Jain, Nishesh; (2021) Reducing the performance gap using calibrated simulation models. Doctoral thesis (Ph.D), UCL (University College London).

Preview

Text 2021_01_21_Complete_Phd_V1.pdf - Submitted Version Download (17MB) | Preview

Abstract

Buildings have a significant impact on the environment. Construction of buildings and their operation accounts for 36% of global final energy use and 40% of energy‐related carbon dioxide (CO2) emissions. Also, as per the 2019 International Energy Agency (IEA) and United Nations Environment Programme (UNEP) report, the building sector has a strong potential to provide long-term energy and greenhouse gas emission savings without high financial costs. Building performance simulation tools, ranging from steady-state calculations to dynamic simulation methods, can calculate the anticipated energy consumption of a building with adequate levels of accuracy. However, there is considerable evidence to suggest that buildings underperform post-completion when compared against the expected performance prediction during the design-stage. The difference between the actual operation and the design intent is termed the ‘performance gap’. While the energy performance gap in buildings is a well-known phenomenon, its in-use interpretation is quite vague. It is important to understand the basis of assumptions and protocols used in design-stage performance calculations to assess the causes of the performance gap. In the context of the performance gap, energy performance is generally the most emphasised. The gap, however, is not limited to energy – it also applies to indoor environmental quality (IEQ) parameters, such as temperature and air quality. Moreover, the pursuit of energy efficiency may have the unintended consequence of compromising IEQ, thereby requiring a comprehensive approach to performance assessment. It is therefore important to consider energy and IEQ performance issues together. This thesis contributes to an improved understanding, quantification and resolution of performance gap related issues by using a novel simulation-based approach that enables systematic identification and classification of the root causes of the performance gap. A new measurement and verification (M&V) framework that is underpinned by building performance simulation and calibration is proposed. A key aspect of this new methodological framework is the identification and separation of the three types of performance gaps because of: 1. Use of inappropriate design-stage calculation methods (such as those used for regulatory compliance), 2. Technical issues with the building, its systems and their operations, and 3. Operational changes that the building has gone through to meet its functional requirements. For the first type of performance gap, CIBSE TM54 (CIBSE, 2013a) already provides guidance to reduce the perceived gap and enable improved estimates of building performance during the design-stage. This thesis focuses on the understanding of operational-stage issues and their detailed causes, related to the second and third types of the performance gap. This thesis is the first study that systematically defines, identifies and separates, • the technical issues that cause the performance gap between design intent and actual operation, and • the deviations of operating conditions from the design that are driven by the building’s function and occupancy. This is achieved by integrating the conventional post-occupancy performance assessment approach with building performance modelling and evidence-based model calibration. Another addition to the conventional approach, explored in this study, is the incorporation of IEQ. The issue of IEQ is addressed in two ways: first, by using zonal temperatures for calibration cross-validation, and second, by assessing the energy-related unintended consequences of IEQ underperformance which may happen during building operations. The calibrated simulation models are operationally accurate virtual representation of the actual building and can help to isolate the performance issues and validate the findings. The new framework proposed in this thesis is better suited than conventional M&V protocols such as ASHRAE (American Society of Heating Refrigerating and Air Conditioning Engineers) Guideline 14 and IPMVP (International Performance Measurement and Verification Protocol). These conventional M&V protocols also propose a calibration-based approach, but they generally focus on broad statistical requirements and are not tied to a framework for a procedural verification of all the most important issues that can cause the performance gap. It is likely that using these conventional protocols will identify some key issues during investigations while leaving other potential issues hidden. The guidance on calibration and validation provided in conventional M&V protocols is commonly used for all model calibration exercises. However, the conventional protocols were developed for calculating energy savings in retrofit applications, and the calibration criteria defined in them are mainly for checking the accuracy of building-level energy use totals. The calibration criteria do not check for the uncertainty or the accuracy of dependent parameters, such as zone temperatures and other environmental outputs, which could cross-validate the model. Mathematically, meeting just the statistical criteria for building-level energy use totals in a highly parameterized model and an under-determined search space can lead to unrealistic solutions also being validated. To better support the calibration accuracy with the new proposed M&V framework, advanced model validation criteria have also been developed. New multi-level calibration criteria are proposed, which factors in data quantity, quality and granularity. In this new advanced validation criteria, the current industry standard of monthly energy use checks is the lowest level of calibration, with higher levels requiring detailed checks, using granular and disaggregated energy use. However, all levels of calibration require minimum dependent parameter checks, such as IEQ checks for typical zone temperatures. Dependent parameter checks are desirable in model calibration; however, current statistical criteria used for calibration are not suitable for these checks. Revised and new metrics and thresholds are proposed and explored in this thesis for use in advanced calibrated model validation checks. Beyond the use of IEQ parameters (e.g. zone temperatures) in model calibration, another area of focus of this thesis is the unintended IEQ underperformance captured during the monitoring. The scope of this assessment is limited to underperformance in IEQ parameters linked to achieving high energy efficiency objectives, thermal comfort and indoor air quality (IAQ). Amongst the various IEQ parameters, thermal comfort and IAQ have complex and dynamic interactions with buildings energy end-uses. Comprising of multiple factors, which are both subjective and empirical, thermal comfort and IAQ performances have a high interrelation with the energy performance objectives. Therefore, along with conventionally tracked parameters of temperature and CO2, additional IAQ parameters (not used during the calibration process), such as NO2, PM2.5 and PM10, are analysed to enhance the understanding of unintended energy-related IEQ underperformance. The new methodology proposed in this study is applied to five case study buildings across four building sectors – offices, schools, hospitals and apartment blocks. These buildings represent a large cross-section of the UK building stock and, therefore, can provide useful insights into the issues in the construction sector that drive the performance gap. While detailed performance assessment and advanced validation is done for all five case study buildings using the proposed framework, in one case study building the multi-level calibration checking criteria is also fully explored. Using this methodology on the various case study buildings, cross-sectoral lessons, related to root causes of the energy performance gap and applicable in the wider industry context, are uncovered. Linking to the three types of performance gaps mentioned earlier, analysis of the results shows that, in most of the case studies, some of the energy performance gap is the perceived gap (related to point 1: use of inappropriate design-stage calculation methods) or is because of operational changes (related to point 3: changes that the building has gone through to meet its functional requirements). However, the most critical cause of the gap is due to technical issues (related to point 2: issues with the building, its systems and their operations) identified across the case studies. These issues were either design errors, improper construction and installation, poor commissioning or shortcomings in building systems and the use of new low-carbon technologies. It was observed that long-term involvement (with responsibility for the operational performance) of the design and construction teams are effective in lowering performance gaps. Issues related to IEQ were also observed across the case studies, such as overheating risks and poor IAQ. These added to the existing knowledge of energy-related IEQ issues and highlighted the need to address IEQ simultaneously with energy through better design, advanced operational controls and by incorporating regular IEQ measurements as part of operations and maintenance protocols. The novel approach presented here builds a case to move building performance calculations towards an operational context, where design projections are done using advanced simulation and with a view of tracking the projections through to operation using measurable performance outcomes. Overall, the study shows the importance of the early involvement of all stakeholders and their accountability to minimise performance issues. Integrating the findings from the case studies, a case could be built for having IEQ performance objectives in energy performance contracts. This can mitigate the trade-offs of IEQ against energy performance that leads to unintended health consequences for occupants. Further, this work promotes a way of integrating dynamic thermal simulation in regular post-occupancy checking and management of buildings.

Download activity - last month

Export as JSONCSVXML

Download activity - last 12 months

Export as CSVXMLJSON

Downloads by country - last 12 months

Export as XMLCSVJSON

Archive Staff Only

View Item