A methodology for auto-calibrating urban building energy models using surrogate modeling techniques

Nagpal, Shreshth; Mueller, Caitlin; Aijazi, Arfa; Reinhart, Christoph

doi:10.1080/19401493.2018.1457722

Cited by 71 publications

(37 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, this method was based on the calibration of each building, which requires more than 1,000 simulations to train the meta-models for each building. Nagpal et al (Nagpal et al, 2019) employed statistical surrogate models with an optimization algorithm to estimate properties of unknown building parameters. Up to 28 unknown parameters with three values each were selected.…”

Section: Question 7: How Can Results From Ubem Be Calibrated?mentioning

confidence: 99%

Ten questions on urban building energy modeling

Hong

Chen

Luo

et al. 2020

Building and Environment

294

121

View full text Add to dashboard Cite

Buildings in cities consume up to 70% of all primary energy. To achieve cities' energy and climate goals, it is necessary to reduce energy use and associated greenhouse gas emissions in buildings through energy conservation and efficiency improvements. Computational tools empowered with rich urban datasets can model performance of buildings at the urban scale to provide quantitative insights for stakeholders and inform their decision making on urban energy planning, as well as building energy retrofits at scale, to achieve efficiency, sustainability, and resilience of urban buildings. Designing and operating urban buildings as a group (from a city block to a district to an entire city) rather than as single individuals requires simulation and optimization to account for interactions among buildings and between buildings and their surrounding urban environment, and for district energy systems serving multiple buildings with diverse thermal loads across space and time. When hundreds or more buildings are involved in typical urban building energy modeling (UBEM) to estimate annual energy demand, evaluate design or retrofit options, and quantify impacts of extreme weather events or climate change, it is crucial to integrate urban datasets and UBEM tools in a seamless automatic workflow with cloud or high-performance computing for users including urban planners, designers and researchers. This paper presents ten questions that highlight significant UBEM research and applications. The proposed answers aim to stimulate discussion and provide insights into the current and future research on UBEM, and more importantly, to inspire new and important questions from young researchers in the field.

show abstract

Section: Question 7: How Can Results From Ubem Be Calibrated?mentioning

confidence: 99%

Ten questions on urban building energy modeling

Hong

Chen

Luo

et al. 2020

Building and Environment

294

121

View full text Add to dashboard Cite

show abstract

“…The full combination of the parameters leads to a massive number of simulations. In the previous UBEM calibration studies, 100 samples [24], 200 to 400 samples [22], and 1000 samples [23] were generated for each building. So, the is the end use result of the reference building for the sample, which is stored in the energy performance database., the end-use results are guessed and aggregated to generate the annual electricity and natural gas EUIs using equation (1).…”

Section: Calibrate the Ubemmentioning

confidence: 99%

Automatic and Rapid Calibration of Urban Building Energy Models by Learning from Energy Performance Database

Chen¹,

Hong²

2020

Preprint

View full text Add to dashboard Cite

Urban building energy modeling (UBEM) is attracting increasing attention in the energy modeling filed. Unlike modeling a single building using detailed building systems information, UBEM generally uses limited high-level building stock data to infer default assumptions about building characteristics and operations. This practice inherently brings uncertainty to UBEM. This study introduced a novel method of automatic and rapid calibration of UBEM based on the annual electricity and natural gas energy use data by learning the correlations between crucial model input parameters and the building energy use from the reference building models. A case study was presented to calibrate 72 large office buildings built before 1978 in San Francisco. Seventeen model parameters were selected and Monte Carlo sampling was used to create 1000 samples that reasonably represent the parameter space. Then 1000 simulations were performed for the reference building model to create an energy performance database. The results showed that by learning from the energy performance database, it took less than four simulation runs on average to calibrate a building model. After the calibration, the distributions of each parameter were obtained to replace their single predefined default values. For example, the default lighting power density of 21.39 W/m 2 was calibrated to be 7.50 W/m 2 on average. The case study successfully demonstrated the effectiveness of the novel calibration method for UBEM in the mild climate. The method will be further tested in future for other climate zones and other building types.

show abstract

“…It is equally important to address the issues of availability and quality of input data (Mangold et al, 2015;Monteiro et al, 2018;Nouvel et al, 2017), and methods to compensate for its deficiencies, e.g. probabilistic building parameter estimation (Burke et al, 2017;Nagpal et al, 2018) or data enrichment (Neun, 2007;Remmen et al, 2016;Schiefelbein et al, 2019). This goes hand in hand with development of approaches for managing the uncertainty embedded in the models and calibration of the models obtained in order to ensure good quality of the modelling outcomes (Sokol et al, 2017).…”

Section: Urban Building Energy Modellingmentioning

confidence: 99%

“…The main advantage of the proposed UBEM is the intensive utilisation of high-resolution metered data on actual energy use. This avoids use of more complex approaches to building energy simulations, including automatic generation of individual building geometry-based 'shoebox' models (Dogan and Reinhart, 2017) and their further calibration (Nagpal et al, 2018). Hence, the UBEM employs a much smaller number of simulations, as they are created for virtual archetype buildings only, making it computationally lighter than its geometry-based analogues.…”

Section: Communicationmentioning

confidence: 99%

“…However, they cannot be expected to become a 'silver bullet' for the urban energy domain, and should instead be developed as a complement to physical models (Nageler et al, 2018). Hence, a number of challenges remain to be addressed, including automated generation of models (Dogan and Reinhart, 2017), integration of various dedicated models (simulations of building energy performance, thermal networks and power grids, urban climate, life cycle analysis) (Zhang et al, 2018) and associated issues of complexity reduction (Remmen et al, 2016), calibration (Nagpal et al, 2018) and co-simulation (Arnaudo et al, 2019).…”

Section: Analyticsmentioning

confidence: 99%

See 1 more Smart Citation

Advancing urban analytics for energy transitions: Data-driven strategic planning for citywide building retrofitting

Pasichnyi¹

View full text Add to dashboard Cite

Decarbonisation of the building stock is essential for energy transitions towards climate-neutral cities in Sweden, Europe and globally. Meeting 1.5°C scenarios is only possible through collaborative efforts by all relevant stakeholders — building owners, housing associations, energy installation companies, city authorities, energy utilities and, ultimately, citizens. These stakeholders are driven by different interests and goals. Many win-win solutions are not implemented due to lack of information, transparency and trust about current building energy performance and available interventions, ranging from city-wide policies to single building energy service contracts. The emergence of big data in the building and energy sectors allows this challenge to be addressed through new types of analytical services based on enriched data, urban energy models, machine learning algorithms and interactive visualisations as important enablers for decision-makers on different levels. The overall aim of this thesis was to advance urban analytics in the building energy domain. Specific objectives were to: (1) develop and demonstrate an urban building energy modelling framework for strategic planning of large-scale building energy retrofitting; (2) investigate the interconnection between quality and applications of urban building energy data; and (3) explore how urban analytics can be integrated into decision-making for energy transitions in cities. Objectives 1 and 2 were pursued within a single case study based on continuous collaboration with local stakeholders in the city of Stockholm, Sweden. Objective 3 was addressed within a multiple case study on participatory modelling for strategic energy planning in two cities, Niš, Serbia, and Stockholm. A transdisciplinary research strategy was applied throughout. A new urban building energy modelling framework was developed and demonstrated for the case of Stockholm. This framework utilises high-resolution building energy data to identify buildings and retrofitting measures with the highest potential, assess the change in total energy demand from large-scale retrofitting and explore its impact on the supply side. Growing use of energy performance certificate (EPC) data and increasing requirements on data quality were identified in a systematic mapping of EPC applications combined with assessment of EPC data quality for Stockholm. Continuity of data collaborations and interactivity of new analytical tools were identified as important factors for better integration of urban analytics into decision-making on energy transitions in cities.

show abstract

A methodology for auto-calibrating urban building energy models using surrogate modeling techniques

Cited by 71 publications

References 24 publications

Ten questions on urban building energy modeling

Ten questions on urban building energy modeling

Automatic and Rapid Calibration of Urban Building Energy Models by Learning from Energy Performance Database

Advancing urban analytics for energy transitions: Data-driven strategic planning for citywide building retrofitting

Contact Info

Product

Resources

About