eplusr: A framework for integrating building energy simulation and data-driven analytics

• Published in: Energy and buildings 2021-04, Vol.237, p.110757, Article 110757
• Main Authors: Jia, Hongyuan, Chong, Adrian
• Format: Article
• Language: English
• Subjects: Alternative energy sources; Archives & records; Bayesian calibration; Building energy simulation; Building performance simulation; Buildings; Computer applications; Data analysis; Datadriven analytics; Energy; Energy modeling; EnergyPlus; Environment models; Integration; Mathematical analysis; Optimization; Parametric simulation; Retrofitting; Simulation
• ISSN: 0378-7788; 1872-6178
• DOI: 10.1016/j.enbuild.2021.110757

[Display omitted] •Developed an R package that integrates EnergyPlus with data-driven analytics.•Structured inputs/outputs format that can be easily piped into data analytics workflows.•Facilitates reproducible simulations through Docker.•Enables flexible and extensible parametric simulations. Building energy simulation (BES) has been widely adopted for the investigation of building environmental and energy performance for different design and retrofit alternatives. Data-driven analytics is vital for interpreting and analyzing BES results to reveal trends and provide useful insights. However, seamless integration between BES and data-driven analytics current does not exist. This paper presents eplusr, an R package for conducting data-driven analytics with EnergyPlus. The R package is cross-platform and distributed using CRAN (The Comprehensive R Archive Network). With a data-centric design philosophy, the proposed framework focuses on better and more seamless integration between BES and data-driven analytics. It provides structured inputs/outputs format that can be easily piped into data analytics workflows. The R package also provides an infrastructure to bring portable and reusable computational environment for building energy modeling to facilitate reproducibility research.