English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Missing data are frequently observed by practitioners and researchers in thebuilding energy modeling community. In this regard, advanced data-drivensolutions, such as Deep Learning methods, are typically required to reflect thenon-linear behavior of these anomalies. As an ongoing research question relatedto Deep Learning, a model's applicability to limited data settings can beexplored by introducing prior knowledge in the network. This same strategy canalso lead to more interpretable predictions, hence facilitating the fieldapplication of the approach. For that purpose, the aim of this paper is topropose the use of Physics-informed Denoising Autoencoders (PI-DAE) for missingdata imputation in commercial buildings. In particular, the presented methodenforces physics-inspired soft constraints to the loss function of a DenoisingAutoencoder (DAE). In order to quantify the benefits of the physical component,an ablation study between different DAE configurations is conducted. First,three univariate DAEs are optimized separately on indoor air temperature,heating, and cooling data. Then, two multivariate DAEs are derived from theprevious configurations. Eventually, a building thermal balance equation iscoupled to the last multivariate configuration to obtain PI-DAE. Additionally,two commonly used benchmarks are employed to support the findings. It is shownhow introducing physical knowledge in a multivariate Denoising Autoencoder canenhance the inherent model interpretability through the optimized physics-basedcoefficients. While no significant improvement is observed in terms ofreconstruction error with the proposed PI-DAE, its enhanced robustness tovarying rates of missing data and the valuable insights derived from thephysics-based coefficients create opportunities for wider applications withinbuilding systems and the built environment.

Opening the Black Box: Towards inherently interpretable energy data imputation models using building physics insight