Abstract
dc:descriptionThe topic of this dissertation is related to thermal flow meters and the flow error compensation due to the different thermo-physical properties of the working medium, which could be other than the medium used for the calibration of the sensor. The motivation of this dissertation was not only to make the thermal flow meters gas independent, but also to upgrade their functionality and make them more intelligent. The adopted compensation approach was based on the use of the thermo-physical properties and the appropriate heat transfer model. This compensation approach requires the accurate measurement of these thermo-physical properties. Consequently, the main focus of this work was on the development of measurement procedures and sensors, which can accurately measure thermal conductivity, volumetric heat capacity, density and dynamic viscosity within gas thermal flow meters. For this reason, two MEMS based sensors were selected, one micro-cantilever beam from atomic force microscopy (AFM) that was deployed as gas density and dynamic viscosity sensor and a second micro-wired sensor for the measurement of thermal conductivity and volumetric heat. The micro-cantilever sensor presented very high sensitivity in respect to density variation 220 Hz/(kg/m^3). The fluid viscous effect on the damping mechanism of the sensor is the dominant effect, this means that thermo-elastic and other secondary damping effects can be considered negligible. Therefore, one advantage of this approach is that the sensor should not be calibrated in vacuum conditions. Six pure gases and sixteen mixtures of them in ambient conditions were investigated. The investigation showed that the accuracy of the determined gas density and viscosity in ambient conditions reached 1.5% and 2.0% respectively. The micro-wired sensor is a thermal property sensor measuring thermal conductivity and volumetric heat capacity. It comprises three free standing parallel micro-wires, one operates as heater, while the remaining two operate as temperature sensor. The proposed modelling approach of the sensor is based on reduced order modelling, in contrast to the traditional analytical modelling approach, which is the standard for this kind of sensors. A parametric model order reduction (para-MOR) scheme based on proper orthogonal decomposition was applied. This led to the reduction of the full order problem with 150000 DOFs-FEM to a reduced problem with 35 DOFs-MOR. The sensor achieved an accuracy in thermal conductivity of 6.5% and in volumetric heat capacity of 3.2%. The higher error in case of thermal conductivity is related to strong variation of the sensor sensitivity in its operational range. The analysis also revealed the limitations of the selected sensor pointing to directions of improvement. Furthermore, the uncertainty analysis of the gas properties error on the mass flow measurement was also investigated for both thermal flow meter principles. The propagation uncertainty theory was used for the calculation of the mass flow uncertainty based on the partial uncertainties of the gas properties. The compensated mass flow uncertainty of the calorimetric principle was on the order of 3.4%.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2018
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Badarlis, Anastasios
- Μπαδαρλής, Αναστάσιος
Subjects
dc:subject × 12Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/42994