Aristotle University Of Thessaloniki (AUTH)
Εκτίμηση της αέριας και σωματιδιακής ρύπανσης στην καμπίνα των επιβατών αυτοκινήτου
Abstract
dc:descriptionThe dissertation focused on the measurement of concentrations of the gaseous pollutants CO, NO, NO2, SO2, O3, which are emitted to a large extent by the transport sector. In addition, the number of particles were measured and grouped by size for PM0.3-0.5, PM0.3-1.0, PM1.0-2.5 and PM2.5-10. Simultaneously with the above, the values of environmental parameters, inside and outside the cabin, were recorded throughout the measurements, as well as information on the operation and motion of each vehicle that participated in the measurements phase. The main objectives of the study were to determine the In/Out ratio of the five gaseous pollutants for different ventilation scenarios and to investigate the possible correlation between the environmental parameters and those of the vehicle with the variation of the concentrations per pollutant inside the vehicle cabin. Two devices (inside and outside the vehicle) of high-precision electrochemical sensors were used to measure the gaseous pollutants. Three approaches were investigated for their calibration, namely the Single and Multiple Linear Regression and the Random Forest algorithm. The devices generally followed the pattern of the time series of the reference station of the Municipality of Thessaloniki in the average resolution of 10 minutes. However, their performance was lower in terms of the reference values, mean error and overall deviation. The calibration significantly improved the agreement between the measurements of the two devices and the reference device. The use of Simple Linear Regression was finally chosen as the basic calibration method. The field measurements were performed in the city of Thessaloniki. Most routes were selected within the residential area, where the highest concentrations of pollutants have been observed. There were four basic scenarios that were repeated most of the time. The choice of these scenarios was based on the literature: closed windows, half-open windows, closed windows with ventilation activated and closed windows with the ventilation activated and air recirculation on. The results reported in the I/O ratio of the concentration of gaseous pollutants show a higher concentration of NO, SO2 and CO pollutants inside the cabin compared to the outside environment for all ventilation scenarios. On the other hand, it seems that the concentration of NO2 and O3 gases is lower inside the vehicle cabin compared to the outside environment. The ventilation scenario is observed to have a significant effect on the ratio of concentrations inside and outside the cabin. With regard to particles, the pattern of change in particle concentrations of different diameters with the ventilation scenario appears to be similar between vehicles. For smaller diameter particles, the concentrations increase as we move from the closed windows scenario with the ventilation activated and air recirculation on to the semi-open windows scenario. The vehicle type seems to affect the concentration of particles, of any diameter, inside the cabin. The exploratory attempt to predict the concentration of a gaseous pollutant in the cabin resulted in the development of linear multiple regression models. Depending on the pollutant and the ventilation scenario, the values of the adjusted R2 ranged from 54% to 70%.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2021
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Margaritis, Dimitrios
- Μαργαρίτης, Δημήτριος
Subjects
dc:subject × 18- Αέρια και σωματιδιακή ρύπανση
- Μετρήσεις στην καμπίνα
- Ηλεκτροχημικοί αισθητήρες
- Στατιστικά μοντέλα
- Gas and particulate emissions
- In-cabin measurements
- Electrochemical sensors
- Statistical models
- Gas- en fijnstofemissies
- Metingen in de cabine
- Elektrochemische sensoren
- Statistische modellen
- Χημεία
- Φυσικές Επιστήμες
- Χημεία, διεπιστημονική προσέγγιση
- Chemical Sciences
- Natural Sciences
- Chemistry, interdisciplinary
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/50433