- PhD,
PhD Defense - Solomon JIDA - ED SPI
"Contribution à l'étude des émissions à l'échappement des véhicules routiers et de leur impact sur l'environnement dans la ville d'Addis Ababa, en Éthiopie" (Contribution to the study of vehicle exhaust emissions and their impact on the environment in Addis Ababa, Ethiopia).
on February 16, 2021
Solomon JIDA will defend his PhD thesis on Tuesday 16 February at 10.00 am by videoconference from Centrale Nantes.
Supervisor: Jean-Francois HETET
Laboratory: LHEEA
Jury Members:
Stephanie LACOUR, Chargée de Recherché HDR, INRAE - UR FRISE (rapporteur)
Gilles VAITILINGOM, Director de recherché HDR, CIRAD (rapporteur)
Pascal HIGELIN, Professor, Université d’Orléans - Polytech’Orléans (Président)
Jean-François Hetet, Professeur, Ecole Centrale de Nantes (thesis director)
Pascal Chesse, Professeur, Ecole Centrale de Nantes (co- thesis supervisor)
Abstract:
Currently, automotive emission is the source of air pollution, in an urban environment. This includes PM2.5, PM10, CO, HC, NOx, and CO2. However, particulate matter emissions from road traffic comprise emissions from exhaust tailpipe, wear, and tear of friction materials of brake and clutch, tire wear, and re-suspension of dust. This study was designed to investigate the level of 24-hr roadside PM2.5 and PM10 concentrations and real-time estimation of CO, HC, NOX, and CO2 emissions from vehicles in Addis Ababa city. To estimate the share of exhaust and non-exhaust particulate matter concentrations from roadside vehicles, two methods were applied. The exhaust-tailpipe emissions were calculated using the European emission inventory Tier II method. Tier I method was used for the non-exhaust emissions like vehicle tire wear, brake, and road surface wear. Roadside 24-hr PM concentrations were estimated from the experimental data collected from 15 different sites. For the modeling, an artificial neural network (the Levenberg-Marquardt and the Scaled Conjugate Gradient) algorithm was used. Moreover, for real-time CO2, CO, HC, and NOx a distance-based emission factors method was applied. The results show that from the total traffic-related particulate emissions in the city, 63% is emitted from vehicle exhaust and the remaining 37% is from non-exhaust sources. The annual road transport exhaust emission, from all vehicle categories, results in approximately 2394 tons of PM. However, from the total yearly non-exhaust particulate matter emissions, tire and brake wear contributes around 65% and 35% is emanated by road-surface wear. Furthermore, vehicle tire and brake wear were found to be responsible for the annual emission of 584.8 tons of coarse particles (PM10) and 314.4 tons of fine particle matter (PM2.5) in the city, whereas surface wear emission was responsible for 313.7 tons of PM10 and 169.9 tons of PM2.5 emissions. This suggests that, for PM emission, non-exhaust sources might be as significant as exhaust sources and have a considerable contribution to the impact on air quality. The experimental results showed that the city average 24-hr PM2.5 concentration is 13%-144% and 58%-241% higher than air quality index (AQI) and world health organization (WHO) standards, respectively. The PM10 concentration also exceeded the AQI standards by 54%-65% and WHO standards by 8%-395%. The model runs using the Levenberg-Marquardt (Trainlm) and the Scaled Conjugate Gradient (Trainscg) and a comparison was performed, to identify the minimum fractional error between the observed and the predicted value. The two models were determined using the correlation coefficient and other statistical parameters. In the Trainscg model, the average concentration of PM2.5 and PM10 exhaust emission correlation coefficient were predicted to be (R2 = 0.775) and (R2 = 0.92), respectively. The Trainlm model has also well predicted the exhaust emission of PM2.5 (R2 = 0.943) and PM10 (R2 = 0.959). The overall results showed that a better correlation coefficient is obtained in the Trainlm model and it could be considered as an optional method to predict transport-related particulate matter emission using traffic volume and weather data for, cities Ethiopia and countries that have similar geographical and development settings. Based on the obtained average emission factors of the vehicle in the selected road type, the result was compared with Euro III and IV standards. It was found that for the entire selected vehicle in rural, urban, and motorway roads, CO emission (g/km) is 42-2309% higher than Euro IV standards, whereas except for Lifan-530 and Glory-330 vehicles, the rest of the vehicles 8-947% higher than Euro–III standards. Lifan-520 and Toyota Vitz deviate far from Euro-III and IV standards, in terms of HC and NOx emission. Besides, HC and NOx emission from the Toyota Corolla-(2003 model) also exceeds relatively the standards. The results indicated that a real driving emission test is more feasible to achieve a definitive emission pattern.
Supervisor: Jean-Francois HETET
Laboratory: LHEEA
Jury Members:
Stephanie LACOUR, Chargée de Recherché HDR, INRAE - UR FRISE (rapporteur)
Gilles VAITILINGOM, Director de recherché HDR, CIRAD (rapporteur)
Pascal HIGELIN, Professor, Université d’Orléans - Polytech’Orléans (Président)
Jean-François Hetet, Professeur, Ecole Centrale de Nantes (thesis director)
Pascal Chesse, Professeur, Ecole Centrale de Nantes (co- thesis supervisor)
Abstract:
Currently, automotive emission is the source of air pollution, in an urban environment. This includes PM2.5, PM10, CO, HC, NOx, and CO2. However, particulate matter emissions from road traffic comprise emissions from exhaust tailpipe, wear, and tear of friction materials of brake and clutch, tire wear, and re-suspension of dust. This study was designed to investigate the level of 24-hr roadside PM2.5 and PM10 concentrations and real-time estimation of CO, HC, NOX, and CO2 emissions from vehicles in Addis Ababa city. To estimate the share of exhaust and non-exhaust particulate matter concentrations from roadside vehicles, two methods were applied. The exhaust-tailpipe emissions were calculated using the European emission inventory Tier II method. Tier I method was used for the non-exhaust emissions like vehicle tire wear, brake, and road surface wear. Roadside 24-hr PM concentrations were estimated from the experimental data collected from 15 different sites. For the modeling, an artificial neural network (the Levenberg-Marquardt and the Scaled Conjugate Gradient) algorithm was used. Moreover, for real-time CO2, CO, HC, and NOx a distance-based emission factors method was applied. The results show that from the total traffic-related particulate emissions in the city, 63% is emitted from vehicle exhaust and the remaining 37% is from non-exhaust sources. The annual road transport exhaust emission, from all vehicle categories, results in approximately 2394 tons of PM. However, from the total yearly non-exhaust particulate matter emissions, tire and brake wear contributes around 65% and 35% is emanated by road-surface wear. Furthermore, vehicle tire and brake wear were found to be responsible for the annual emission of 584.8 tons of coarse particles (PM10) and 314.4 tons of fine particle matter (PM2.5) in the city, whereas surface wear emission was responsible for 313.7 tons of PM10 and 169.9 tons of PM2.5 emissions. This suggests that, for PM emission, non-exhaust sources might be as significant as exhaust sources and have a considerable contribution to the impact on air quality. The experimental results showed that the city average 24-hr PM2.5 concentration is 13%-144% and 58%-241% higher than air quality index (AQI) and world health organization (WHO) standards, respectively. The PM10 concentration also exceeded the AQI standards by 54%-65% and WHO standards by 8%-395%. The model runs using the Levenberg-Marquardt (Trainlm) and the Scaled Conjugate Gradient (Trainscg) and a comparison was performed, to identify the minimum fractional error between the observed and the predicted value. The two models were determined using the correlation coefficient and other statistical parameters. In the Trainscg model, the average concentration of PM2.5 and PM10 exhaust emission correlation coefficient were predicted to be (R2 = 0.775) and (R2 = 0.92), respectively. The Trainlm model has also well predicted the exhaust emission of PM2.5 (R2 = 0.943) and PM10 (R2 = 0.959). The overall results showed that a better correlation coefficient is obtained in the Trainlm model and it could be considered as an optional method to predict transport-related particulate matter emission using traffic volume and weather data for, cities Ethiopia and countries that have similar geographical and development settings. Based on the obtained average emission factors of the vehicle in the selected road type, the result was compared with Euro III and IV standards. It was found that for the entire selected vehicle in rural, urban, and motorway roads, CO emission (g/km) is 42-2309% higher than Euro IV standards, whereas except for Lifan-530 and Glory-330 vehicles, the rest of the vehicles 8-947% higher than Euro–III standards. Lifan-520 and Toyota Vitz deviate far from Euro-III and IV standards, in terms of HC and NOx emission. Besides, HC and NOx emission from the Toyota Corolla-(2003 model) also exceeds relatively the standards. The results indicated that a real driving emission test is more feasible to achieve a definitive emission pattern.
