Poisson Ensemble Inversion for Atmospheric Particle Size Distribution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining atmospheric particle size distributions are resource-intensive, require extensive data collection, and rely on assumptions about particle properties, leading to errors and uncertainties, especially when dealing with non-spherical particles and larger diameters.
Innovation Solution
The method involves obtaining a measured ensemble property distribution function using an extinction probe, generating theoretical distribution functions based on Poisson statistics, and fitting these to the measured data using a forward inversion algorithm, eliminating the need for scattering models and reducing computational requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If individual particle measurements are used to generate particle size distributions, then measurement precision can be achieved, but loss of time and productivity decrease due to requiring large numbers of measurements and long flight paths
Solution Approach 1:
The patent combines multiple particle measurements into an ensemble measurement approach, where the total scattering signal from multiple particles is measured simultaneously rather than measuring individual particles sequentially. This merging of measurements into an ensemble allows rapid data collection while maintaining distribution accuracy through statistical analysis of the combined signal.
Solution Approach 2:
The patent inverts the traditional approach by not trying to determine individual particle sizes from scattering signals, but rather directly inferring the particle size distribution from the ensemble scattering signal. This inversion of the measurement and analysis approach eliminates the need for time-consuming individual particle measurement and sorting.
2Productivity
If scattering models with fundamental assumptions about particle properties are used, then particle size estimation can be performed, but reliability decreases due to errors when particles are non-spherical or larger than 50 um
Solution Approach 1:
The patent changes the fundamental parameters of the measurement approach by measuring ensemble scattering properties rather than individual particle scattering. This parameter change allows the use of statistical methods that are insensitive to individual particle shape assumptions, thereby improving reliability for non-spherical and large particles.
Solution Approach 2:
The patent uses simple spherical scattering models as computational approximations that are easy to calculate, accepting small errors in exchange for computational efficiency. These simplified models serve as starting points that are refined through iterative fitting to match the actual ensemble scattering data, providing both speed and accuracy.
3Productivity
If forward scattering spectrometer technology is used, then particle size can be estimated quickly, but measurement precision decreases for particles greater than 50 um due to sensitivity to optical property deviations
Solution Approach 1:
The patent merges measurements of multiple particles into an ensemble signal, which stabilizes the measurement for large particles by averaging out the effects of individual particle optical property variations. This ensemble approach maintains the speed of forward scattering while improving precision for particles greater than 50 um.
4Measurement precision
If optical array probes are used to form shadow images, then particle size distribution can be determined, but device complexity increases and measurement precision decreases due to defocus errors and sample volume uncertainties
Solution Approach 1:
The patent replaces the complex mechanical and optical imaging system of OAPs with a simpler light extinction measurement system. Instead of forming shadow images requiring precise focus and positioning, the invention uses transmission measurements that are inherently more robust to alignment errors and sample volume variations.
Solution Approach 2:
The patent extracts only the essential measurement information (light transmission through the particle ensemble) while eliminating the complex image formation and processing requirements of OAPs. This extraction of the core measurement principle simplifies the device while maintaining measurement capability.
5Measurement precision
If intensive real time processing capabilities are used to transmit, store, and process signals from individual particles, then measurement precision can be maintained, but loss of time and use of energy increase
Solution Approach 1:
The patent combines multiple particle measurements into a single ensemble signal that requires minimal processing. Instead of transmitting, storing, and processing individual particle signals requiring intensive computational resources, the ensemble approach processes a single aggregated signal through statistical analysis, dramatically reducing energy consumption while maintaining distribution accuracy.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach allows for accurate estimation of particle size distributions with minimal processing and data collection time, reducing errors and uncertainties, and effectively handling non-spherical particles without assumptions about their properties.
Implementation Method 1
obtaining a measured ensemble property distribution function for one or more events using an extinction probe
Implementation Method 2
generating one or more theoretical ensemble property distribution functions based on Poisson statistics using one or more model parameters
Implementation Method 3
determining an event parameter property distribution function by fitting one or more of the theoretical ensemble property distribution functions to the measured ensemble property distribution function using a forward inversion algorithm
Data Source
AI summary
A method for determining a distribution of events, a method for determining a distribution of particle sizes in a sample of air, and an apparatus for performing the same are provided. The method includes obtaining a measured ensemble property distribution function for one or more events, generating one or more theoretical ensemble property distribution functions based on Poisson statistics using one or more model parameters, and determining an event parameter property distribution function by fitting one or more of the theoretical ensemble property distribution functions to the measured ensemble property distribution function using a forward inversion algorithm.