Abstract
Measurements from low-cost gas sensor systems for air quality monitoring are affected by cross-sensitivities, interferences with environmental factors, sensor aging, concept drift, and unit-to-unit variability. As a consequence, ensuring metrological traceability and low mea- surement uncertainty, two properties that characterize reliable measurements, for larger sensor networks (e.g., in smart cities) is considerably challenging. The following work is dedicated to reliable air quality monitoring with low-cost gas sensor systems in such cases, proposing several tools and strategies to either increase or track the reliability of those measurements.Firstly, the relocation problem with field calibrated systems is revised and traced back to lacking representativeness of the calibration data, typically followed by concept drift. With this knowledge, a compact continuous-flow automaton that allows characterizing cross-sensitivities and interferences with environmental factors as well as resolving spatial and temporal reloca- tion problems is presented. It generates orthogonal atmospheres, i.e., gas mixtures at different relative humidities and temperatures, in an efficient manner using fractional factorial designs for the simultaneous calibration of an array of low-cost sensor systems in the laboratory.
For field calibrated systems, which are heavily affected by concept drift, machine learning algorithms that monitor the trustworthiness of incoming measurements are proposed and dis- cussed. Anomalies are detected by estimating the support of the sensor signal distribution and by assessing the position of new signals with respected to this support. Moreover, it is demon- strated how such algorithms might be evaluated with strategies from software validation.
Lastly, a theoretical concept for the stochastic online recalibration of gas sensor networks by means of mobile reference instruments is presented. Recently developed gradient update rules such as RMSProp (with and without momentum) are explored. The analysis demonstrates that the reliability of the measurements could be maintained in this manner as sensor aging and concept drift are continuously compensated for.
| Date of Award | 30 Nov 2023 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Edward Smart (Supervisor) & Victor Becerra (Supervisor) |