Particle discrimination based on multi-angle polarized elastic light scattering

A method of monitoring an aerosol environment for a particular particle, the method comprising the steps of:

• initializing testing with an optimized neural network model for a particle type from a particle type database;
• using a monitoring apparatus to sample said aerosol environment and provide a series of datasets signals based on each particle observed over a testing period and storing each dataset in a sample file for examination, wherein the monitoring apparatus is for real-time detection of a particle type in an aerosol, and the monitoring apparatus comprises:
• an interrogation volume through which a stream of aerosol is drawn;
• a focused light beam illuminating said interrogation volume;
• a plurality of polarization sensitive detectors positioned with respect to said interrogation volume, wherein at least two of said detectors are configured to detect backscattered horizontally polarized light reflections from said particles;
• a plurality of polarization sensitive detectors positioned with respect to said interrogation volume to detect forward scattered light, wherein at least one of said detectors is configured to detect horizontally polarized light and at least one of said detectors is configured to detect vertically polarized light;
• and
• a processing unit in communication with said detectors, wherein the processing unit collates the output of all said detectors when a particle interacts with said light beam into a dataset for said sample file;
• verifying each dataset in the sample file for validity;
• and
• examining each validated dataset sample with the optimized neural network model and storing the results and timestamp to a statistical database for display.

A method of monitoring an aerosol environment for multiple particle types, the method comprising the steps of:

• initializing testing with an optimized neural network model for each particle type to be investigated from a particle type database;
• using a monitoring apparatus to sample said aerosol environment and provide a series of datasets signals based on each particle observed over a testing period and storing each dataset in a sample file for examination, wherein the monitoring apparatus is for real-time detection of a particle type in an aerosol, and the monitoring apparatus comprises:
• an interrogation volume through which a stream of aerosol is drawn;
• a focused light beam illuminating said interrogation volume;
• a plurality of polarization sensitive detectors positioned with respect to said interrogation volume, wherein at least two of said detectors are configured to detect backscattered horizontally polarized light reflections from said particles;
• a plurality of polarization sensitive detectors positioned with respect to said interrogation volume to detect forward scattered light, wherein at least one of said detectors is configured to detect horizontally polarized light and at least one of said detectors is configured to detect vertically polarized light;
• and
• a processing unit in communication with said detectors, wherein the processing unit collates the output of all said detectors when a particle interacts with said light beam into a dataset for said sample file;
• verifying each dataset in the sample file for validity;
• and
• wherein each optimized neural network model examines each validated dataset sample and provides the results and timestamp for each sample to a statistical database for display and further statistical operations.