault
interface. prediction and predictive maintenance. Raspbian Jessie OS
A classical Bluetooth module has been used to transfer has been ported raspberry pi-based embedded system and
data from developed embedded system to smartphone-based further developed software package transferred on raspberry
app interface and further to cloud for future analysis. Other pi 3-based embedded system.
than this, a 2G GSM module has been used as an alterna-
tive data transfer to the server. Developed embedded system
power is provided by a switched mode power supply (SMPS) Smartphone‑based android app and Web
that provides cleaned and filtered ± 12 V to all circuit mod- page‑based data visualization and analytics
ules. Developed embedded processing and control unit has interface
been interfaced with various commercial sensors, which are
already integrated with RO plant for measurement of differ- Data communication from embedded RO plant status moni-
ent parameters. toring unit to cloud or Web server is a challenging and dif-
ficult task in different Indian remote locations. One of the
Software package development major challenges faced during the installation of RO plant
status monitoring units at different villages is the availabil-
Various modular programs have been written in C+++ with ity of network for data transfer. Initially, 2G GSM module
GCC compiler for the functioning of developed embedded has been used to communicate data from RO plant status
RO plant status monitoring unit. Open-source Qtopia plat- monitoring system to the server, but various problems have
form has been used for graphical user interface development been faced related to network availability. The strength of
for data visualization and user interaction. Developed soft- different firm’s network changes from village to village. A
ware package provides the number of facilities for hassle- new innovative approach of data transfer from embedded
free operation and regular maintenance. Few of the major unit to the cloud has been implemented to tackle above-
features are on-site set point window for location-specific mentioned problems. Proposed approach consists of two-
thresholds of pH, TDS, flow rate, real-time data visualiza- step data transfer: First is from embedded RO monitoring
tion of various RO plant parameters, multiple data transfer unit to the smartphone-based android app, and second is
options and RO plant operation steps in the local language. from smartphone to cloud using operators smartphone. A
The main software module initializes the various water qual- dedicated smartphone-based app interface has been devel-
ity sensors, flow sensors, level sensors, pressure sensors, oped to transfer data from RO plant status monitoring unit
analog-to-digital converter, E2ROM, Bluetooth module, to smartphone and further to cloud. Open-source android
serial port, GSM module, port expander module, phase fail- studio platform has been used for smartphone-based app
ure monitoring unit and also diagnosis of overall embedded interface development. Developed smartphone-based app
unit. Microcontroller unit communicates with developed interface provides a front end view to display different RO
dedicated hardware unit using the I2C protocol to drive plant parameters such as pH, TDS, feed and product water
various actuators and also to acquire information from vari- flow, input power status, input and output tank level, input
ous sensors. Developed software package has menu-driven pipeline status, plant modes running status and the overall
option with four different modes to operate RO plant status status of the plant. The developed app also provides options
monitoring unit: First is data visualization mode, second is to transfer per minute RO plant data along with location
location-specific set point mode, third is data transfer mode, information to Google firebase cloud. Further, a dedicated
and the last one is training mode. Data visualization mode Web page has been developed to access the cloud for real-
provides facility to access various RO plant parameters in time data visualization. The developed Web page provides
real time and also to drive plant in different modes such as real-time information of all installed RO plant process vari-
process mode, rinse mode, chemical wash mode and water ables such as feed and product pH, TDS, flow rate, pressure
wash mode. Location-specific set point mode has options to along with various operation modes (process mode, rinse
set upper and lower thresholds for critical sensors response mode, chemical wash mode and water wash mode) duration
such as pH, TDS and flow. System triggers alarm for RO information. The user can easily access real-time data or sta-
plant operator as real-time sensors response crosses set tus of any RO plant by just clicking on its location on Google
points. Data transfer mode provides real-time data trans- map. Android app also has the ability to save per day data
fer options to cloud and server using 2G GSM module or on smartphone memory in Excel file format. It also provides
operator’s smartphone-based app interface while the train- facility to take a picture RO plant location and transfer to
ing mode provides stepwise points for RO plant operation cloud as a backup of the location information. Android app

13
Applied Water Science (2018) 8:159 Page 7 of 10  159

also integrates all RO plants data with location information osmosis (Ruiz-García et al. 2017; Yiantsios et al. 2005), but
to Google map for easy understanding, quick judgment and no significant effort has been observed in the direction of
action. Overall data transfer scheme (Fig. 5) from RO plant an portable embedded system development for early fault
status monitoring unit to user interface is as follows. prediction and predictive maintenance for reverse osmosis
plant. Developed RO plant status monitoring unit has been
integrated with 2000-LPH reverse osmosis plant installed in
Algorithms for early fault prediction institute campus and continuously operated six hours daily
and predictive maintenance for 10 days. Real-time data from various sensors have been
acquired episodically using developed RO plant status moni-
Reverse osmosis (RO) membrane fouling and scaling are one toring unit. Real-time flow rate, pressure and water quality
of the main challenges for the membrane manufactures, the sensors data of feed and product water have been acquired
scientific community and industry professionals. Membrane for 7 days continuously. Water wash process has been per-
fouling and scaling phenomenon have a negative effect on formed to clean the RO membrane after each day data acqui-
the RO desalination system. Predicting fouling and scaling sition process. It has been observed that feed water param-
in RO systems are the key challenges and also important to eters are almost constant while product water parameters
evaluate the long-term performance and cost. As membrane also change very slowly over the time period. The sampling
is one of the important and costly elements of reverse osmo- rate for data acquisition of sensors integrated with RO plant
sis plant, continuously chemical wash and water are required has optimized based on the study and observation and fixed
to increase the life of the membrane. Regular chemical and at one hour (Fig. 6). Real-time pressure sensors (P1, P2 and
water wash decreases the chances of early membrane foul- P3 in Fig. 4) and flow sensors (F1, F2 in Fig. 4) responses
ing and scaling. Researchers have presented various tech- have been used to measure the overall flow rate of feed and
niques and methods for membrane fouling prediction for product water. pH and TDS sensors responses have been
different water treatment methods such microfiltration (Liu used to judge the quality of feed and product water. Over-
et al. 2009; Davis 1992; Duclos-Orsello et al. 2006), reverse all acquired data have been used to design the models to

Fig. 5  Overall communication scheme for developed embedded RO plant status monitoring unit

13
 159   Page 8 of 10 Applied Water Science (2018) 8:159

Fig. 6  Flow rate and water quality variation in feed and product water

predict membrane fouling and scaling events, and based on data, and it has been observed that efficiency of the reverse
this information system generates an early warning and also osmosis plant decreases over the time period of operation.
predicts various predictive actions such as chemical wash, It also has been observed that efficiency of the reverse
water wash, membrane replacement. Membrane efficiency osmosis plant depends over the membrane chemical wash,
analysis (Fig. 7) has been carried out using real-time sensors water wash and backwash frequencies. Efficiency analysis

70.00 67.66

64.10
65.00 62.82
Reverse Osmosis Plant Efficiency (in %)

61.84

60.00 57.89 57.89 58.14

57.26
56.00 55.41 Membrane scaling
55.00
event critical point
(need chemical
wash)
50.00
45.76
45.00

40.00 38.50

34.81
35.00

28.69
30.00
25.59
25.00
0 2 4 6 8 10 12 14 16
Time Period (in days)

Fig. 7  Efficiency analysis of reverse osmosis membrane over the time period of 15 days

13
Applied Water Science (2018) 8:159 Page 9 of 10  159

can clearly indicate the membrane scaling event, as mem- regulation have been implemented in MATLAB 2015a to
brane efficiency suddenly goes down monotonically up to achieve the best results. Further, performance of both train-
25 percent. Artificial neural network (ANN)-based tech- ing algorithms (Table 1) has been compared on the basis of
niques have been implemented in MATLAB 2015a to pre- mean square error, number of neurons and accuracy, and the
dict membrane fouling and scaling event. Multi-input- and final weight and bias values have been updated on RO plant
single output-based architecture has been used to predict status monitoring unit to predict plant efficiency, and based
membrane fouling and scaling event. Five plant parameters on the variation in membrane efficiency, system generates
such as feed and product water flow rate, product water a silent alarm indicator to indicate the membrane fouling
quality parameters (pH, TDS) and water wash frequency and scaling event.
along with plant operation time have been used as input in Total samples dataset (75 samples) has been divided into
input layer and predict single output membrane efficiency in three different datasets such as training, testing and valida-
output layer. The optimal number of neurons in hidden layer tion. Training dataset consists of 70% of the dataset while
has been achieved using trial-and-error method. Hidden testing and validation dataset consists of 15% each. Training
layer neurons have been varied from 1 to 7, and obtained process has been stopped as mean square error stops improv-
results have been compared in the context of complexity, ing. Mean square with the variation in the number of neu-
computation time and accuracy. Two mostly explored train- rons in the hidden layer has been analyzed for both training
ing algorithms such as Levenberg–Marquardt and Bayesian algorithms (Fig. 8). Out of both training techniques, Bayes-
ian regulation training algorithm performs the best with the
lowest mean square error (1.2814 × 10−10), less number of
Table 1  Performance comparison of two different training techniques neurons in hidden layer (5) and with the highest accuracy
used of membrane scaling and fouling prediction (R2 = 99.99%).
Training algorithm Number of Testing R2 MSE (testing)
neurons
Testing and field trial results
Bayesian regulation 1 99.97 0.129687
Bayesian regulation 3 99.99 8.8855 × 10−8
Developed embedded RO plant status monitoring unit has
Bayesian regulation 5 99.99 1.2814 × 10−10
been tested with RO plant installed at institute campus hav-
Bayesian regulation 7 99.99 4.0055 × 10−9
ing a capacity of 2000 LPH. Tested RO plant status moni-
Levenberg–Marquardt 3 99.952 0.243723
toring unit has been installed along with various preexisting
Levenberg–Marquardt 4 100 1.1371 × 10−5
RO plants having a capacity of 500–2000 LPH at ten differ-
Levenberg–Marquardt 6 99.998 0.028779
ent nearby villages of Rajasthan such as Tamkor, Googne
Levenberg–Marquardt 8 100 0.00025786
ki Dhani, Raboodi, Mukha ka bas, Jhunjhunu to monitor

0.3

0.243723
0.25

0.2
Mean Square Error

0.15 0.129687

0.1

0.05 8.88553E-08 0.028779

4.00552E-09
1.13712E-05 1.28141E-10 0.000257861
0
0 1 2 3 4 5 6 7 8

-0.05
Number of Neurons in Hidden Layer

Levenberg Marquardt Bayesian Regulerization

Fig. 8  Mean square error variation with the number of neurons in hidden layer

13
 159   Page 10 of 10 Applied Water Science (2018) 8:159

the real-time status of RO plants. It has been observed that credit to the original author(s) and the source, provide a link to the
developed RO plant status monitoring unit provides real- Creative Commons license, and indicate if changes were made.
time data of RO plant parameters and it is also able to pre-
dict various events related to membrane behavior such as
membrane scaling and fouling. System also predicts various References
maintenances regarding actions such as membrane chemi-
Abbas A (2006) Model predictive control of a reverse osmosis desali-
cal wash, water wash, replacement of membrane, wrong nation unit. Desalination 194(1-3):268–280
operation of RO plant, emergency power shutdown. The Armbrust N, Sbarbaro D (2010) On the robust tuning of controllers
developed system also has autofault diagnosis capability to with recycle compensators. IFAC Proc Vol 43(21):211–217
detect and report internal errors related to sensors and circuit Bartman AR et al (2010) Minimizing energy consumption in reverse
osmosis membrane desalination using optimization-based control.
damage. J Process Control 20(10):1261–1269
Chaabene AB, Sellami A (2013) A novel control of a Reverse Osmosis
desalination system powered by photovoltaic generator. In: 2013
Discussion and conclusion International conference on electrical engineering and software
applications (ICEESA), IEEE
Davis RH (1992) Modeling of fouling of crossflow microfiltration
This paper reported a portable reverse osmosis status membranes. Sep Purif Methods 21(2):75–126
monitoring unit for early fault diagnosis and predictive Duclos-Orsello C, Li W, Ho C-C (2006) A three mechanism model
maintenance. Developed systems monitor various plant to describe fouling of microfiltration membranes. J Membr Sci
280(1-2):856–866
parameters such as feed and product water flow rate, pres- Gambier A, Andrea W, Essameddin B (2006) Optimal control of a
sure, pH, TDS, over- and undersupply voltage, input and reverse osmosis desalination plant using multi-objective optimiza-
output water tank level. Developed system provides early tion. Computer aided control system design, 2006 IEEE interna-
fault detection related to membrane fouling, scaling, sensors tional conference on control applications, 2006 IEEE international
symposium on intelligent control, 2006 IEEE
damage, dry running of pumps, a wrong operation of plant Hwang T-M et al (2010) Prediction of membrane fouling rate by neural
and many more. The system also suggests various predictive network modeling. Desalin Water Treat 15(1–3):134–140
actions to plant operator at regular interval to avoid faults. Janghorban Esfahani I et  al (2016) Control performance evalua-
A smartphone-based app interface with various advanced tion of reverse osmosis desalination system based on model
predictive control and PID controllers. Desalin Water Treat
features such data collection from embedded RO plant sta- 57(55):26692–26699
tus monitoring unit further transfers it to cloud. Developed Kumar VV, Rao VSR, Chidambaram M (2012) Centralized PI control-
app interface also provides facility to integrate real-time lers for interacting multivariable processes by synthesis method.
plant data to the Google map with location information for ISA Trans 51(3):400–409
Lee YG et al (2009) Artificial neural network model for optimizing
quick understanding and rapid maintenance action. Differ- operation of a seawater reverse osmosis desalination plant. Desali-
ent ANN-based algorithms have been explored to predict nation 247(1–3):180–189
membrane efficiency using various reverse osmosis plant Liu Q-F, Kim S-H, Lee S (2009) Prediction of microfiltration mem-
parameters as input. The present idea to monitor the status of brane fouling using artificial neural network models. Sep Purif
Technol 70(1):96–102
reverse osmosis plant and early predict various faults avoids Ruiz-García A, Melián-Martel N, Nuez I (2017) Short review on pre-
the situation of non-operation of RO plant and also increases dicting fouling in RO desalination. Membranes 7(4):62
the life of the overall plant. It also guides plant operator Sobana S, Rames CP (2011) Identification, modelling, and control of
regarding various operation steps, faults and maintenance continuous reverse osmosis desalination system: a review. Sep Sci
Technol 46(4):551–560
actions at regular interval. The overall developed system is Taherkordi A, Frank E (2016) Scalable modeling of cloud-based iot
cost-effective, portable and easy to use for villagers or any services for smart cities. Pervasive computing and communication
unskilled person. workshops (PerCom Workshops), IEEE
Torky OM, Irraivan E, Hanif NHHBM (2009) PC based SCADA sys-
Acknowledgements  Authors wanted to thank Director CEERI for tem for reverse osmosis desalination plants. In: 2009 IEEE student
allowing us to develop such technology and also to publish this work. conference on research and development (SCOReD), IEEE
Authors also wanted to extend their gratitude to Department of Sci- Yiantsios SG, Sioutopoulos D, Karabelas AJ (2005) Colloidal foul-
ence and Technology (DST), New Delhi, for funding such innovative ing of RO membranes: an overview of key issues and efforts
project. Authors wanted to thank group members for providing their to develop improved prediction techniques. Desalination
support in system testing and installation. 183(1–3):257–272
Zilouchian A, Jafar M (2001) Automation and process control of
reverse osmosis plants using soft computing methodologies.
Open Access  This article is distributed under the terms of the Crea- Desalination 135(1-3):51–59
tive Commons Attribution 4.0 International License (http://creat​iveco​
mmons​.org/licen​ses/by/4.0/), which permits unrestricted use, distribu-
Publisher’s Note Springer Nature remains neutral with regard to
tion, and reproduction in any medium, provided you give appropriate
jurisdictional claims in published maps and institutional affiliations

13