Abstract
In spite of progress in theory and technology of sensors, especially gas sensors, there is still difficult to obtain good metrological properties of them. The most important disadvantages are nonlinearity of transfer curve and quite big sensitivity for additional quantities, mainly temperature, humidity but also other gases in case of gas sensors (i.e. non-selectivity). To apply such sensors in gas concentration measurement transducers some effective methods of measurand reconstruction are required. There are described and compared four numerical methods in this article. The general assumption was that as well the main inverse transfer characteristic as the influence characteristics are described by the polynomials of order not higher than two. The first method does not include correction of additional quantities influence and was included only to compare with other methods result. Efficiency of proposed methods was tested using set of data obtain experimentally for two kind of multi-component (matrix) gas sensors based on Taguchi sensors and integrated temperature and humidity sensor. In conclusion was stated that it is possible to reach quite good efficiency of correction when a multiplicative inverse model of sensor is used, described by 18 coefficient (the 3-rd method). In this case the calibration procedure is rather laborious, but a procedure of coefficient determination and correction procedure are quite simple. Efficiency of the 4-th method is very high, but determination of inverse model coefficients is very complicated.
| Translated title of the contribution | Reconstruction of gas concentrations for multicomponent nonselective gas sensors |
|---|---|
| Original language | Polish |
| Pages (from-to) | 118-121 |
| Number of pages | 4 |
| Journal | Przeglad Elektrotechniczny |
| Volume | 86 |
| Issue number | 10 |
| Publication status | Published - 2010 |
ASJC Scopus subject areas
- Electrical and Electronic Engineering
Fingerprint
Dive into the research topics of 'Reconstruction of gas concentrations for multicomponent nonselective gas sensors'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver