Site Loader

Fire Controller System using Fuzzy Logic for Safety Protection

Abstract

1: Introduction
Human life is more valuable. In past, there are so many incidents of fire disaster at home, buildings and in industries. There can be so many reasons of fire, the problem is that how to detect fire, how to control fire and how to safe human life? For this purpose we purposed a solution using Artificial Intelligence technique called Fuzzy Logic.
The idea of Fuzzy Logic was proposed by Dr.Lotfi Zadeh of the University of California at Berkeley in 1960s. Fuzzy Logic has been applied to many fields in which control system, neural networks and artificial intelligence are included. Fuzzy Logic is based on “degrees of truth” which is many valued logic rather than binary logic. Binary Logic(two-valued logic) often consider 0 which mean false and 1 which mean true. However, Fuzzy Logic deals with truth values which lies between 0 and 1, and these values are considered as degrees of truth or intensity of truth.
Fire is an undesirable event. It is important to detect fire at early stage as soon as possible to safe human life. Fire controller system proposed in this paper integrates the use of instruments and connectivity. Earlier to detect fire single sensor were used, which provide faulty output1. After that multi sensor were used to detect fire which give better output, and more reliable results, as well as overcome the false alarm system. In proposed work with application of Fuzzy Logic, flame sensor and temperature sensor are used. which provide better result.

Fig1: Block diagram of FLS

2: Literature Survey
Several year before, fuzzy control become a one of the most beneficial area where research done in the application of fuzzy set theory. There were many technologies which is proposed for early fire detection for safety, such as Neural Network, image processing, Video based techniques, fuzzy logic etc. Fire detection is become an important research and development topic, many technologies are used for the surveillance of early fire detection2.
In 3, author build “Smart Forest Fire Early Detection Sensory System(SFFEDSS)”, by combining wireless sensor networks with artificial neural networks(ANNs). In this system, low-cost sensor nodes, temperature, light and smoke data spread out on the forest in order to get information, and this information is taken as input to ANN models that is converted into intelligence. Without any human involvement this SFFEDSS system monitor the forest.
In4, in the Mediterranean regions, to detect fire at early stage, author analysis the application of Spinning Enhanced Visible and Infrared Imager SEVIRI images. In this techniques, two or more images are compared with each other after interval(15-min).
In5, author proposed an fire alarm system, by using less expensive instruments, connectivity and wireless communication. It is real time monitoring system, in which system detect presence of fire as well as captures images by using camera and display images on screen. In this system two controller are used, Controller 1 send signals to GSM and Controller 2, and Controller 2 will turn ON the screen. When it find that temperature, humidity, CO2 and fire is increased above the threshold, these outputs are taken from sensors. One or more arduino are connected and trigger automatically.
In6, in this paper author proposed a System Safe From Fire(SFF) by using multi sensors, actuators and operated by micro-controller unit(MCU) is an intelligent self controlled smart fire extinguisher system. Sensors placed in different areas for monitoring purpose, and input signals taken from that sensors, and combines integrated fuzzy logic to detect fire breakout location and severity, and discard false fire situation, such as cigarette, smoke, welding etc. SFF notifies fire services and other by text messages and telephone calls when fire is detected.

3: Methodology
In this paper, fuzzy control algorithm is used to detect fire and control fire in residential area. Sensors are placed in different areas, and used to detect any abnormal behavior where it is placed. Whenever any abnormal behavior is detected, fire detection sensors are used to set an alarm and other control mechanism, if somewhere fire is detected. Rule based fuzzy logic is implemented on data which is collected from different sensors.
In MATLAB, fuzzy logic toolbox is used for simulation with more accuracy, flexibility and scalability with other system. The model of Fuzzy Logic is consist of Fuzzification, Fuzzy Rule, Fuzzy Inference System and Defuzzification process.

Diagram: Flowchart of fire controller system

Diagram: Block Diagram of fire controller system

3.1. Fuzzification:
In fuzzification, crisp value is converted into fuzzy linguistic variables using membership functions. Membership function are used to correlate input variables into real world parameters. Membership function has a value of 0 to 1. Selection of membership function is based on process knowledge.2

Mamdani inference system are most widely used. In mamdani form Two or more inputs and one output are described by collection of rules by using IF-THEN.
In this paper 3 inputs are used, change rate of temp, humidity and flame, flame presence remain constant so we used 2 inputs in simulation change rate of temp and humidity. Change rate of temp is taken by comparing 2 temp, previous temp and current temp, that how much temp is changed. Such that: if (temp1 is low) and (temp2 is low) then (change rate of temp is low). Output is the probability of fire presence.
The membership function if change rate of temp and humidity is LOW, MIDIUM, HIGH and probability of fire is V.LOW, LOW, MIDIUM, HIGH, V.HIGH.

FIG:MF for change rate of temp
FIG:MF for humidity

3.2. Inference Rules:
In fuzzy inference system, rule is constructed to control output variable. Fuzzy rules are so easy to construct, understand, it is just simple IF-THEN rule with a condition. We can say that fuzzy rules controls the complete fuzzy system. For example if(temp is low) and (target is high) then (AC is on). In IF part knowledge is captured and in THEN part is used to give conclusion or output in linguistic variable form. In this paper rules are constructed carefully because it is used to control fire in residential area, to avoid any disaster. In MATLAB we use rules viewer to add, delete and edit rules which shows how each rule behave in a system.

Sr. Change the rate of temp Humidity Flame Probability of Fire presence Solution
1 Low High Present V.Low Alert
2 Low Medium Present Low Alert
3 Low Low Present Low Alert
4 Mid High Present Mid Alert
5 Mid Medium Present Mid Alarm/ Water Shower(Low)
6 Mid Low Present High Alarm/ Water Shower(High)
7 High High Present High Alarm/ Water Shower(Low)
8 High Medium Present V.High Alarm/ Water Shower(High)
9 High Low Present V.High Alarm/ Water Shower(High)
Table: Fuzzy rules for fire controller system

For instance, according to above mentioned rules table the rules can be read ad follows:
If (change rate of temp is LOW) and (change rate of Humidity is HIGH) then (fire presence is Low) and (alert service is ON).
If () and () then () and ().
If () and () then () and ().
If () and () then () and ().
If () and () then () and ().
If () and () then () and ().
If () and () then () and ().
If () and () then () and ().
If () and () then () and ().

3.3 Defuzzification:
Defuzzification is a process of converting fuzzy set to crisp values. By using rules set, the output of fuzzy set is calculated. Rules are simple in the form of “If-then” statement. The input of fuzzy system is fuzzified and rules are applied to this fuzzified input. After applying rules, each rules generate fuzzy output, and this fuzzy output is converted into crisp value. This process is called defuzzification. There are many defuzzification methods, but centroid defuzzification method is used in this paper.

sensors

4. Results and conclusion
In this paper, two sensors Temperature Sensor(DHT-22) and Flame Sensor are used. Temperature sensor give the value of temperature and humidity. Humidity is the presence of water vapor in the air. After getting data from sensors, fuzzy algorithm is applied. For detecting fire, flame must be present so that flame value is not used in simulation because it is taken as Boolean value (yes\no). Flame presence? if yes then fuzzy controller will work.
In this paper, change rate of temperature and change rate of humidity is used. Because in different cities there are different temperature and humidity. So to use change rate of temperature and humidity the result is more accurate.

REFERENCES

1 IJESRT (aiswarya muralidharan, fiji joseph)
2 Vikshant khanna, Rupinder kaur cheema, “Fire Detection Mechanism using Fuzzy Logic”, International Journal of Computer Applications (0975-8887), Volume 65 – No. 12, March 2013.
3 Hamdy soliman, Komal Sudan, Arish Mishra, “A Smart Forest Fire Early Detection Sensory System, Another Approach of Utilizing Wireless Sensor and Neural Networks”, IEEE SENSORS 2010 Conference, 2010.
4 Giovanni laneve, “Continuous Monitoring of Forest Fires in the Mediterranean Area Using MSG”, IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, Vol.44, NO.10, OCTOBER 2002.
5 Digvijay Singh, Neetika Sharma, Mehak Gupta, Shubham Sharma, “Development of System for Early Fire Detection using Arduino UNO”, International Journal of Engineering Science and Computing, Vol.7, NO.5, May2017.
6 Md Iftekharul Mobin, Md Abid-Ar-Rafi, Md Neamul Islam, Md Rifat Hasan, “An Intelligent Fire Detection and Mitigation System Safe From Fire(SFF)”, International Journal of Computer Application(0975-8887), VOL,133, NO.6 January 2016.

Post Author: admin