10th International Aerosol Conference September 2 - September 7, 2018 America's Center Convention Complex St. Louis, Missouri, USA
Abstract View
Toward a More Reliable Optical Smoke Detector: Scattering Matrix Analysis of Fire and Non-fire Aerosol for Classification
QIXING ZHANG, Jia Liu, Jie Luo, Feng Wang, Jinjun Wang, Yongming Zhang, University of Science and Technology of China
Abstract Number: 727 Working Group: Low-Cost and Portable Sensors
Abstract Point type photoelectric smoke detector, or namely optical smoke detector, is the most common and widely used fire detector in fire protection engineering. This kind of detector senses the light scattered by the smoke entered into the chamber for fire alarm. The optical smoke detector is widely accepted for its features such as low cost, compact, long life, high sensitivity, etc. However, it suffers by the high false alarm rate caused by non-fire aerosols, such as dust, steam (water vapor), cook-generated aerosol, etc. Many efforts have made to develop alternative sensors for building fire alarm, such as temperature sensor, gas sensor, or the combination of these sensors. However, the optical smoke detector still keeps a dominant market share in applications.
Development of a more reliable optical smoke detector is an urgent demand of the fire protection industry. Due to the rigorous requirement of low cost, we constrained the improvement by light scattering characterization for aerosol classification. A better knowledge of the optical properties of the aerosols from fire and nuisances will therefore help to improve the performance of existing detectors, reduce the amount of false alarms, and serve as guideline in the development of the next generation smoke detector. Scattering matrix containing several independent elements, depend on the characteristics of the particles, including size distribution, morphology, and refractive index, has an advantages over non-polarized light scattering for aerosol classification. Here, we present a study of polarized light scattering properties of typical fire and non-fire aerosol through the numerical simulation and experimental measurement of the angular distribution of scattering matrix.
Cement dust and water droplets are selected as typical non-fire aerosols, compared to fresh smoke from several flaming and smoldering test fires. The complete scattering matrix for these aerosols are measured as a function of scattering angle from 5° to 160° at a wavelength of 532 nm by a polarization-modulated scatterometer. Other related characteristics, such as particle size distribution, chemical composition, refractive index and micro morphology are also analyzed. Modelling the scattering matrix of these aerosols considering the detailed morphology are presented and verified by experimental measurement results. Discrimination methods based on the angular distributions of their scattering matrix elements were discussed. The principle guide for detector design by scattering matrix analysis were proposed. A prototype of a smoke detector based on polarized light scattering will also be presented.