Abstract:
Automatic and intelligent fire detection system is important for early detection and promptly extinguishing fire. Fire is a danger leads to quick destroying of assets and human lives. Moreover, it can propagate in fast manner according to several factors such as wind direction, fired contents (gas, fuel), etc. The fire is giant phenomena which can be controlled in its early stages but this period does not take except few seconds, there after it turns into a formidable force where to be difficult to control. With the scientific progress, the man found himself before requirements that imposing obligation upon him to consider the velocity of fire discovery where the scientific progression has been exploited in the field of communications and information transfer towards developing means of fire alarm and linking these means with Automatic Firefighting System where the time element is one of the most important factors in the process of fighting and controlling the fire.
The area of this research is limited to south Khartoum in Elshagra GAS depots and briefly the problem concentrates in intelligent fire alarm and monitoring system because they have a problem to know where the fire location and sometimes the firefighting unit miss the location of fire, also they have not scientific scenarios for people and vehicles to exit by safest way, and one of the main reasons is the lack of firefighting resources, the research here applies the concept of resource sharing by enabling communication among the nearest firefighting units. Thus, enables them to remotely monitor the gas depots and knowing the exact situation in case of fire occurrence.
iii
In this research, we survey previous studies to make our contributions of sensor networks to early fire detection and find the safest way to exit when there is alarm detected and also short path for firefighting unit to reach fired area.
The idea starts by sensing the signal coming from the sensors to define the fire location, direction and size. Then concentrate on how to reach the fire location and exit safe way for employee and people when there is a fire. The result of this research give strong justification that prove and verify our methodology with a clear contribution in Improved dijkstra algorithm for firefighting like Forced-Dijkstra and Geo-dijkstra. Furthermore, an evaluation of the above-mentioned algorithms has been done showing very promising and realistic results.
