Data-Driven Smart Buildings

LBNL “Building 59” Office Building

The LBNL Office Building (i.e. Building 59 or Wang Hall) is a medium-sized office building located inside the Lawrence Berkeley National Laboratory (Berkeley Lab) campus in Berkeley, California. The building has 10,400 m²of conditioned spaces on four floors. The lower level provides space for mechanical systems, the second level is the National Energy Research Scientific Computing Center (NERSC), and the third and fourth levels are office spaces. The ground office floor (third floor) is primarily closed office space, while the second office floor (fourth floor) is primarily open office space. The building was built in 2015 and retrofitted in 2019 to improve its energy efficiency. Model predictive control (MPC) technology was implemented in the building automation system (BAS) to optimize HVAC operations (supply air temperature setpoint, air damper position, fan speed, hot water valve position) for saving energy.

Project Information

Location

Berkeley, California, USA

Building Typology

Office Building

Technology Installed / Proposed

Digital twin: building performance dataset and related metadata semantic models, model predictive control, occupant measurement.

Data Availability

High-resolution three-year time-interval data from more than 300 sensors and meters including whole-building and end-use energy consumption, HVAC system operating conditions, indoor and outdoor environmental parameters, as well as occupant counts and Wi-Fi connection counts.

Status

Operational - Results Available

Project Aim

Reducing energy waste in buildings and optimising building operations require access to a diverse and integrated set of data. However, it is currently time consuming and hard to find datasets that have adequate data coverage, good data quality and clear documentation. Measuring ground truth at high resolution in all buildings is impractical and challenging. Therefore, it is critical to collect, curate and make publicly available high-resolution data from a small number of buildings that have broad applicability to a variety of high-impact use cases.

The building has rich sensing and monitoring systems, which provide data for implementing and testing MPC to validate its performance against the baseline rule-based controls. The data was curated and published to support public research on building energy and controls.

The uniqueness of this dataset includes:

A rich high-resolution three-year time-interval data of a real office building, which includes two years of pre-pandemic data, and the year 2020 when the COVID-19 pandemic started.
The building was used for model predictive control research and field testing.
The dataset has camera-based occupant count measurements as well as proxy virtual sensing from the Wi-Fi-connected devices.
A metadata semantic model based on the Brick schema, an open-source effort to standardize semantic descriptions of the physical, logical, and virtual assets in buildings and the relationships between them was developed to represent the metadata of the sensors, meters, and HVAC systems.
A semantic description of the dataset (including building and system characteristics, and information on data curation, data quality, data categories, and application aspects) was developed.

Additional Information

The dataset is hosted at Dryad website: https://doi.org/10.7941/D1N33Q.

The Python code for detecting and filling the data gaps, as well as for modifying outlier values, is available at the dataset’s GitHub page: https://github.com/LBNL-ETA/Data-Cleaning.

N. Luo, Z. Wang, D. Blum, C. Weyandt, N. Bourassa, M.A. Piette, T. Hong. (2022). A three-year dataset supporting research on building energy management and occupancy analytics. Scientific Data, 9(1), 156.

N. Luo, T. Hong. Energy and occupancy analytics to improve understanding and efficiency of building operations – A case study of an office building in Northern California. International COBEE Conference, Montreal, July 2022.

D. Blum, Z. Wang, C. Weyandt, D. Kim, M. Wetter, T. Hong, M.A. Piette. (2022). Field Demonstration and Implementation Analysis of Model Predictive Control in an Office HVAC System. Applied Energy, 318, 119104.

Four RTUs serving the two office floors of the building

Architecture of the Model Predictive Control (MPC) implemented in the building

For more information on the Case Study

Contact Person: Dr Tianzhen Hong, Dr David Blum

Lawrence Berkeley National Laboratory agree that the case study information of LBNL “Building 59” Office Building can be shared under CC BY-NC-ND 4.0 license. This license allows others to download your works and share them with others as long as they credit you, but they can't change them in any way or use them commercially.