Assessing Robustness Regarding Weather Uncertainties for Energy- Efficiency-Driven Building Refurbishments

Chinazzo, Giorgia; Rastogi, Parag; Andersen, Marilyne

doi:10.1016/j.egypro.2015.11.021

Cited by 19 publications

(8 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Each usage type has the same form, area and operation schedules. The residential buildings are described by Chinazzo et al (2015). Variation in the outputs is also introduced by considering several years of historical weather data from many climates (weather stations) and augmenting this data with synthetic weather generated for some climates (Rastogi, 2016).…”

Section: Energyplus Datasetmentioning

confidence: 99%

Tuning machine learning models for prediction of building energy loads

Seyedzadeh

Rahimian

Rastogi³

et al. 2019

Sustainable Cities and Society

Self Cite

166

View full text Add to dashboard Cite

There have been numerous simulation tools utilised for calculating building energy loads for efficient design and retrofitting. However, these tools entail a great deal of computational cost and prior knowledge to work with. Machine Learning (ML) techniques can contribute to bridging this gap by taking advantage of existing historical data for forecasting new samples and lead to informed decisions. This study investigated the accuracy of most popular ML models in the prediction of buildings heating and cooling loads carrying out specific tuning for each ML model and using two simulated building energy data generated in EnergyPlus and Ecotect and compared the results. The study used a grid-search coupled with cross-validation method to examine the combinations of model parameters. Furthermore, sensitivity analysis techniques were used to evaluate the importance of input variables on the performance of ML models. The accuracy and time complexity of models in predicting heating and cooling loads are demonstrated. Comparing the accuracy of the tuned models with the original research works reveals the significant role of model optimisation. The outcomes of the sensitivity analysis are demonstrated as relative importance which resulted in the identification of unimportant variables and faster model fitting.

show abstract

Section: Energyplus Datasetmentioning

confidence: 99%

Tuning machine learning models for prediction of building energy loads

Seyedzadeh

Rahimian

Rastogi³

et al. 2019

Sustainable Cities and Society

Self Cite

166

View full text Add to dashboard Cite

show abstract

“…Moreover, asymmetrical urban canyons are a feasible option, so long as taller buildings cast shadows on the canyon ground and shorter buildings, allowing for openness to the sky for convective cooling. The obtained UTCI thermal stress could be reduced significantly if the calculations were made on a typical day rather than an extreme one, not to mention the uncertainties of the weather file (Chinazzo et al, 2015). Given the shorter time required for computer simulations, which exceeds a thousand runs per day, this study could be applied in various climatic contexts.…”

Section: Discussionmentioning

confidence: 99%

A parametric optimisation study of urban geometry design to assess outdoor thermal comfort

Ibrahim

Kershaw

Shepherd

et al. 2021

Sustainable Cities and Society

View full text Add to dashboard Cite

“…Notably, there are multiple uncertainties in the building operation phase. This can include occupant behavior [22], climate and weather data [23,24], and economic or regulatory boundaries [25]. Robustness assessment determines how the performance is affected under changing internal and external conditions under different plausible future scenarios [26].…”

Section: Robust Decision-making For Retrofitsmentioning

confidence: 99%