D. Phil. (Energy Studies) / Quantifying energy consumed and emissions produced by transport is essential for effective policy formulation and urban environmental management. Current first-world methods for determining vehicle emissions factors are technology and resource intensive, and results cannot be applied directly to cities in other parts of the world. There is a need for alternative cost-effective and accurate methods for determining real-world fuel consumption and emissions from vehicles in cities of the developing world. In this thesis, a new emissions simulation and inventory model is developed and implemented as a software tool. A novel application of low cost on-board diagnostics equipment and Global Positioning System sensors is devised to survey engine-operating parameters, driving conditions and vehicle usage profiles needed by the model. An emissions inventory is produced for the City of Johannesburg using the software tool and surveying method to demonstrate the overall process. The core contribution of this thesis is the logical development of data structures and software tools which link base engine-operating patterns (of engine speed and engine load), derived from the literature, to measured engine-operating patterns and vehicle activity from real-world driving. A range of real-world driving cycles and emission factors published by the Swiss Institute of Materials Science and Technology are transformed to produce the base engine-operating patterns and their corresponding emissions factors. The calculation of emission factors for real-world driving involves matching measured engineoperating patterns to combinations of the base engine-operating patterns using numerical methods. The method is validated using a cross validation technique. The emissions inventory application integrates measured engine-operating patterns, vehicle activity, fleet structure, fuel sales and the emissions simulation procedure to calculate total emissions. Fuel consumption and emissions of interest are CO2, CO, HC, NOx. Measurements of engine operating parameters and vehicle usage patterns were recorded for 30 privately owned passenger vehicles from the Johannesburg fleet. The selection included Euro-0 (a mixture of pre Euro-1 vehicles), Euro-2 and Euro-3 petrol vehicles, and Euro-2 diesel private passenger vehicles. Fifteen billion vehicle kilometres were driven in Johannesburg by private passenger vehicles per year consuming 325 million litres of diesel and 1 524 billion litres of petrol. iv Total emissions were estimated to be 4.13 Mt CO2, 82.77 kt CO, 9.15 kt HC, and 24.49 kt NOx. Between 88 and 93% of the total emissions were from vehicles which fall into the Euro-0 petrol category. Diesel vehicles did not make a significant contribution to CO and HC emissions but contributed 14% of the NOx and 19% of the CO2 emissions. During weekdays, 28 to 31% and 25 to 27% of the total fuel consumption and emissions were due to the morning commute and the evening commute periods respectively. Although minibus taxis, buses, freight and vehicle age significantly impact on total fuel consumption and emissions in cities they were not considered within the scope of this study. Vehicle usage patterns are analysed to produce spatial maps and diurnal charts of congestion on suburban roads, streets and highways within the Johannesburg municipal area. Times and locations of congestion are presented in terms of a standard congestion index, and suggestion given on how and where congestion problems could be addressed. This study shows that vehicle emissions inventories can be cost effectively produced by surveying engine-operating parameters and vehicle usage profiles using on-board diagnostics and Global Positioning System sensors and simulating emissions factors using a new emissions simulation and emissions inventory model.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:uj/uj:6611 |
| Date | 16 November 2009 |
| Creators | Goyns, Philip Hugh |
| Source Sets | South African National ETD Portal |
| Detected Language | English |
| Type | Thesis |
Page generated in 0.0016 seconds