| With the continuous development of rail transportation, rail transportation fire safety has become increasingly important topic. Based on the composition and working principle of the automatic fire alarm system and other aspects of research, in order to fast-track development zones station, for example, a system designed for automatic fire alarm problem, do a general alarm system design station, introduced implementation of the project, a gas fire extinguishing system and method of operation of fire alarm systems.Then according to the characteristics of fire detection, fire detection for the phenomenon of nonlinear structure proposed use in fire detection fuzzy neural network technology to reduce fire detection and false alarm rate improved fire detection accuracy. This paper describes the application of neural networks and fuzzy systems in fire detection, and fuzzy control, neural networks and fuzzy neural network development and principles were reviewed. In this paper, fuzzy network in fire detection can not only be common features humanoid fuzzy systems and neural networks thinking nonlinear structure function and play, and can learn from each other, complementary symbiosis.Finally, through the design and simulation of fuzzy neural network analysis, concluded: fuzzy neural network algorithm can be a good judge of a fire situation, and can easily lead to false alarms to identify interfering signals, with good intelligence. For smoldering fire or flame is difficult to determine the probability of about0.5, when the duration of the introduction of smoke to give a more accurate determination of whether a fire occurs. In fire detection technology is the application of fuzzy neural network is feasible and efficient, and the traditional neural network system in the application of fire detection compared fuzzy neural network is more powerful ability to adapt to the environment, and for this particular signal fire, more suitable for the application of fuzzy neural network to detect. |
PDF Full Text Request