Gas Sensor Calibration Models for Device Variation Compensation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Chemoresistive gas sensors exhibit device-to-device variations due to manufacturing inconsistencies, necessitating individual calibration, which is costly and time-consuming, and their accuracy is compromised by environmental changes without recalibration.
Innovation Solution
A gas sensing device uses a calibration model based on data from a plurality of test sensor units, reducing the need for individual calibration and enabling accurate concentration determination despite manufacturing variations and environmental changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If individual calibration is performed for each sensor, then measurement accuracy is improved, but manufacturing time and cost increase
Solution Approach 1:
Multiple individual sensor calibrations are merged into a single group calibration process. Sensors are calibrated collectively rather than individually, reducing total calibration time while maintaining accuracy through statistical analysis of group data to determine calibration parameters.
Solution Approach 2:
A universal calibration approach is implemented where a single calibration model derived from multiple test sensors can be applied to production sensors. This multi-functional calibration system serves both characterization and production calibration purposes, eliminating the need for separate individual calibration steps.
2Measurement precision
If individual calibration is performed for each sensor, then measurement accuracy is improved, but manufacturing cost increases
Solution Approach 1:
Multiple individual sensor calibrations are merged into a single group calibration process. Sensors are calibrated collectively rather than individually, reducing total calibration time while maintaining accuracy through statistical analysis of group data to determine calibration parameters.
Solution Approach 2:
Calibration data and models are copied and applied across multiple sensors. Instead of performing unique calibration for each sensor, a calibration model derived from test sensors is replicated and applied to production sensors, significantly reducing calibration costs while maintaining acceptable accuracy.
3Measurement precision
If sensors are calibrated one by one, then calibration accuracy is improved, but productivity decreases
Solution Approach 1:
Multiple individual sensor calibrations are merged into a single group calibration process. Sensors are calibrated collectively rather than individually, reducing total calibration time while maintaining accuracy through statistical analysis of group data to determine calibration parameters.
Solution Approach 2:
Comprehensive calibration data is collected in advance from multiple test sensors under various conditions. This preliminary characterization data is used to create calibration models that can be quickly applied to production sensors, eliminating the need for time-consuming individual calibration steps while maintaining accuracy.
4Reliability
If calibration data from multiple test sensor units is used, then device-to-device variations are compensated, but calibration model complexity increases
Solution Approach 1:
The calibration approach transitions from individual sensor parameters to statistical parameters derived from multiple sensors. By changing from deterministic to statistical parameter representation, the system handles device variations more efficiently with simpler models that capture population-level characteristics rather than individual deviations.
Solution Approach 2:
Individual sensor-specific variations are extracted and separated from the common calibration model. The calibration system isolates the essential calibration parameters that apply universally while accounting for device variations through statistical methods, simplifying the overall calibration model structure.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for determining a calibrated measurement value for a concentration of the target gas comprises a step of obtaining a measurement signal based on the concentration of the target gas. The method further comprises a step of determining the calibrated measurement value based on the measurement signal and based on a calibration model. The calibration model is based on calibration data of a plurality of test sensor units having the same type as the sensor unit.