Fuel Consumption Estimation and Analysis of the University of Alberta Fleet Vehicles
Bibliographic record
Abstract
This thesis aims to create a platform to estimate and monitor the University of Alberta (UAlberta) fleet vehicles’ fuel consumption and Carbon Dioxide (CO2) emissions. The main objective is to collect and analyze fleet vehicles information to reduce energy consumption and greenhouse gas emissions from university vehicles. To this end, this thesis creates a data collection platform for real-time monitoring and analysis of fleet activity, utilizing onboard diagnostics (OBD) data from each vehicle. By processing the collected data, this thesis seeks to identify the causes of high fuel consumption in the fleet and determine the optimal vehicle type for different applications and driving cycles. Two machine learning methods, including random forest (RF) and artificial neural network (ANN), were investigated to estimate fuel consumption based on OBD and actual fuel consumption data. The study used data from a Ford Escape plug-in hybrid electric vehicle (PHEV) and a Ford F-350 vehicle during real-world urban and highway driving on a 100-km route. The machine learning models utilized OBD parameters such as engine load, engine speed, intake manifold absolute pressure, air-fuel equivalence ratio, and throttle position. The validation results indicated that the RF model was more accurate than the ANN model, achieving an estimation accuracy of 99% for the Ford Escape PHEV and 100% for the Ford F-350. These findings confirm that utilizing machine learning models can effectively estimate vehicular fuel consumption; thus, these models can be used to monitor fleet vehicles’ energy consumption, and design strategies to reduce the fuel consumption from the UAlberta fleet vehicles. Additionally, this thesis investigated the energy consumption and cost of a conventional vehicle (Ford Escape S) with an internal combustion engine (ICE) and a PHEV (Ford Escape PHEV) from the UAlberta fleet. The vehicles were driven 243 times on a 20-km route in Edmonton, Canada, during 2021 - 2022. The route included both urban and highway areas. The research also explored the impact of ambient temperature (Tamb) on the operations and energy consumption of the vehicles, considering different powertrains and electrification levels. This study reveals that for warm start tests, the total energy consumption increased by decreasing the Tamb from 32 °C to -24 °C. Modes that entail continuous operation of the electric motor are especially affected. Among the modes, Auto EV (i.e., electric and hybrid electric) mode demonstrated the highest increase in energy consumption, rising by almost 452% when the Tamb drops from 29°C to -24°C. Similarly, during cold start tests, there was an increase in energy consumption as the Tamb decreased from 29 °C to -18 °C. The mode that showed the highest increase in energy consumption was EV Now (i.e., all-electric) mode, with an increase of 527% by reducing the Tamb from 29°C to -13 °C. The thesis also examined the effect of start-stop technology on conventional vehicles’ energy consumption and operational costs. To conduct the study, three vehicles from the UAlberta fleet were tested, and the effect of start-stop technology was evaluated on four different applications of UAlberta fleet vehicles. The findings indicated that the energy and cost savings achieved by vehicles equipped with start-stop technology could be significant, depending on the vehicle drive cycle and idling percentage, as well as engine size. The fuel savings are anticipated to increase during the cold season operation of fleet vehicles.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".