Least square regression method for estimating gas concentration in an electronic nose system

Area of Science:

• Analytical Chemistry
• Sensor Technology
• Machine Learning Applications

Background:

• Electronic Nose (ENose) systems offer a promising approach for chemical sensing.
• Accurate identification and quantification of analytes are crucial in various scientific fields.

Purpose of the Study:

• To develop and validate an ENose system capable of identifying analyte types.
• To estimate the concentration of identified analytes using advanced algorithms.

Main Methods:

• The ENose system comprises seven sensors: five gas sensors with varied heater voltages, one temperature sensor, and one humidity sensor.
• Machine learning techniques, specifically Support Vector Machine (SVM), were employed for gas discrimination.
• Least squares regression was utilized for predicting analyte concentrations.

Main Results:

• The developed ENose system demonstrated the ability to identify different types of gaseous analytes.
• The system successfully estimated the concentration of the detected analytes with reasonable accuracy.

Conclusions:

• The integrated approach of SVM and least squares regression enables effective analyte identification and concentration estimation.
• This ENose system shows potential for real-world applications requiring sensitive chemical detection.