CHAPTER – 1

Design and Development of Embedded System Based Multi-Detector

For Fire Detection and Control in Strategic Industries

Section 1.1
 INTRODUCTION

Science is the beginning of civilization, it has continuously engaged in

converting the natural resources into useful products. During industrialization, many

technologies gave rise to manufacturing and production industries, but there remains

a significant gap in the area of saving life and property from fire accidents. The fire

poses a significant risk in the work place especially material manufacturing

industries such as strategic industries, metallurgical industries, chemical factories etc.

During the midst of 19th century, numbers of mechanical, electrical devices were

developed to substitute human senses in detecting the environmental changes created

by fire, viz.heat, smoke, gases (visible/non-visible).

“A vast technology has been developed to prevent, reduce, or terminate

exhausting labor and physical damage with fire accidents. It is now, dedicated to

the production of the most trivial conveniences and comfort”.

As per the NCRB statistical reports, averages of 59 industrial workers arelosing

their life every day in India due to fire accidents.Fire accidents in manufacturing

industry are increasing now-a-days due to poor fire protection and monitoring, lack

of adequate fire alarms and emergency control systems while the risk of life andloss

of property(Assets, data, and environment) are being increased. As per the statistics

of National crime records bureau (NCRB) statistics, accidental deaths due to fire

accidents in Mumbai in 2015 were 68. The largest causes of fire have been gas

leakage, electrical short circuits, fireworks and other or minor causes. Fire accidents

in manufacturing industry is increasing now-a-days, due to poor fire protection and

monitoring, lack of adequate fire alarm systems, emergency control systems while

the risk of life is being increased and loss of property(including people, assets, data,

and the environment). Industrial workshops / plants have increased their trust on fire

preparedness and care, considering advanced fire detection and alarm systems.

Different approaches stem from different backgrounds of researchers dealing with

fire such as computer science, geography, earth observation and fire safety. In fact,

fire accidents take away more lives than any other natural calamity or disaster.

Fire protection systems(including control) are classified according to the time

elapsed between the start of the fire and the actuation of the system as ultrahigh-

speed (to 0.1 second), high-speed (to 3 seconds), and standard speed (to180 second).

The fire-extinguishing substance can be applied for periods ranging from 30 second

to 3600 second [1]. The first step toward halting a fire is to properly identify the

incident, raise the occupant alarm, and then notify emergency response professionals.

This is often the function of the fire detection and alarm system. If not detected and

corrected at the right time, it can also be life alarming [2].

The present study focuses on an early stage of fire, during which an initiated

fire in the cellar/ compartment develops to trigger a fire detector installed, to

elucidate fire phenomena and predict accurately heat transfer. It has been recognized,

however, that in the initial stage of fire, combustion, unsteady heat condition of

circumstances plays a major roles.

Fire is always of great concern in hot working in metallurgical industries where

LPG fired furnaces are extensively in use. One can be pulled in to false sense of
security in working industrial gas fired furnaces, however, even these circumstances

an electrical fire can burn insulation, which will prove highly toxic and can kill or

incapacitate a work crew quickly. A fire accident is defined as “an undesirable

event which emits heat, smoke or flame. It has the potential to cause damage, may

require intervention either mechanical or human or has a cost implication”.

Generally, the source of Fire Accidents occurs due to the combustible materials, not

taken care of storage and operating failures, i.e., gases, oil or other chemical fires in

manufacturing of super alloys;

1. Stocking of combustible dust and flammable liquids and gasses

2. Faulty equipment and heating
3. Electrical fire due to short circuit

When fire fighting units engage in such situations there is a high possibility of

loss to the life of fire personnel’s. In everyday life, it is not feasible to always rely on

human patrol for detecting and extinguishing fire at an accident scene. So, to cope up

these disasters we need to design a smart detection and extinguishing system on the

basis of above mentioned reasons. There has been a huge effort over the years to

design such systems. Automatic fire detection and control is becoming very essential

to reduce the fire such as in production installations where critical materials are

placed and in operation of strategic industries. A key aspect of fire protection is to

identify a developing fire emergency in a timely manner and alert and this is the role

of fire detection and alarm system [3]. Generally fire detectors are designed to

respond at an early stage to one or more of the four major characteristics of

combustion such as heat, smoke, and flame or gas. No single type of detector is
suitable for all types of premises of fires. Heat detectors are respond to the

temperature rise associated with a fire and smoke detector respond to the smoke or

gas generated due to fire. To reduce the threat of smoke entering exit paths during

evacuation from a structure, smoke control systems can be used.This is a feature

intended to enhance life safety. An embedded system with multi-detector is

developed equipped with an alarm and control system focusing on fire detection,

protection of critical equipment from the fire hazard.

The present work involves in the fire alarm system will be effective and

tailored as per the industrial processes including the building and the types of fire

that could develop. The design of thesystem developed with due understanding of the

functions and limitations of the equipment chosen to obtain maximum efficiency and

safety.To achieve the desired level of protection, many fire alarm systems containing

acombination of smoke detectors, heat detectors, and manual pull boxes.

Section 1.2

A Brief Review of Earlier Experimental Techniques

 Most of the fire detection systems use single detector and operate without any

controlling mechanism. Those devices either use temperature sensing or smoke

detection concept to identify the presence of fire (4). Microcontroller based LPG Gas

leakage Detector using GSM module, in this system where used gas detector, GSM

module, microcontroller, if the gas concentration is increases the gas sensors will

sense the gas and then send to the microcontroller. The then GSM module is

connected to the microcontroller. Then the GSM module is connected to the

microcontroller which will gives the command to stop the main supply. The system

maintenance cost is very low.G. A. Arun Kumar(5) et al proposed design for gas

leakage detection and identification of accurate location of gas leakage system for

safety in industries using Wireless Sensor Networks (WSN), ARM7 and ZigBee for

detection and communication of data. The system is easy and overcomes the

shortcomings of traditional systems. Numbers of sensors are placed at different

locations to detect the leakage of gas and in case of emergency data is send to

observer’scoordinate system. This overcomes the weakness of existing system.

Suneel Mudunuru (6)et al proposed designed for smoke and fire detection using

ZigBee for real time security system. This paper presents a detection of fire in early

stages and gives alarm signals to save lives and property. It uses LP2148

microcontroller, photoelectric smoke detectors, wireless detector network.

Sudipta Bhatatachrjee (7)et al explained about detection of fire, and alarm

system for Board and Pillar mines using wireless sensor network (WSN). This paper

describes detection of exact fire location and timely for response. In case of

emergencies system also generates alarm to save the natural resources and the human
life from fire.Jan (8)et al has applied wireless sensor network for carbon monoxide

detection and autonomous counter measurement system for a mill. The CO sensor

module is connected to a Telos B node and interfaced with ZigBee wireless

connectivity to the central controller. The system comprised of the central controller,

a high-end PC, connected to the Telos B wireless sensor module via USB and to the

actuator circuit through RS-232.

Sudipta Bhatatachrjee et al (9) developed a wireless sensor network system for

smart gas (CO) monitoring. The system represents an energy management that

involves three sensor levels; first the network coordinator uses information from the

PIR sensor and neighboring nodes to detect the presence of a gas concentration. Next

it modulates the duty cycle of the node and the sensor board designed with a wireless

sensor network (WSN) node. Then autonomously sends the recorded data wireless.

Cetin Elms (10) et al analyzed about the data fusion framework with novel hybrid

algorithm for multi-agent decision support for forest fire. A framework for use of

wireless sensor networks in forest fire detection and monitoring, computers

environment and urban systems. Yunus ere Aslan (11)et al discussed a framework

for use of wireless sensor networks in forest fire detection and monitoring.

Rifat Hussain (12)et al most systems make use of Fire Alarm Control Panel

fitted with a digital alarm communicator transmitter, which sends information to the

central station. However, only some panels are fitted with this digital alarm

communicator transmitter. This brings about obsolescence due to the requirement of

additional hardware and interfacing. The current fire alarm systems also do not have

a means of communicating to the outside world. The sensors and detectors are
designed to report only to the Fire Alarm Control Panel. This suggests that

individuals outside cannot know the situation inside a building in case of fire. The

fire fighter cannot determine the extent to which the fire has dynamically spread

within the building from outside it. Zhang (13)et al shown in his study that a home

owner, who travelled cannot know the status of his burning house from his location.

Suvan Kumar (14)at el in his study, that major loss of lives and properties

could occur before a fire hazard can be detected and tackled. Some remote

monitoring systems can be designed by using wireless sensor networks, Ethernet and

other digital technologies, but they are faced with the issues of being too complex,

expensive and having some levels of redundancy, as well as low compactness, all

this brought to the development of an SMS based system that is configured to report

to specified number of occupants and a nearest Fire station. Shepherd (15)at el has

introduced the idea of using Bluetooth wireless technology as a cable replacement

using the wireless interconnectivity which can be implemented using radio home

automation system method. Shrikanthan (16)at el explained an automated system

based on Bluetooth wireless technology which allows the user to monitor and control

different appliances that are connected over a Bluetooth network based on a mobile

host controller. Maqsood.Jat.el alarms (17)at el implemented techniques and

provided a viable solution to realize the home automation system which constitutes

bluetooth control via Android app development for in-house control and GSM

(Global System for Mobile Communication) technology for mobile control using

Ardino. Most of the works are concerned in the open area fire detection algorithms

and architecture. The proposed system is focused on the different sources which
causes fire accident in the strategic industry (the smoke, temperature and gas

leakages). A design which contains a home network including a GPRS/GSM

gateway and three kinds of wireless security sensor nodes is presented. This system

has a user interface and it can respond quickly to alarms incidents.A new method of

moving object detection by combination of pixel illumination with its chrome in

UUV color space is implemented. The algorithm of maintenance with three key

values is discussed in this project. In case of swaying objects, it is very robust and

effective way of false in another paper presented the detection and description based

on an object oriented, statistical multi future analysis of video sequences.

Z. Bing, G. Yunhunget al (18)Safe from fire; it is an algorithm which is used

for fire detection. The technique which uses this algorithm is known as Safe fromFire

technique. In this technique three sensors are used which are flame sensors, gas

sensors and temperature sensors. Arduino is a microcontroller which is used for

collecting data from the sensors. The technique consists of two parts that are

software module and hardware module. The software part consists of webpage

through which user will login in the system and the sensors along with Arduino will

start working. The reading of sensors is passed to the Arduino to through which it is

stored in the database. Database maintains record of all the readingsof concerned

sensors. Whenever any fluctuation occurs in the readings(at least in anyone of the

two or more sensors) then an alarm alerts automatically displaying which sensors

detects the fire showing the exact place with exact time that has been occurred with

the help of image processing of the placed sensors near that particular location.

Wireless sensor networks (WSN) provide a low cost solution with respect
maintenance and installation and in particular, building refurbishment and retrofitting

are easily accomplished via wireless technologies. Fire emergency detection and

response for building environment is a naval application area for the deployment of

wireless sensor networks. In such a critical environment, timely data acquisition,

detection and response are needed for successful building automation. This paper

presents an overview of research activity on communication protocols that are

suitable for this problem and described work on the use of WSNs to improve fire

evacuation and navigation (19). The author discussed the system will detect and

control the fire accidents on running train. In-house parameters such as temperature

and humidity in the each coach can be monitored in real time. From the information

collected by the sensor system decisions for firefighting, alarming and automatic

water sprinkler or engine driver. After receiving the signal, the engine driver will

stop the train and take necessary action. The trains are moderate vehicles used for

transporting people and goods. Mostly, the people prefer the train journey for longer

distance as it is cheaper. Since induction of train for public transportation, the fire

accidents are not catered seriously by the Indian railways. The notices showing “Do

not smoke” Do not carry inflammable material are the only precautionary warnings

about the fire in each compartment. However, because of failure in routine

maintenance systems or by the activities of illegal social elements, the fire accidents

occur in train frequently [20].

There are many concerns in automatic fire detection of which the most

important ones are about different sensor combinations and appropriate techniques

for quick and noise-tolerant fire detection (21). An ideal instrument for early
detection of aircraft cargo compartment fires would combine high sensitivity, rapid

response time, automatic self-testing, and would have a zero false alarm rate. In

addition, such systems could provide continuous data to the pilots, giving them the

ability to monitor the growth and possible extinction of a fire after activation of the

on board suppression system (22).Fully engineered sensors will achieve rapid

response with low false alarm rates, will be small, lightweight, fully automatic, self-

checking, and use little power. The reduction in false alarms is accomplished by

measuring the concentrations of multiple gases in order to detect incipient

combustion. This multi-parameter detection approach has been examined by

NIST(23)Target gases for the Southwest sciences instrument includes carbon

monoxide (CO), carbon dioxide (CO2), hydrogen cyanide (HCN) and acetylene

(C2H2). Carbon dioxide is an important gas to monitor because its non-zero ambient

concentration allows the detection system to automatically confirm correct

performance.

In addition, in flaming combustion carbon dioxide concentration is well in

excess of ambient levels (350ppm) and can reach concentrations in excess of 1500

rpm(24). In contrast smoldering fires tend to produce large amounts of carbon

monoxide(25). Monitoring both CO and CO2 reduces false alarm because the ratio of

these gases due to combustion is known and should be significantly different from

the ratio due to emissions from biological cargo. Already many solutions has been

proposed and implemented for early detection fire. These systems make use of one of

the following techniques; video surveillance system, video camera sensitive to smoke

in day time, camera sensitive to fire flame at night, detection of heat flux using IR
thermal imaging cameras and LIDAR system which detects the smoke particles by

backscattering of laser light. All this system has some limitations due to atmospheric

conditions such as dust visual camera that take snapshots of the forest to detect the

communication towers(26).A rotating motor is installed to provide a full round view

of the forest. The images obtained from the camera are processed using program or

MATLB code and are compared with the reference images taken at initial stage. This

system also had limitation of high false alarm rate. Also the cost of installation of

visual cameras on communication towers was very high.

Another method is the use of satellite system to detect the forest fire. The main

components of the system are satellite(s) and the base station that collects the data

send by the satellite(s) and runs the analyzing algorithm. The raw data from the

satellite(s) is processed and then advance very high resolution radiometer (AVHRR)

instrument is used to detect presence of Hot spots. However, the clouds greatly affect

the system(27-28)Forest fire surveillance system which consists of WSN was also

proposed for detection of forest fires in South Korea. The WSN determines the

temperature and humidity after which middleware programmed and web application

analyzes the collected data. However, in this approach of detection of forest fire there

was some loss of data during communication(29) WSN consists of temperature

sensor setup and GPS module was also proposed for detection of forest fire. In this

temperature data was transmitted to base station via primary and main antenna using

satellite. Some of the antennas; continuous power was required to both temperature

sensor setup and antennas.

 In addition to this climatic/seasonal changes can affect the system(30).

Networking without pre-exist infrastructure reduces the wiring cost greatly. In recent

years, wireless sensor networks (WSN) are widely deployed in environmental

monitoring, structural health monitoring and industrial monitoring. It provides low

cost solutions for such applications. It consists of small size, low-power, and low-

cost devices that integrated with limited computation, sensing, and radio

communication capabilities. In order to implement large-scale remote monitoring,

networking techniques are introduced. Traditional wired networking technique has

been widely used in fire alarm system. But, deployment cost bring by wired network

is expensive, and cabling is difficult in some environment. Wireless networks are

dense wireless of small, low cost sensors, which collect disseminate environment

data(31). Therefore wireless sensor networks can be an alternative in these cases

since WSN are deployed without the need for any pre-existing infrastructure and

with little maintenance(32). It was known fact that, the reliability of wireless

communication is lower than wired communication. Faouzi Derbel researched the

reliability of wireless communication for fire detection systems in commercial and

residential areas and analyzed parameters’ that can influence the radio transmission

within buildings(33). In order to real-time monitor and control fire-fighting

operations, this work focuses on self-organization, fault tolerant, mobile localization.

A fire detection and rescue support framework with wireless sensor networks is

proposed(34). The progress on fire detection technologies has been substantial over

the last decade due to advancement in sensor and microelectronics(35-36)A review

of progress in various emerging sensor technologies for fire detection and monitoring
is elaborated. Mostly, fire detection technologies are categorized into two groups,

one is vision based technique that analyzers video frames and process images to

detect fire and another one is sensor based fire detection(37-38).With the increase of

number of surveillance cameras, vision based fire detection with the same equipment

sets become an attractive opportunity. Since, it does not incorporate additional

hardware budget. However, video frame detection based approach is not appropriate

for early stage fire detection. Because the early stage fire there might be only smokes

or very minimal fire flames. Video frame detection based approach can miss

detecting the fire at that particular moment,while with gas sensor fire can be detected

even before occurring of inflammation by identifying the type of leaking gas (39-40).

To detect smoke with cameras several smoke detection researchers have been

published. Since, smoke is grayish and semi-transparent edges of high frequency

image frames losses their sharpness and becomes an indicator for smoke Also smoke

is distinguished by checking the variations of background color tones, segmentation

of smoke colored pixels, blur background, illumination etc(41-42).

However, it is not clear though how this technique can distinguish between

foggy weather and smoke. To overcome this problem motion analysis is also

included in vision based technique to detect smoke accurately(43-44). Computer

vision based fire detection algorithm for fire color modeling and motion detection,

sensor networks are combined These combined approach seems very attractive

however it could increase the expenses of the system and the system complexity will

increases for installation and deployment. Compared to the above techniques and

approaches stated before SFF strategy of detecting fire is simple, less expensive and
effective to handle deceptive fire scenario’s. It is also effective for early fire hazard

occurrence detection(45-47).A fire accident is defined as an undesirable event which

emits heat, smoke or flame. It has the potential to cause damage, may require

intervention either mechanical or human or has a cost implication. A design which

contains a home network including a GPRS/GSM gateway and three kinds of

wireless security sensor nodes is presented. This system has a user interface and it

can respond quickly to alarms incidents(48). A new method of moving object

detection by combination of pixel illumination with its Chrome in Uncoloured space

is implemented. The algorithm of maintenance with key values is discussed in this

project. In case of swaying objects, it is very robust and effective way of false

alarms. In another paper presented the detection and description based on an object

oriented, statistical multi future analysis of video sequences. Safe from fire: it is an

algorithm which is used for fire detection. The technique which uses this algorithm is

known as Safe from Fire technique. In this technique three sensors are used which

are flame sensors, gas sensors and temperature sensors. Arduino is a microcontroller

which is used for collecting data from the sensors. The technique consists of two

parts that are software module and hardware module. The software part consists of

webpage through which user will login in the system and the sensors along with

Arduino will start working. The reading of sensors is passed to the Arduino to

through which it is stored in the database. Database maintains record of all the

reading of sensor whenever fluctuation occurs in the reading at least any two sensors

then automatically alarm rings.

 Current gas detection system also fall short of the ideal; tests of commercially

available CO detectors showed various failures that included false alarms at low

levels, and worse, no alarms at dangerous CO levels. A southwest science is

developing a novel optical absorption spectroscopy technique for fire detection. Key

gases produced by incipient fires are detected by optical absorbance measurements

using inexpensive, rugged mid-infrared light emitting diodes (LEDs). Fully

engineered sensors will achieve rapid response with low false alarm rates, will be

small, lightweight, fully automatic, self-checking, and use little power. This multi-

parameter detection approach has been examined by NIST.Target gases for the

Southwest sciences instrument include carbon monoxide (CO), carbon dioxide

(CO2), hydrogen cyanide (HCN) and acetylene (C2H2). Carbon dioxide is an

important gas to monitor because its non-zero ambient concentration allows the

detection system to automatically confirm correct performance. In addition, in

flaming combustion carbon dioxide concentration is well in excess of ambient levels

(350ppm) and can reach concentrations in excess of 1500ppm In contrast

smouldering fires tend to produce large amounts of carbon monoxide(49).

Monitoring both CO and CO2 reduces false alarm rates, because the ratio of these

gases due to combustion is known and should be significantly different from the ratio

due to emissions from biological cargo. Improving the Safety of equipment

maintenance practice, which includes combustion analysis, enables the boiler

technician to fully verify and maintain the equipment operating specifications for

safe and efficient operation.

Section 1.3
 Role of embedded systems -instrumentation
Instrumentation is a technology of measurement and control which serves not only science

but also all branches of engineering, medicine and etc. Now a day’s embedded

system is a very powerful tool in instrumentation (measurement and control)

systems(50-51). The advanced embedded based measurement is fast, highly reliable,

precise, accurate and all flexible to make it perfectly for any given task. An

embedded system is a computing system that is embedded to controlling some non-

computing hardware like washing machine, a car engine or a missile. Computing

machines were always the attraction for fast development of science and technology

in the field of automation for industrial needs was the motivation of science and

technology in the field of automation for industrial and domestic products.

Automation being one of the objectives, the computational needs was the motivation

for the development of computing machines and devices. The computing system is

part of procedure or process.

The sidebar “describes the historical progression of embedded systems design

practices” three Generations of embedded Systems Design. Current trends are

moving away from the specificity that particular programming languages and

implementation platforms offer toward greater generosity. Practices typically use

higher levels of abstraction, as in model based design, and apply either critical or

best-effort systems engineering(52-53).Microcontroller is an electronic single chip

microcomputer that is more suited for measurement and control of instruments.

Microcontroller has been treated as a tool for computing and communication with

other devices. Knowledge of microcontrollers is meaningful and very rewarding if it

is applied to design a product that is useful in the industry or for the society in
general(54). Microcontrollers are more suited for control and automation of process,

analytical, biomedical industries and etc. There are two types of architectures

available namely, Harvard architecture and Von Neumann architecture.

Hardware architecture

Harvard architecture uses separate memories for program and data with their

independent address and data buses because of two different streams of data and

buses. The more designs of microcontrollers are followed by Harvard architecture.

Not only the architecture supports parallel buses for address and data, but also it

allows a different internal organization such that instruction can be pre-fetched and

decoded while multiple data are being fetched and operated on. Further, the data bus

may have different size than the address bus. This allows the optimal bus widths of

the data and address buses for fast execution of the instruction. The examples of

Harvard architecture are MCS-51 from Intel Corporation and PIC from Microchip.

Von Neumann architecture

In Von Neumann architecture, program and data share the same memory space.

Von Neumann architecture allows storing or modifying the programs easily.

However, the code storage may not be optional and requires multiple fetches to form

the instruction. Program and data fetches are done using time division multiplexing

which affect the performance. The example of Von Neumann architecture is 68HC11

from Motorola.An embedded system can be defined as a computing device that does

a specific focused job. Appliances such as the air-conditioner, VCD, DVD player,

etc., are example of embedded systems. Each of these appliances will have a
processor and special hardware to meet the specific requirement of the application

along with the embed software that is executed by the processor for meeting that

specific requirement. The embedded software is also called firmware. The desktop /

laptop computer is a general – purpose computer. It canbe used for variety of

applications such as playing games, word processing accounting, and software

development and so on. In contrast, the software in the embedded systems is always

fixed. Some of special features are listed below;

● Embedded systems do a very specific task

● Embedded systems have limited resources that have particularly the memory

● A specific job has to be completed within a specific time. In some embedded

systems is called real-time systems

● Embedded systems consume very low power. Many embedded systems have been

operating through a battery power

● Embedded systems are highly reliable

● Some of embedded systems operate in extreme environmental conditions such as

very high temperatures and humidity.

● More embedded systems are addressed to the consumer market for example,

electronic toys, mobile phones, home appliances etc.

The embedded system market is one of the highest growing areas as these

systems are used in consumer electronics, office automation, industrial automation,

bio-medical engineering, wireless communication, data communication,

telecommunication, transportation, military etc. Embedded system hardware includes

components such as user interface, memory size, Display, Source of power, memory
 cards, timers and other discrete components like detectors, signal conditioners,

hooters, valves, water pumps, sprinklers etc., All the hardware components were

explained in detail in the previous chapter including the total processes flow of

signals and data.

Embedded system software is a collection of codes written to perform a

specific function. Most embedded system software can be run on microcontrollers.

This software can be written in several ways. Some of the methods are Machine

Codes. These codes are usually hexadecimal codes and provide basic instructions on

how each processor will work. Machine codes are replaced by a programming

language.

Programming Language

The software used in the embedded system is Open w.r.t, PikeOS, eCos,

Fusion RTOS, Nucleus RTOS, RTEMS, INTEGRITY, UC / OS, QNX, FreeBSD

and OSE. The code is usually written in C or C ++, but some high-level

programming languages like Python and JavaScript are now also used for

microcontrollers and embedded systems.

Software Features

The main role of the software in the present study is to govern the following

activities;

● To make the data to the system to convert the analogue signal of the detectors
output into corresponding digital information for the data processing in
microcontroller.
 ● To detect and measure the Smoke, leakage of LPG, and Heat.
● To check the detectors functionality by means of software to find out missing
signals of the positioned detectors.
● To make the different functional units of system work in a systematic and
sequential manner.
● To compute, display and storage of information of smoke, concentration of LPG
and heat.
● To indicate the hardware defects if any.

The necessary software in the present study is developed in C language to

implement these tasks for effective functioning of the system.

Section 1.4

Purpose and scope of the present study

Nowadays, fire accident in industries is posed as a serious threat to the loss of

human life, health and property. The largest causes of fire have been fuel gas

leakage, electrical short circuits and fireworks. The main purpose of the present
research work is to design and develop an embedded system based Multi-detector for

fire detection, alarm and control, which can be used especially in strategic industries.

The Multi-detector gives the ultimate in protection against both slow and fast

developing fires. Multi-detector comprises true multi-criteria units and the output

levels such as smoke level, temperature level and fuel gas concentration are

continually monitor by the detector and onboard processor using algorithms

developed specifically for each task. An alarm signal is only enabled in the control

unit once the processor is satisfied that an incipient fire has been detected. By using a

combination of inputs, the incidence of nuisance alarms are reduced at the same time

actual fire is not impaired. The multi-detector achieves high reliability and accuracy.

The ability of the detection system is accurately predicting the performance of the

smoke, fuel gas and temperature detections, which are an integral part of the

measurements associated with fire safety design.

In present methods, the fire detectors (smoke and gas leakage detectors)

located in given occupancy at cellar or away from work place confined space which

cause significant delay in the detection and activation of smoke or fire alarm. This

delay causes serious injuries or deaths to the human being. The low concentration of

smoke, gas leakage and temperature rise the industrial working atmosphere which

becomes a serious fire hazard and potential threat for human life and valuable

property.The early detection of smoke and gas leakage is essential in strategic

industrial working environment which protect from the danger and save the working

people and properties.

 It is essential to detect and measure the concentration of smoke and gas leakage

to activate the alarm and control the fire at the early stage.No doubt, several

investigators developed the techniques for the measurement of fire detection and

control in different methods and few manufacturers are producing a variety of

instruments for early warning, but these are conventional, indirect and suffer from

many disadvantages. But the attempts to design and develop the embedded system

based systems for detection of fire and control is rather scarce particularly in India

through they offer many advantages.

Hence, in the present study, an attempt is made to design and development of

embedded system based Multi-detector for early detection and control of fire for life

safety and other applications in strategic industries.

REFERENCES

1. M.H.Adam," Automatic Fire Fighting System" IOSR,ISSN(e);2250-

3021,ISSN(p);2278-8719,Vol.04,Issue 12 (December 2014),Version-4 pp-64-68.

2. D.Das,J.Saika,J.Deka,T.Ismail,GSM based gas leakage detection system with

preventive measures" International Journal of Electronics and Communication

Engineering and Technology(IJECET), ISSN 0976-6464,vol.5, Issue 5,

May(2014),pp 122-135.

3. Asif M.Mukashi, Dr. U.V, Wail “Smart fire detection system in industrial

application using CAN Protocol" International journal of Recent Trends in

Engineering and Research".

4. Majid Bahrepour, Nirvana Meritnia, Paul Havinga Pervasive Systems Group,

University of Twente: “Automatic Fire Detection: A Survey from wireless sensor

network perspective”.

5. L.Liu,R.Sun,S.Al-Sarawi," A smart bushfire monitoring and detection system

using GSM technology”, International Journal of Computer Aided Engineering

and Technology, 2010,2(2/3),pp.218-232.

6. G.A.Arunkumar et al(G A Arunkumar, K.Rajashekar, BVV Satyanarayana and

K.Suryanarayana Murthy "Implementation of Real time Detection of Gas leakage

in industry using ARM7 & Zigbee" International Journal of Engineering

Research & Technology ) IJERT, Vol.1,Issue 7,September-2012.

7. Suneel Mudunuru et al, V. Narshimhanayak, G. Madhusudhana Rao,

K.SreenivasaRao,"Real-time security control system for smoke and fire

detection using Zigbee", International Journal of Computer science and

information technologies, vol.2(0),2011,2531-2540.

8. A.Somov, A.Baranov, A.Savkin, D.Spirjakin,A.Spirjakin, R.Passerone,

"Development of Wireless Sensor Network for Combustible Gas Monitoring",

Sensors and Actuators A:Physical 171(2011)398-405.

9. Sudipta Bhattacharjee “Fire alarm system based on SMS" Int,J Communications

Network and system Science, 2014,7,386-395.

10. Certin Elms A data fusion framework with novel hybrid algorithm for multi-

agent decision support system for forest fire Expert systems with applications,

vol.38,2011,pp.9225-9236.

11. Yunus Ere Aslan vol.36, 2012, pp.614-625.

12. Rifat Hussain, "An intelligent fire detection and mitigation system safe from

fire", Dept. of Computer science and Engineering University of Liberal

Arts,Bangladesh"2010.

13. Zhang L.at.el, “Design and implementation of Automatic Fire Alarm System

Based on Wireless Sensor Networks". Proceedings of the International

symposium on information processing microcontroller. Retrieved March 16,

2006.

14. Suavan Kumar 2015,"GSM based Industrial Security System" Dept of Electrical

Electronics Engineering, Galilea institute of Technology.

15. R.Shephered,"Bluetooth wireless technologies in the home" journal of

Electronics and Communication Engineering, vol.13,no.5, pp, 195-203,

Oct.2001.
16. N.Srikanthan and Tan Karand "Bluetooth Based Home automation system",

Journal of Microprocessor and Microsystems, vol.26,pp.281-289,Elsevier

Science B.V., 2002.

17. Y.Zhao and Z.Ye,”Low cost GSM/GPRS BASED wireless home security

system”, IEEEE Trans. Consumer Electron, vol.56,no.4,(2007) anuary,pp.546-

567.

18. Bing, G.Yunhung,L.Bo,Z.Guangwei and T.Tian,”Home Video Security

Surveillance”, Info-Tech and Infonet,2001, Proceedings, ICII 2001-Beijing.2001

International Confrance, vol.3, pp.202-208.

19. Bing, G.Yunhung,L.Bo,Z.Guangwei and T.Tian,”Home Video Security

Surveillance”, Info-Tech and Infonet,2001, Proceedings, ICII 2001-Beijing.2001

International Confrance,vol.3,pp.202-208.

20. Majid Bahrepour, Nirvana Meritnia, Paul Havinga Pervasive Systems Group,

University of Twente: “Automatic Fire Detection: A Survey from wireless sensor

network perspective”.

21. R. Pitchai Ramaswamy1, M.PraveenKumar1, S.SaratKumar2 and R.RaghuRaman

22. Bike, D. Aircraft cargo Compartment smoke detector alarm incidents on the US

registered air craft, 1974-1999.DOT/FAA/AR-TN00/29,2000

23. Mike J. and McAvoyT.Analysis of Fire and Non-fire Signatures for

Discriminating Fire Safety Science – Proceedings. Fifth international Symposium

1997; 819-82Mike J. and McAvoyT.Analysis of Fire and Non-fire Signatures for

Discriminating Fire Safety Science – Proceedings. Fifth international Symposium

1997; 819-82.
24. Pfister, G., Fire Safety Journal, 1983; 6: 165-174, and Jackson.M. And Robins, I

Gas sensing for fire detection; measurements of CO, CO 2, H2, O2 and smoke

density in European standard fire tests, Fire safety journal 1994; 22: 181-205.

25. Z.Liu and A.K Kim,”Review of recent developments in fire detection

technologies”. Journal of fire protection Engineering, vol.13, no 2 pp 129-151,

2003.

26. T.Celik, H Demirel, h Ozkaramanli, and M.Uyguroglu, “Fire detection using

statistical color model in video sequences”. Journal of Visual Communication

and Image Representation”, vol.18 no.2 pp 176-185, 2007.

27. Y.Dedeoglu,B.UToreyin.U.Gudukbay, and A.E.Cetin “Real-time fire and flame

detection in video (in ICASSP(2),2005,pp 669-672.

28. A.Somov, D. Spirjakin, M. Ivanov, I.Khromushin, R.Passerone,A.Baranov, and

A Savkin, “Combustible gasses and early fire detection: an autonomous system

for wireless sensor networks”, in Proceedings of the 1st International Conference

on Energy-efficient Computing and Networking.ACM.2010,pp.85-93.

29. D.Kristinic,D.Stipanicev, and T.Jakovcevic. “VHistogrambased smoke

segmentation in forest fire detection system”. Information technologies and

control, vol.38,no 3.2015.

30. C.B.Liu, and N.Ahuja,”Vision based fire detection, “in pattern Recognition, 2004

ICPR2004,Proceedings of the 17th international conference on vol.4

IEEE,2004,pp.134-137

31. Gross handler, W.A Review of measurements and candidate signatures for early

fire detection. NISTIR 5555, January 1995.

32. Osterlind,F,: Pramsten,e; Roberthson, D. Eriksson,J: Finne, N; Voigt, T.

Integrating building autorotation systems and wireless sensor networks

Proceedings of Emerging Technological and Factory Automation, 2007, 1376-

1379.

33. FaouziDerbelReliable wireless communication for fire detection systems in

communication and networking, 2003, 654-65.

34. KeweiSha,Weisong Shi Watkins,O Wayne State Univ. Detroit, MI; Using

wireless sensor networks for fire rescue applications; requirements and

challenges,IEEE International Conference on Electro / Information

technology,2006,239-244.

35. J.San-Miguel-Ayanz and N.Ravail,”Active fire detection for fire emergency

management: Potential and limitation for operational use of remote sensing

“Natural Hazards, vol.35, no.3.pp.36-376, 2005.

36. Z.Liu and A KL Kim, “Review of recent developments in fire detection

technologies, “journal of fire protection Engineering vol, 13, no 2. Pp, 129 – 151,

2003.

37. T.Celik.HDemirel,HOzkaramanli and M.Uyguroglu, “Fire detection using

statistical color model in video sequences”, Journal of Visual Communication

and Image Representation”. Vol. 18 no pp,176-185,2007.

38. Y.Dedeoglu B.U Toreyin U Gudukbay, and A.E Cetin “Real-time fire and flame

detection in video”inICASSP(2),2005,pp.669-672.

39. A. Somov, D.Spirjakin, MIvanov,IKhromushin, R.Passermone, A Baranov, and

A.Savkin, “Combustible gases and early fire detection: an autonomous system

 for wireless sensor networks, “in Proceedings of the 1 st International Conference

on EnergyEfficient Computing and Networking , ACM,2010 pp 85-93

40. KrstinicD.Stripanicev, and T.Jakovcevic,”Histogram based smoke segmentation

in forest fire detection system” ‘Information Technology and Control, vol.38 no.

3, 2015.

41. C.B.Liu and N.Ahuja “Vision based fire detection”, in Pattern Recognition,2004

ICPR 2004 in proceedings of the 17 th Intentional Conference on vol.4,IEEE

2004,pp.134-137.

42. B.U Toreyin,Y.Dedeoglu, and A.E cetin “Wavelet based real-time smoke

detection in video”, in Signal Processing Conference 2005 13 th European

IEEE,2005,pp.1-4.

43. D.KrstinicD.Stripanicev, and T.Jakovcevic,”Histogram based smoke

segmentation in forest fire detection system” ‘Information Technology and

Control, vol.38 no. 3, 2015.

44. C.B.Liu and N.Ahuja “Vision based fire detection”, in Pattern Recognition,2004

ICPR 2004 in proceedings of the 17 th Intentional Conference on vol.4,IEEE

2004,pp.134-137.

45. T.Celik,HOzkaramanali, andHDemirel, Fire and Smoke detection without

sensors: image processing based approach “in 15 th European signal processing

conference,EUSIPCO,2007, pp 147-158.

46. (K.Angayarkkani and N.Radhakrishnan,”Efficient forest fire detection system: a

spatial data mining and image processing based approach “International Journal

of Computer Science and Network Security, vol.9.3.pp 100-107,2009.

47. M.Bahrenpuor, N. Mertinia and PJ Havinga, “Automatic fire detection: a Survey

from wireless sensor network perspective”2008.

48. A Ollero J Maritinez-De Dios, and B Arrue, “Integrated systems for early forest

Research 14th Conference on Fire and Forest Fire Research 14 th Conference on

Fire and Forest Meteorology, Luso, vol.16,1998.p.20.

49. L.Yu,NWangand X Meng,”Real-time forest fire detection with wireless sensor

networks“in Wireless Communications International Conference vol. 2

IEEE,2005pp,1214-1217.

50. BCNakra,KChaudharyÏnstrumentation Measurement and

Analysis”,TataMcGrawHillPublishers,NewDelhi(1995).

51. CSTangan,GRSharma and VSNMani, Instrumentation Devices and

Systems”,2ndEdition,TataMcGraw-Hill Publishers,NewDelhi(2002).

52. Thoam A Henziner, Joseph Sifakis ”The Discipline of Embedded Systems

Design “Published by the IEEE Computer Survey, 0018-9162/07-2007IEE.

53. KVK Prasad, Embedded / Real systems; Concepts, Design and Programming”,

Black Book, Reprint Edition, Published by Dreamtech press, New Delhi (2007).

54. Ajay Deshmukh,” Microcontrollers theory and Applications”, Eight reprint, Tata

McGraw – hill publishers, NewDelhi.