Multi-parameter Fire Detection Method Based On Dual Wavelegnths

Fire detector is essential for the fire proctection system.With the development of semiconductor technology and microelectronics process,fire detectors are lower cost and more stable nowadays,such as the optical fire smoke detector which is wildly used in recent couple of years.The optical fire smokes detector triggle a fire alarm when the light power scatted by fire smoke exceed a threshold.However,non-fire aerosols would also scatter the light and triggle a false alarm.The development of the optical fire smoke dector is seriously restrictied by its principle defect.Therefore,it becomes a research interest to develop the technology to distinguish the fire smokes generated in the early stage of fire from non-fire aerosols.This thesis begins with analyzing the operating principle of the optical fire smoke detector,and the cause of nuisance alarm by non-fire aerosols.After that,some improvement approaches are discussed.In order to develop an optical fire smoke detector with lower cost and less false alarm,the research of this thesis focuses on four aspects as follows:Firstly,Mie theory indicates that the scattering effiency relates to the particle size.In order to investigate the particle size characterization of aerosols,we established an aerosol testing and particle sizing experimental platform,and measured the particle size distribution of typical fire smokes(such as smoldering cotton,smoldering wood,open fire of n-heptan,open fire of polyurethane and so forth)and non-fire aerosols(such as standard dusts,water vapor and so on).The test results showed that the particle size distributions of fire smoke and non-fire aerosols are different since they are generated in different ways.Related reports and our experimental results indicated that the particle size distributions of non-fire aerosols are usually bigger than those of fire smokes.Therefore,we can utilize the particle size distribution and its statistical parameters to distinguish the non-fire aerosols form fire smokes.In this way,the fire detector could resist the false alarms caused by non-fire aerosols.Secondly,existing optical fire detector measures the volume(mass)concentration of aerosol as fire alarm criterion.However,volume concentration only is not enough to distinguish non-fire aerosols from fire smokes.According to the general description of particle scattering effect which is given by Baron et al.,the scattering power would indicate the volume concentration of aerosol,or the surface area concentration by adjusting the wavelength of incident light.When the volume concentration of fire smoke is equal to that of non-fire aerosol,the surface area concentration of fire smokes would higher than that of non-fire aerosol,since the particle size distrtibution of fire smokes is smaller.Base on the difference of volume concentration and surface area concentration between fire smokes and non-fire aerosols,this paper promotes a multi-parameter fire detection method by dual wavelengths technology.The volume concentration,surface area concentation and Sauter mean diameter are obtained by the scattering signals of different wavelenths,and the nuisance fire alarms caused by non-fire aerosols can be resisted by these parameters.A simulation model of non-paraxial optical system for fire detector is establelished based on Mie theory,which is utilized to optimize the dual wavelengths multi-parameter fire detector to improve the measurement accuracy for volume concentration and surface area concentration.A prototype detector was manufacted and verified by monodisper DEHS aerosols,typical fire smokes and non-fire aerosols.Thirdly,the measurement results of volume concentration and surface area concentration of black smokes by dual wavelengths multi-parameter fire detection method are lower than the true value,since that the response sensitivities of optical detector to black smokes are lower than those to white smokes.In order to triggle an alarm timely,we improve the multi-parameter fire detection method with multiple scattering channels to enhance the sensitivity to black smokes.The multiple-channel detection method would adjust the concentration conversion factors of white smokes and black smokes by multiple scattered singals.And we tested the multiple-channel prototype detector by real smokes and simulations.The results show that the detector can measure the volume concentration and surface area concentration of white smokes and black smokes correctly.