Mechanical and Mechatronics Engineering

Permanent URI for this collectionhttps://uwspace.uwaterloo.ca/handle/10012/9912

This is the collection for the University of Waterloo's Department of Mechanical and Mechatronics Engineering.

Research outputs are organized by type (eg. Master Thesis, Article, Conference Paper).

Waterloo faculty, students, and staff can contact us or visit the UWSpace guide to learn more about depositing their research.

Browse

Browsing Mechanical and Mechatronics Engineering by Author "Blackmore, Daniel"

Filter results by typing the first few letters
Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Bayesian Inference for Truck-based Methane Quantification Uncertainty
    (University of Waterloo, 2024-09-24) Blackmore, Daniel
    Methane emissions from the oil and gas sector are one of the most important factors to address with respect to human-driven forces of climate change. Within Canada, the United States, and other jurisdictions worldwide, significant progress has been made in the measurement and regulation of methane emissions. While this progress has been beneficial for methane emission reduction, far less work has been performed in the understanding of uncertainties associated with methane emission measurements. Understanding these uncertainties is crucial for regulation, repair activities, and inventorying of emissions to be performed. This thesis covers a multi-year project related to the investigation of methane emissions quantification uncertainty, with a focus on the development of an uncertainty model for truck-based emissions estimates using a generalized Bayesian inference. A literature review of uncertainty analysis for methane quantification technologies is presented, as well as a detailed overview of specific technologies that were investigated during controlled release field measurement campaigns. The controlled release measurements are detailed, as well as the empirical results for the technologies that were evaluated. Subsequent chapters focus on truck-based tunable diode laser absorption spectroscopy measurements, combined with atmospheric data in the Gaussian plume model. The Gaussian plume model is derived, and the method of modelling the errors associated with this measurement technique is described. Bayesian inference is used to quantify the emission estimate uncertainty, which relies upon a CFD investigation into the errors associated with the Gaussian plume model. This thesis presents the details of how the Bayesian inference is performed – namely the form of the likelihood function, the treatment of priors, and the construction of credible intervals on the resulting posterior distributions. Then, the procedure for investigating the model error using high-fidelity detached eddy simulations is detailed. Next, the results of the Bayesian inference on the controlled release data are presented. It was found through the analysis of the applicability of the credible intervals to the true emission rates that the procedure resulted in an accurate representation of the true uncertainty of the measurement technique. Further investigation into the factors affecting the uncertainty of emission estimates revealed the measurement distance to be a significant contributor to the uncertainty, as well as very low wind speeds being a potential limitation to the technique. This thesis concludes with some discussion of the implications of the results, what limitations are present in the study, and some recommendations for future research relating to this work.