IoT based milk monitoring system for detection of milk adulteration

Abstract

Sustenance security in provincial and urban zones is an extremely huge, as it nearly influences the soundness of nationals. Late examinations detect that crude milk contains pathogenic life forms which could bring about contamination if devoured which can build the rate of infections and break down the personal satisfaction. Thus, creating apparatuses for constant and shrewd detecting is required for quality checking and to settle on reasonable and opportune choice. The work aimed to present some aspects regarding milk quality and quantity estimation. The Internet of Things (IoT) based system allows users to know the groupings of gases in crude milk continuously. As the milk is kept for several days, the expansion of bacterium will get increased which ends up in undesirable smell, style and harmful substances. Hence there is a necessity for monitoring system to discover and determine the spoilage of milk and turn out into a healthy product. Consequently, the toxic substances in milk are identified before to maintain a strategic distance from entanglements in the underlying stage for a decent last item. In this proposed system, Microbial activity is determined using gas sensor, high quality milk should have no salinity, so salinity of the milk is measured by using a salinity sensor and also level of the milk will be measured by using a level sensor. In addition to that customer should have their own card for accessing the milk diaries.

Full text

IoT BASED MILK MONITORING SYSTEM FOR
DETECTION OF MILK ADULTERATION
Dr. G. Rajakumar1, Dr. T.Ananth Kumar2, Dr. T.S. Arun Samuel3, Dr. E.Muthu
Kumaran4
1Dept. of Electronics and Communication Engineering, Francis Xavier
Engineering College, Tirunelveli, India.
2Dept. of Electronics and Communication Engineering, IFET College of
Engineering, India.
3Dept. of Electronics and Communication Engineering, National Engineering
College, Kovilpatti, India
4Dr.B.R.Ambedkar Institute of Technology, Port Blair, India.
Abstract.
Sustenance security in provincial and urban zones is an extremely huge, as it nearly
influences the soundness of nationals. Late examinations detect that crude milk contains pathogenic
life forms which could bring about contamination if devoured which can build the rate of infections
and break down the personal satisfaction. Thus, creating apparatuses for constant and shrewd
detecting is required for quality checking and to settle on reasonable and opportune choice. The work
aimed to present some aspects regarding milk quality and quantity estimation. The Internet of Things
(IoT) based system allows users to know the groupings of gases in crude milk continuously. As the
milk is kept for several days, the expansion of bacterium will get increased which ends up in
undesirable smell, style and harmful substances. Hence there is a necessity for monitoring system to
discover and determine the spoilage of milk and turn out into a healthy product. Consequently, the
toxic substances in milk are identified before to maintain a strategic distance from entanglements in
the underlying stage for a decent last item. In this proposed system, Microbial activity is determined
using gas sensor, high quality milk should have no salinity, so salinity of the milk is measured by
using a salinity sensor and also level of the milk will be measured by using a level sensor. In addition
to that customer should have their own card for accessing the milk diaries.
Keywords:Internet of Things (IoT) based system, Arduino, EM-18 module.
1. INTRODUCTION
The Internet of things (IoT) describes network of material devices, means of transport, and
erstwhile equipment surrounded with electronics, software, sensors, actuators, and system
connectivity which allow these items to gather and swap over data. The IoT permit things
to be intelligence or forced distantly diagonally active system communications, make
chance for supplementary straight combination of the material earth into computer-based
organization, and resultant in enhanced effectiveness, correctness and financial advantage
in calculation to compressed human being interference. Milk is a perishable product.
Consequently, it is typically handled locally inside a couple of hours of being gathered [1].
In the United States, there are a few hundred thousand dairy ranches and a few thousand
milk preparing plants. Dairy cows are milked twice a day using mechanical vacuum
milking machines. The raw milk flows through stainless steel or glass pipes to a
refrigerated bulk milk tank where it is cooled to about 40° F (4.4° C).
International Journal of Pure and Applied Mathematics
Volume 118 No. 9 2018, 21-32
ISSN: 1311-8080 (printed version); ISSN: 1314-3395 (on-line version)
url: http://www.ijpam.eu
Special Issue ijpam.eu
21

2
It’s more harmful for child and pregnant women [4]. To avoid this there is a need of real
time monitoring system which always keeps an eye on the quality of the milk [5]. It is
serious concern and steps have been taken towards it, and many existing systems are there
which only monitors the microbial activity. By this system will be able to monitor
maximum aspects of milk like microbial activity, adulteration, quality etc. Accordingly,
this work helps in early identification of dangerous substances in milk to maintain a
strategic distance from entanglements in the underlying stage for a decent last item [6].
2. LITERATURE REVIEW
Lucas de Souza Ribeiro et. al. [2] states that using a cryoscope, detection of water
adulteration in milk can be performed. The GaAsSb sensors, which show quick reaction
and great affectability to the NIR range, were utilized to distinguish diffusely reflected
light. The proposed instrument was tried on milk tests corrupted with water. The outcomes
displayed high coefficients of assurance, higher than 0.99. In this manner, the created
framework might be utilized for identification of milk debasement. Carla Margarida
Duarte et. al.[9] developed a attractive counter that identifies the nearness of Streptococcus
agalactiae (a Group B Streptococci) in crude milk. This gadget permits the investigation of
crude milk without crossing over the microfluidic channels, making this incorporated stage
exceptionally appealing for quick bacteriological pollution screening. Wesley Becari et.al.
[7] developed a methodology for the detection of bovine milk adulteration by applying
electrical impedance measurements. The classification of the results is proposed through
ak- nearest neighbors algorithm that allows to quantitatively qualify the samples of pure
and adulterated milk. Pallavi Gupta et. al [5] displayed another framework, which is
utilized for the location and estimation of corruption of clarified butterfat, a classification
of anhydrous milk fat. Identification of defilement by at least 20% of creature muscle
versus fat's in clarified margarine is effectively and monetarily done. Dari de O. Toginho
Filho and Vanerli Beloti [3] proposed a model of a computerized photometer,
microcontrolled, versatile gadget, which utilizes three LEDs with discharge in the NIR
area and was created without the utilization of focal points, filters or moving parts. The
outcomes demonstrate that the model reaction resembles the one of a business cryoscope,
yet quicker.
3. PROPOSED SYSTEM
In this IoT system, we aimed to present some aspects regarding milk quality and quantity
estimation. So, in this proposed system each customer should have their own card for
accessing the milk diaries. High quality milk must have no salinity, so salinity of the milk
is measured by using a salinity sensor for detecting adulteration of milk and level of the
milk will be measured by using a level sensor for measuring the quantity of the milk.
When the milk is stored for long, the microbial activity gets started which gives the milk a
foul smell which can be detected using a gas sensor. In the existing system only the gas
sensors are used for detection of early microbial activity which makes it useless when it
comes to detection of adulteration of milk, our proposed system detects both aspects,
adulteration as well as early microbial activity in milk.
3.1. Working Principle
Here, IoT based Arduino Microcontroller is used which can drive by 5V DC supply; the
quality of the milk is maintained by using the smart sensors the temperature sensor helps in
monitoring the temperature of the milk. The viscosity sensor measures the viscosity of the
International Journal of Pure and Applied Mathematics Special Issue
22

3
milk, the gas sensor used to detect the odour of the milk, the milk level sensor is used to
measure quantity of milk, and the salinity sensor detects the salinity of the milk. The RFID
reader reads the RFID card which consists of the customer details and the payment details,
if the predefined customer makes the successful payment the motor will be switched on
denoting the milk is filling. If there is an unknown entry or the insufficient payment the
buzzer will be blown, all these statuses will be shown figure 1 in LCD.
Figure 1 Proposed IoT based System
3.2. Arduino Uno
The Arduino Uno is a microcontroller board based on the IoT ATmega328. It has 14
digital input/output pins (of which 6 can be used as PWM outputs), 6 analog inputs, a 16
Arduino
Milk Level
Sensor
Gas Sensor
Temperature
Sensor
Viscosity
Sensor
Salinity
Sensor
Buzzer
Motor
LCD
RFID Reader
Power Supply
RFID Tag
International Journal of Pure and Applied Mathematics Special Issue
23

4
MHz ceramic resonator, a USB connection, a power jack, an ICSP header, and a reset
button as shown in figure
2.
Figure 2 Arduino Board
Specification
• Microcontroller: IoT Enable ATmega328
• Operating Voltage: 5V
• Input Voltage (recommended):7-12V
• Input Voltage (limits): 620V
• Digital I/O Pins: 14 (of which 6 provide PWM output)
• Analog Input Pins: 6
• DC Current per I/O Pin: 40 mA
• DC Current for 3.3V Pin: 50 mA
• Flash Memory: 32 KB (ATmega328)
• SRAM: 2 KB (ATmega328)
• EEPROM: 1 KB (ATmega328)
• Clock Speed: 16 MHz Power
3.3. Power Supply
The AC voltage, commonly 220V, is associated with a transformer, which steps that air
conditioner voltage down to the level of the coveted dc yield. A diode rectifier at that point
gives a full-wave redressed voltage that is at first sifted by a basic capacitor channel to
deliver a dc voltage. This subsequent dc voltage generally has some swell or air
conditioning voltage variety.
3.4. LM35 Temperature Sensor
The voltage yield of a LM335 increments by roughly 10 mV for each 1 degree Kelvin of
ascend in temperature. Note that 1 degree Kelvin is equal to 1 degree Celsius. In the
circuit, the output of the LM335 is fed into a 741 op-amp (any standard op-amp may be
used) which is configured as a voltage follower. The primary capacity of the operation
International Journal of Pure and Applied Mathematics Special Issue
24

5
amp, consequently, is simply to buffer the LM335 yield with the goal that it isn't
influenced by whatever heap is associated with this temperature sensor circuit as appeared
in figure 3.
Figure 3 LM-35 Temperature sensors
3.5. Level Sensors
Level sensors sense the level of fluid and other liquid and fluidized solids, that show a
higher free surface. Material that flow becomes flat in their containers since of gravity
while mainly mass solids mound at an angle of repose to a peak as shown in figure 4.
Figure 4 Level Sensor
3.6. Salinity Sensor
Measure water with a wide assortment of salinities, from salty water to sea water, and even
hyper-saline conditions. You can likewise think about how saltiness influences lightness or
screen saltiness esteems in estuaries where new water blends with sea water as appeared in
figure 5.
International Journal of Pure and Applied Mathematics Special Issue
25

6
Figure 5 Salinity Sensor
3.7. Viscosity Sensor
The deflector diverts flow into the sensor to continually renew the sample in the
measurement chamber. A built-in temperature detector (RTD) senses the actual
temperature in the measurement chamber as shown in figure 6.
Figure 6 Viscosity Sensor
3.8. MQ6 Gas Sensor
Figure 7 MQ6 Gas Sensor
This sort of gas sensors is produced using tin dioxide (sno2) semiconductor which delivers
a low conductivity in clean air. TGS 813 sensor is much touchy in nature to propane,
International Journal of Pure and Applied Mathematics Special Issue
26

7
Methane and furthermore butane. This kind of sensor is appropriate for checking LPG
Gases and furthermore it reasonable in recognizing extensive variety of gases for
mechanical applications. The huge preferred standpoint of this sensor is it is of minimal
effort. MQ6 gas sensor instantly finds the focus in air where its basic electrical circuit
changes over the conductivity charge in to a yield flag and its gas fixation relates thusly.
This is a Figaro gas sensor which has images and sticks in the external body of the external
circuit of sensor which is associated with yield over the heap resistor which increments
quickly as protection Rs diminishes where it relies upon centralization of gas as appeared
in figure 7.
3.9. RFID
Information inside a tag may give ID to a thing in fabricate, merchandise in travel, an area,
the personality of a vehicle, a creature or person. For instance, the shade of paint for an
auto body entering paint splash zone on the creation line, the set-up directions for an
adaptable assembling cell or the show to go with a shipment of merchandise as appeared in
figure 8.
Figure 8 RFID
4. RESULT & DISCUSSION
This section shows the implementation results, which finally shows the different aspects of
the milk. The whole process is controlled by an Arduino board. The analog data is sent to
the Arduino using different sensory system. It is used to detect milk adulteration and early
microbial activity by continuously monitoring the milk using various techniques.
International Journal of Pure and Applied Mathematics Special Issue
27

8
Figure 9 Experimental Setup
In this paper, figure 9 intended to detect adulteration and microbial activity in milk. As the
system is switched on, all the sensory system gets active and starts sensing for any
abnormalities in the milk one after another.
If there is any abnormality found by the sensor, it informs the Arduino which
automatically stops the process. In addition to this RFID is used for milk booths as each
user can have their own card for accessing the milk booths. Thus, providing good quality
milk to the people using sensors like MQ-6 gas sensor, salinity sensor and viscosity sensor
etc. The system is intended to detect adulteration and microbial activity in milk so the
system must be fast as well as accurate. With the help of multiple sensors abnormalities in
milk is detected, the detection must be continuous as well as automated. Thus, in this work
the various sensors keep on monitoring the milk and transmit data to Arduino controller.
The Arduino controller carries out the process of filling of milk as well as showing
continuous result in the LCD in figure 10.
Figure 11. Power Analysis of Proposed System Component
Figure11 represents the Power Analysis of Proposed System Component which
involves the power consumption of Milk Level Sensor, Gas Sensor, Temperature Sensor,
Viscosity Sensor and Salinity Sensor. Figure 11 shows the design of proposed system with
optimized power, because sensors are the major role play in the proposed system which is
consuming the maximum power in the system. Hence, we should know about each sensor
power consumption before going to design the proposed system.
F
i
g
u
r
e
1
0.033
0.099
0.095
0.081
0.081
0
0.02
0.04
0.06
0.08
0.1
0.12
POWER(W)
Power(W)
12
12
7
5
3
0 2 4 6 8 10 12 14
8051 Microcontroller based system
8086 Microprocessor based system
ARM based System
FPGA based System
Proposed Arduino based System
Power(W)
International Journal of Pure and Applied Mathematics Special Issue
28

9
2. Power Analysis of different system
Figure 12 displays the Power Analysis of different system which is helpful to detect the
adulteration level in the milk and power consumption. Figure 12 represents the battery
backup level indication of existing and proposed system, which is helpful for battery
backup requirement of each system. Finally, how much amount of power utilizes by
proposed will be inferred from the figure 12. Hence the proposed system power
consumption is 3 watts.
Figure 13 Adulteration Detection Speed in different platforms
Figure 13 represents the Adulteration Detection Speed with different platform which is
helpful to show the fastest to detect the adulteration.
F
i
g
u
r
e
1
4
W
e
i
ght analysis of different system
Here, the adulteration detection speed depends on the computation time of proposed
system also how fastest way to detect the adulteration in order to avoid latency. Finally,
Internet of Things (IoT) based proposed i.e Arduino Internet of Things (IoT) based
12
8
5
2
0
2
4
6
8
10
12
14
C++ Software
Java Software
FPGA
Implementation
based System
Proposed
Embedded based
System
Adulteration Detection Speed (ns)
0.9
0.9
0.7
0.6
0.4
Weight (Kg)
8051 Microcontroller based system
8086 Microprocessor based system
ARM based System
FPGA based System
International Journal of Pure and Applied Mathematics Special Issue
29

10
proposed is better performance than the C++, Java and FPGA based system. Hence the
proposed system adulteration detection speed only 2ns.
Figure 14 shows the weight analysis of different system which is representing the weight
of each system. Here, the proposed system weight represents, which is helpful locate the
system at which place and execution purpose. Because we can relocate the system
according to their weight. Hence the proposed system carrying only 0.4 Kg weight.
5. CONCLUSION AND FUTURE ENHANCEMENT
In this paper, we developed a IoT based system which gives faster and more accurate
results. In our proposed system, Microbial activity is determined using gas sensor. The
high-quality milk should have no salinity, so salinity of the milk is measured by using a
salinity sensor and level of the milk will be measured by using a level sensor. In addition
to that customer should have their own card for accessing the milk diaries. The milk
collection parameters such as weight, FAT & CLR are measured by this system gives fast
and more accurate results than the existing systems which are more costly than the
developed one. As a future work, it is intended to implement IOT and DBMS for billing
system, in which each user will have a data base of their own, in which the data is recorded
for the amount of milk taken, in this payment can be done using debit or credit cards,
payment can be done on monthly basis. Further this system will be used by the
management for tracking the milk production and marketing, all the information from milk
production to marketing will be stored in management’s website which can be accessed by
any user having account in that firm.
6. REFERENCES
[1] Pedinti Sankaran Venkateswaran; Abhishek Sharma; Santosh Dubey; Ajay
Agarwal; Sanket Goel “Rapid and Automated Measurement of Milk Adulteration
Using a 3D Printed Optofluidic Micro Viscometer (OMV)” IEEE Sensors
Journal, Year: 2016, Volume: 16, Issue: 9, DOI: 10.1109/JSEN.2016.2527921
[2] Lucas de Souza Ribeiro; Fábio Augusto Gentilin; José Alexandre de França; Ana
Lúcia de Souza Madureira Felício; Maria Bernadete de M. França “Development
of a Hardware Platform for Detection of Milk Adulteration Based on Near-
Infrared Diffuse Reflection” IEEE Transactions on Instrumentation and
Measurement, Year: 2016, Volume: 65, Issue: 7, DOI:
10.1109/TIM.2016.2540946
[3] Maurício Moreira; José Alexandre de França; Dari de Oliveira Toginho
Filho;Vanerli Beloti; Alberto Koji Yamada; Maria Bernadete de M. França; Lucas
de Souza Ribeiro “A Low-Cost NIR Digital Photometer Based on InGaAs
Sensors for the Detection of Milk Adulterations with Water” IEEE Sensors
Journal, Year: 2016, Volume: 16, Issue: 10, DOI: 10.1109/JSEN.2016.2530873.
[4] Jinying Yin; Siqi Zhang; Hongyan Yang; Lijie Wang; Zhen Zhou “Influence of
Fat Particle Size on Light Scattering Properties in Milk Quality Testing” Year:
2014, DOI: 10.1109/IMCCC.2014.157
[5] Pallavi Gupta; Anwar Sadat; Mohd Jamilur Rahman Khan “An Opto electro
mechanical Sensor for Detecting Adulteration in Anhydrous Milk Fat” IEEE
Sensors Journal, Year: 2014, Volume: 14, Issue: 9, DOI:
10.1109/JSEN.2014.2319113
International Journal of Pure and Applied Mathematics Special Issue
30

11
[6] Ahmed Gad; Shesha H. Jayaram “Effect of Electric Pulse Parameters on
Releasing Metallic Particles From Stainless Steel Electrodes during PEF
Processing of Milk” IEEE Transactions on Industry Applications, Year: 2014,
Volume: 50, Issue: 2, DOI: 10.1109/TIA.2013.2278424
[7] Gabriel Durante; Wesley Becari; Felipe A. S. Lima; Henrique E. M. Peres
“Electrical Impedance Sensor for Real-Time Detection of Bovine Milk
Adulteration” IEEE Sensors Journal, Year: 2016, Volume: 16, Issue: 4, DOI:
10.1109/JSEN.2015.2494624.
[8] G.Rajakumar, Dr.D.Manimegalai, “VLSI Implementation of DIP Based Edible
Oil Adulteration Identification”, International Journal of Advanced Research in
Computer Science and Software Engineering (IJARCSSE), Volume 3 Issue 5,
Page No 771 – 777, May 2013, ISSN: 2277 – 128X.
[9] Carla Margarida Duarte; Ana Carolina Fernandes; Filipe Arroyo Cardoso;
Ricardo Bexiga; Susana Freitas Cardoso; Paulo J. P. Freitas “Magnetic Counter
for Group B Streptococci Detection in Milk” IEEE Transactions on Magnetics,
Year: 2015, Volume: 51, Issue: 1, Article Sequence Number: 5100304, DOI:
10.1109/TMAG.2014.2359574
[10] G.Rajakumar, Dr.D.Manimegalai, “FPGA Implementation of DIP based
Adulteration Identification in Food Samples”, International Journal of Computer
Applications (IJCA), Volume 35, No.1, Page No 06 - 11,December 2011, ISSN
0975 – 8887.
International Journal of Pure and Applied Mathematics Special Issue
31

32