Gas Analyzer Learning Unknown Gas Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing gas concentration measurement systems face challenges in accurately detecting unknown gases due to absorption spectra overlapping, leading to significant cross-interference and reduced measurement precision, as conventional filtering methods and mathematical algorithms are not robust enough to handle unknown interference gases effectively.
Innovation Solution
A method and system that involve injecting standard gas samples into a measuring cell to establish a feature library, scanning for absorption spectra, and using a squared loss function to learn and train parameters to identify unknown gas features, which are then incorporated into the library for precise concentration calculation, thereby reducing cross-interference and improving measurement accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If spectral detection method is used for gas concentration measurement, then detection accuracy is improved, but measurement precision deteriorates due to absorption spectra overlapping and cross interference from unknown gases
Solution Approach 1:
The patent segments the gas detection problem by separating known gases from unknown gases in the spectral analysis process. It divides the absorption spectrum into multiple wavelength segments and applies different processing strategies: using standard spectral libraries for known gases and machine learning algorithms for identifying and handling unknown gases, thereby resolving the cross-interference issue
Solution Approach 2:
The patent introduces an intermediary machine learning model that acts as a mediator between the raw spectral data and the final concentration measurement. This intermediary layer learns to identify and compensate for the interference caused by unknown gases, enabling accurate measurement of target gases even in the presence of spectral overlapping
2Measurement precision
If filtering method is used to eliminate unknown gas interference, then measurement accuracy is improved, but device complexity increases due to additional chemical filters and reaction systems
Solution Approach 1:
The patent replaces the mechanical/chemical filtering system with a computational approach. Instead of using physical filters and chemical reactions to remove unknown gases, it uses machine learning algorithms to identify and compensate for their interference in the spectral data, significantly reducing device complexity while maintaining measurement accuracy
Solution Approach 2:
The patent introduces an intermediary machine learning model that acts as a mediator between the raw spectral data and the final concentration measurement. This intermediary layer learns to identify and compensate for the interference caused by unknown gases, enabling accurate measurement of target gases even in the presence of spectral overlapping
3Measurement precision
If conventional mathematical algorithms are used to eliminate unknown gas interference, then measurement accuracy is improved, but robustness deteriorates due to sensitivity to singular points and high sample requirements
Solution Approach 1:
The patent employs dynamic machine learning models that can adapt and learn from data, making the system robust to varying conditions and singular points. Unlike static conventional algorithms, the learning model can adjust its parameters and continue to provide accurate results even when encountering unexpected spectral patterns or interference from unknown gases
Solution Approach 2:
The patent implements feedback mechanisms where the machine learning model continuously learns from measurement results and improves its ability to identify and compensate for unknown gas interference. This feedback loop enhances the robustness of the system by allowing it to adapt to new interference patterns and maintain high measurement accuracy under varying conditions
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 effectively eliminates the impact of unknown interference gases on measurement precision, significantly enhancing the accuracy and precision of gas concentration measurements by learning and training parameters to accurately identify and quantify unknown gas features.
Implementation Method 1
spectral detection technology is widely used for quality monitoring of water, air and soil in China, particularly for detection of concentrations of industrial pollution gases
Data Source
AI summary
The present application discloses a method and a system for measuring a concentration of a gas to be measured by identifying features of an unknown gas. According to the method, a function model is constructed according to features of a known gas and features of a gas mixture, the features of the unknown gas are finally acquired by learning parameters in a training function, and the features of the unknown gas are incorporated into a feature library of the known gas, to provide more accurate parameters for obtaining the concentration of each component in the gas mixture through inversion calculation. The system includes a light source, a measuring cell, a spectrometer, a storage module, a learning and training module, an initialization module, and an inversion module.