Collaborative VOC and PM Source Apportionment Using Correlated Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods fail to perform collaborative source apportionment of particulate matter (PM) and volatile organic compounds (VOCs) simultaneously, lacking an algorithm to identify common pollution sources, which hinders effective collaborative control in atmospheric environments.
Innovation Solution
A method utilizing a deep learning model based on a one-dimensional convolutional neural network, self-attention mechanism, and multi-layer perceptron for PM source apportionment, combined with a positive matrix factorization model for VOCs, to correlate and attribute PM and VOCs sources with a high correlation coefficient, enabling collaborative source apportionment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If separate source apportionment models are used for PM and VOCs, then each pollutant can be analyzed independently, but collaborative source apportionment cannot be achieved and common pollution sources cannot be identified
Solution Approach 1:
The patent combines separate PM source apportionment (using optimized single particle classification model) and VOCs source apportionment (using PMF model) into a unified collaborative analysis system. By merging the results through correlation calculation, the system achieves both independent analysis capability and collaborative source identification, resolving the contradiction between separate analysis precision and collaborative control versatility.
Solution Approach 2:
The patent creates a multi-functional source apportionment system that can simultaneously perform PM analysis, VOCs analysis, and collaborative source identification. The system processes multiple pollutant types through different specialized models while integrating results to provide universal source attribution, enabling both specific pollutant analysis and broader collaborative control.
2Measurement precision
If deep learning models are used for PM source apportionment, then analysis accuracy is improved, but model complexity increases
Solution Approach 1:
The patent segments the complex source apportionment task into distinct components: an optimized single particle classification model for PM analysis and a PMF model for VOCs analysis. Each segment is specialized and relatively simple, but together they achieve high overall accuracy through their coordinated operation and correlation-based integration.
3Productivity
If collaborative source apportionment is implemented, then common pollution sources can be identified for better control, but computational requirements and processing complexity increase
Solution Approach 1:
The patent performs preliminary source apportionment analysis separately for PM and VOCs using optimized models before conducting the collaborative correlation analysis. This preliminary processing organizes the data in advance, making the subsequent integration and correlation calculation more efficient and reducing the overall computational burden compared to performing a single complex collaborative analysis from scratch.
Data Source
AI summary
A method for collaborative source apportionment of volatile organic compounds (VOCs), a product, a medium, and a device are provided. The method includes: establishing a single particle classification model; training and optimizing the single particle classification model with a local pollution library, and analyzing pollution sources of single particle mass spectrometric data to be apportioned to obtain a time series of the pollution sources contributing to the particulate matter (PM); obtaining VOCs factors of the pollution sources and a time series thereof; performing correlation calculation on the time series of the pollution sources contributing to the PM and the time series of the VOCs factors to obtain a correlation coefficient; and attributing the PM and the VOCs factor having the correlation coefficient higher than a set threshold to a same pollution source, and identifying a common pollution source of the PM and the VOCs.