Investigation Of Fire Localization Technique Based On Temperature Sensors Arrays

Automatic fire localization technology is an important topic in the automatic fire detection field. It can identify the position of fire source, and shorten time of extinguished a fire. In limited space, it is a better choice for the fire source localization by using temperature sensors arrays. This approach is based mainly on characteristics of hot gases caused by fire combustion and cross-correlation principle of array signal. Currently, by using of temperature sensors arrays to locate fire source automatically is still at the study stage. There are some different ideas and methods on the localization principles, delay time estimation, data processing and filter options.The thesis analyzes characteristics of temperature field in an early stage of fire and frequency trait of hot gases signals caused by fire combustion, designs parameters of temperature sensors, sample frequency of fire localization system and style and parameters of filter.The paper utilizes bandpass equiripple FIR filter to realize pretreatment of sample signals, compares with highpass FIR filter, the former has better performance than the latter. At the same time, the thesis points out more calculated data and more accurate on fire localization results. It is beneficial to decrease fire localization error by increasing band width of FIR filter to solve the problem of insufficient sampling points at an early stage of fire.This thesis introduces 1-bit algorithm, according to steps of Martin Berentsen fulfills signal time delay estimation by using 1-bit algorithm with changeable threshold. Through simulation analysis, the paper calculates fire source position and fire localization errors. It can draw a conclusion that 1-bit algorithm is feasible and valid method for signal time delay estimation replacing with cross-correlation principleThe paper gives an overview about the localization principle with temperature sensors arrays based on the far-field algorithm and the near-field algorithm, and discusses the performance of fire source localization principle based on combination of the two algorithms. The author deduces calculation formula of hot gases velocity in the near-field algorithm research. It can be verified to achieve the pleasing result of fire localization estimation by using simulation tools.