Files
Belleflamme_38281800_2024.pdf
Embargoed access until 2025-06-24 - Adobe PDF
- 5.62 MB
Details
- Supervisors
- Faculty
- Degree label
- Abstract
- The objective of this thesis is to analyse and correct the effect of hysteresis in micro-sensors. Hysteresis is a non-linear phenomenon commonly observed in various physical systems. This hysteresis poses significant challenges for signal modelling and system control. This thesis focuses on the identification, modelling and mitigation of the hysteresis effect in a gas micro-sensor. The standard score is used for normalisation. For filtering, we use wavelets with a filtering threshold and a Savitzky-Golay filter. For curve modelling, sigmoids are used and for hysteresis simulation a Taylor expansion is used. The model achieves good result. It cancel the hysteresis effect on the trained data, with an MSE of 0.074.