Study On The Characteristics Of Light Scattering By Compare Polydisperse With Monodisperse Of Fire Smoke Particles，and With Interfering Particles

Smoke is one of the most important characteristics of the early fire. Thetechnology which was most widely used in the field of fire detection isphotoelectric smoke detection technology, it achieves fire detection based on theprinciple of light scattering of smoke particles. Currently, the common model usedto approximate the shape of smoke particles for light scattering calculations werethe simple spherical or spheroid models. Actually, the fire smoke particles’ SEMfigure show clearly that they were a particle group which have a fractal structureapproximately, number of their basic particles approximately obeys logarithmnormal distribution, and the space orientation is randomly. Based on the method ofT-matrix, a detailed comparison of the distribution of light scattering Muller matrixelements with scattering angle on the polydisperse model, the monodisperse modeland the equal-volume-sphere model of fractal structure of smoke particles which isproduced by flaming n-heptane, under the laser wavelength of300nm,633nm and800nm respectively. The results show that scattering light could reflect thecharacteristics of the fractal structure of smoke particles and the distributioncharacteristics of primary particles in smoke particles when λ is equal to300nm,then the light scattering characteristics of fractal structure of polydisperse modeland monodisperse model become identically when λ is equal to633nm, meanwhile,the light scattering characteristics of three models about the smoke particlesproduced by flaming n-heptane tend to be consistency when λ is equal to800nm.The photoelectric smoke detector alarms fire by detecting the intensity of thesmoke particles’ scattered light directly, it is easily interfered by non-fire smokeparticles such as water vapor、dust and other nuisance, where dust is a widespreadfalse alarm source. This paper was based on the T-matrix method, taking thewavelength be300nm、633nm、900nm, the light scattering characteristics werecalculated for the smoke particles and dust particles which obey certain particlesize distribution and under the condition of random orientation, and also make acomparative analysis of the feature when their light scattering Muller matrixelements change with the scattering angle.The results show that: the normalizedmatrix elements of the smoke particles which denoted as F22(θ) F11(θ)is closing to 1gradually while the incident light’s wavelength is increasing, but the dustparticles’ normalized matrix elements is less than1obviously, this shows thenon-spherical features, so it is easy to judge out of the smoke particles and dustparticles. The other Muller matrix elements change while the incident light’swavelength and scattering angle change, this could be an entry point for the futureresearch, this has an important guiding significance on researching newphotoelectric smoke detector to reduce the fire detection misstatement and alarmomission.