Design Of Electrical Fire Monitoring System Based On Embedded System

This thesis has designed a kind of intelligent fire monitoring system base on Embedded Linux system, The system used in embedded technology, field bus technology, and fuzzy neural network technology. The system enables online monitoring of multiple parameters, will be monitoring the line of leakage current, copper temperature, ambient temperature and environmental tobacco smoke and other parameters are displayed through the LCD. Through this system can keep track electric line or electrical equipment, fire risk and abnormal changes of parameters of state, and the ability to detect fire hazards, through sound and light alarm and interlocks to eliminate fire hazards.This research is based on ARM9 core platform embedded system for electric fire monitoring system, software using LINUX as operating system., emphasizing on the software design based on ARM9+LINUX platform and the design of intelligent signal processing method. First, we analyze the system according to the function request. The system requirements for processor computing speed, system response time is short, low power consumption, but also a wealth of peripheral interfaces, the software needed to support a graphical user interface and the Pinyin input method support, consider these factors and costs, select the ARM920T core SAMSUNG S3C2410 processor as the core of the system processor, and select the open source LINUX as the operating system. Then we have the hardware platform of the overall planning, the modular processing hardware platform, into the CPU, memory, CAN bus, audio, Ethernet, power supply modules, each module standard between interfaces to facilitate future expansion. Based on the hardware platform, we have also been planning software system, based on the LINUX operating system, hardware drivers written, and then prepared the man-machine interface to realize the electrical fire control system software platform. In the final chapter of this article introduces the genetic algorithm optimization of fuzzy systems and artificial neural network characteristics, and how to combine the two applied to the electrical fire monitoring system, and collected a variety of signals intelligence processing, the system alarm is more accurate and reliable simulation results show that the system can meet the expected demands, with very low false positive rate.At last we summarize the whole system and give some considerations to improve the system.