Search

Measuring building thermal performance is important for meeting goals in energy efficiency, emissions reductions, and housing quality. Energy Performance Certificate (EPC) assessments performed in the UK are limited in their capacity to assess the in-situ thermal performance of dwellings, while approaches are expensive and impractically invasive for use on a mass-scale. The UK roll out of smart meters will result in widespread collection of energy data. However, to-date no established method exists for inferring thermal performance from such data. This research aims to develop a method of measuring the thermal performance of existing dwellings using smart meter data and weather data derived from location which can be applied to large numbers of dwellings. This research presents the “Deconstruct” method to estimate dwelling thermal characteristics, notably the Heat Loss Coefficient (HLC), from smart meter and weather data. This method uses a grey-box building physics model to describe the relation between dwelling thermal properties and measured energy demand. A 'post-hoc control trail' approach identifies data subsets which enable robust model parameter estimates. This enables dwelling characterisation without the need for a controlled experiment with on-site monitoring. Evaluation of the Deconstruct confirmed that steady-state modelling assumptions made were reasonable for the majority of dwellings from a range of datasets. HLC values were inferred with a mean uncertainty of 15% for 63% of 7529 sites from the Energy Demand Research Project (EDRP) dataset. Future monitoring studies are proposed which address shortcomings in existing datasets. Limitations of the method are discussed. Deconstruct could enable rapid estimation of key thermal performance parameters for millions of UK dwellings, with a range of applications including policy evaluation, retrofit assessments, and advice to dwelling occupants.

Flexible bluff bodies exposed to a uniform fluid flow undergo vibration due to vortex-shedding from the body. This phenomenon is known as vortex-induced vibration (VIV). Structural members, such as cables, conduits, and pipes, are susceptible to VIV. Vortex-induced vibrations result in structural stresses due to elastic structural deformation from the flow-induced loading. Cylindrical structures in proximity to planar surfaces can undergo vortex-induced vibrations when the gap between the cylinder and the planar surfaces is above a critical value. This laboratory study investigated the mechanism of vibration and planar wall proximity effect on the cylinder response to the vortex shedding process. The vibration mechanism at subcritical gap ratios is hypothesized to be movement-induced vibrations (MIV) caused by unsteady fluid flow interactions. On the other hand, vortex-induced vibration has been established as the vibration mechanism at large gap ratios in uniform flow (Blevins, 1990). A circular cylinder, mounted on two degree-of-freedom (DOF) leaf springs at each end, was positioned in a tolerant, open return subsonic wind tunnel at Bucknell University. At a reduced velocity of Ur = 5 and a Reynolds number, using the cylinder’s diameter, of 1.73 x 104, a series measurement of two-DOF structural accelerations along with the fluctuating wake velocity at a fixed position relative to the cylinder was recorded. This study provides data on the relationship between two DOF vibrations and the proximity of a circular structure to a planar surface. For gap ratios of G/D > 1.0 and δ/G ≈ 0.30, the alternate shedding of vortices (the von Kármán vortex street) produces a fluctuating y-direction acceleration of the body at the same frequency as that of vortex shedding. Furthermore, the transverse y-direction acceleration has an associated streamwise x-direction acceleration at twice the vortex shedding frequency, indicating a fluid-structure interaction due to von Kármán vortex shedding as expected for a cylinder in a uniform flow. For G/D < 0.5 and δ/G ≈ 0.61, it is theorized that the bistable upstream wall boundary layer separation bubble periodically detaches and reattaches to the outer front top quarter surface of the cylinder as the cylinder moves upstream and towards the wall. As a result, coherent vortices are shed only from the outer side of the cylinder. The cylinder’s resulting MIV from the fluid forces has an oblong acceleration trajectory, reinforcing the single-sided vortex shedding from the cylinder. The bistable wall boundary layer is suggested as a possible mechanism for the near-wall region.

Landslides on transportation networks are becoming increasingly prevalent in response to environmental change, at the same time as demand placed on road and rail infrastructure is growing in line with population growth and economic development. To minimise the impact of slope instability on vital transport infrastructure, as well as on terrestrial and aquatic ecosystems, public safety, communities, local heritage, and the economy, and to better inform decision making, there is a need for remote-condition-monitoring systems to reveal processes that are precursors to slope failure. Here, the application of a novel electrical resistivity tomography (ERT) system (Proactive infrastructure Monitoring and Evaluation (PRIME)) for monitoring unstable slopes that affect transport infrastructure is examined. The system provides near-real-time data collection via a telemetric link. 4-Dimensional resistivity models (3-D with a time-lapse element) are presented in the context of moisture content and soil suction, two parameters known to influence slope stability. Building upon relationships evident in the field, the development of laboratory-based petrophysical relationships that relate electrical resistivity to moisture content and soil suction directly is also discussed. 4-D ERT models were calibrated using these petrophysical relationships to provide insights into complex spatial and temporal variations in moisture content and soil suction. Two study areas are considered here, enabling different controls on electrical resistivity to be examined. The Ripley Landslide in the Thompson River Valley, British Columbia, Canada, threatens the serviceability of two national railway lines. Here, 4-D ERT models reveal complex hydrogeological pathways, providing new insights into the subsurface processes in an active, natural landslide. Freeze-thaw cycles also impact the subsurface resistivity here, which highlights the need to consider the impact of temperature when inferring changes in moisture content from ERT data. The Old Dalby Railway Cutting in Leicestershire, United Kingdom, is also considered in this thesis. Subsurface changes in resistivity at this site across both wooded and grassed areas of the slope enable an improved understanding of hydrogeological pathways and allow examination of the use of ERT for assessing the influence of vegetation on subsurface moisture pathways and on slope stability. This thesis highlights the utility of geoelectrical monitoring for assessing slope stability in the context of moisture-driven landslides. It aims to emphasise the benefits of 4-D geoelectrical monitoring and points towards the future incorporation of ERT monitoring in integrated decision support and early warning condition monitoring systems.

This thesis is an exploration of the design of successful urban places, those that are filled with people enjoying the vitality present in a density of life. The inspiration for this thesis came from a statement made by Jason Bergen. While walking through downtown Blacksburg, unsuccessful in our quest to find a place sit down and eat lunch, Jay said, "You know Jon, there are no good places to sit in Blacksburg." Without realizing it, Jay had made a simple observation that can be used to assess the quality of an urban environment.