Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.802463
Title: Developing a risk assessment model using fuzzy logic to assess groundwater contamination from hydraulic fracturing
Author: Milton-Thompson, O.
ISNI:       0000 0004 8510 764X
Awarding Body: University of Exeter
Current Institution: University of Exeter
Date of Award: 2020
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
Technological advances in directional drilling has led to rapid exploitation of onshore unconventional hydrocarbons using a technique known as hydraulic fracturing. This process took off initially in the US, with Canada following closely behind, but brought with it controversial debates over environmental protection, particularly in relation to groundwater contamination and well integrity failure. Prospective shale gas regions lie across areas in Europe but countries such as the UK are facing public and government turmoil surrounding their potential exploitation. This extent of energy development requires detailed risk analysis to eliminate or mitigate damage to the natural environment. Subsurface energy activities involve complex processes and uncertain data, making comprehensive, quantitative risk assessments a challenge to develop. A new, alternative methodology was applied to onshore hydraulic fracturing to assess the risk of groundwater contamination during well injection and production. The techniques used deterministic models to construct failure scenarios with respect to groundwater contamination, stochastic approaches to determine component failures of a well, and fuzzy logic to address insufficiency or complexity in data. The framework was successfully developed using available data and regulations in British Columbia (BC), Canada. Fuzzy Fault Tree Analysis (FFTA) was demonstrated as a more robust technique compared with conventional Fault Tree Analysis (FTA) and implemented successfully to quantify cement failure. A collection of known risk analysis methods such as Event Tree Analysis (ETA), Time at Risk Failure (TRF) and Mean Time To Failure (MTTF) models were successfully applied to well integrity failure during injection, with the novel addition of quantifying cement failures. An analytical model for Surface Casing Pressure (SCP) during well production highlighted data gaps on well constructions so a fuzzy logic model was built to a 93% accuracy to determine the location of cement in a well. This novel application of fuzzy logic allowed the calculation of gas flow rate into an annulus and hence the probability of well integrity failure during production using ETA. The framework quantified several risk pathways across multiple stages of a well using site-specific data, but was successfully applied to a UK case study where there existed significant differences in geology, well construction and regulations. The application required little extra work and demonstrated the success and limitations of the model and where future work could improve model development. This research indicated that risks to groundwater from hydraulic fracturing differ substantially depending on well construction. Weighing up the risk to groundwater compared with financial gain for well construction will be essential for decision-makers and policy. To reduce the social anxiety of hydraulic fracturing in the UK, decision-makers who face criticism must ensure information is disseminated properly to the public with a well-defined risk analysis which can be interpreted easily without prerequisite knowledge. Finally, although this research is based on onshore hydraulic fracturing, the risk assessment techniques are generic enough to allow application of this research to other subsurface activities such as CO2 sequestration, waste injection disposal and geothermal energy.
Supervisor: Javadi, A. ; Kapelan, Z. Sponsor: Engineering and Physical Sciences Research Council (EPSRC)
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.802463  DOI: Not available
Share: