Linear Model for Concentration Measurement of Mixed Gases

Area of Science:

• * Chemical Sensors
• * Analytical Chemistry
• * Signal Processing

Background:

• * Electronic noses are valuable tools across various industries, including food preservation and environmental monitoring.
• * A significant challenge in their application is the cross-sensitivity of gas-sensing responses, hindering accurate mixed gas concentration measurement.
• * Existing methods struggle to precisely quantify individual gas concentrations within complex mixtures.

Purpose of the Study:

• * To develop a robust methodology for accurately determining mixed gas concentrations using electronic noses.
• * To establish accurate sensor response and concentration measurement models that account for cross-sensitivity.
• * To validate the proposed models through simulation and experimental data.

Main Methods:

• * Determining the number of state variables influencing cross-sensitivity from experimental data.
• * Establishing a state space model derived from an equivalent circuit model.
• * Utilizing parameter correlation iterative algorithms and a Kalman filter for parameter estimation.
• * Developing sensor response and mixed gas concentration measurement models.

Main Results:

• * The developed state space model effectively represents the electronic nose's behavior under mixed gas conditions.
• * The sensor response model accurately predicts sensor outputs influenced by multiple gases.
• * The concentration measurement model demonstrates high accuracy in quantifying individual gas concentrations within mixtures.
• * Both simulation and experimental results confirm the models' predictive and measurement capabilities.

Conclusions:

• * The proposed state space modeling approach effectively addresses the cross-sensitivity issue in electronic noses.
• * The developed models provide a reliable method for accurate mixed gas concentration measurement.
• * This advancement enhances the applicability of electronic noses in complex sensing environments.