One Building, 122,518,205 Time Series Data Points: This Is How You Maintain Safety and Performance.

Matt Gudorf, University of California, Irvine
Christopher Hartley, Altura Associates

Lab buildings are complex energy intensive spaces. In order to realize energy savings and enhance safety and performance, more dynamic controls were added. No longer are labs controlled with a simple building automation system (BMS) that provide constant air volumes over narrow ranges. Dynamic air handlers and exhaust fans now supply and remove variable volumes of air, at optimized static pressures to and from precise air flow valves, that are controlled by temperature, sash, and demand controlled ventilation sensors. These systems must perform as intended or energy savings and safety could be compromised. Monitoring of two or three systems may be possible, however the number of systems has grown not only within the building but exploded across campuses of buildings.

The next step in UCI's Smart Labs program is the deployment of a cloud-based data analysis platform that collects and normalizes data points from disparate, propriety databases. This platform was used to conduct a monitoring based commissioning (MBCx) project at Reines Hall, which is a 156,514ft2 mixed use office and wet/dry laboratory building built in 1996 and retrofitted as a Smart Lab in 2014. This data analysis platform is capable of conducting automated fault detection diagnostics (FDD), and can be used to optimize building air and water system performance during new construction, monitoring-based, continuous, and retro-commissioning projects. The optimized systems can be tracked and maintained so that their benefits can persist throughout the day-to-day operations of the building.

Data analytics enable the manual and automatic analysis of current and historical time series data that irrespective of manufacture is collected from the BMS, venturi valves, demand controlled ventilation system, hydronic and electrical metering. Data is collected using native communications protocols (Modbus, BACnet, SQL, Obix, Sedona, Haystack, XML) and stored at specified time intervals or after a change in value. The time series data can then be analyzed using a built-in toolset or customized based on user need. This enables data from many sources to be leveraged in order to make informed facility and energy management decisions. As an example, equipment start and stop times can be analyzed, and the facility manager can be alerted via email if specific equipment operates outside of the designated window, enabling targeted and direct action. Environmental Health and Safety personnel can have additional targeted notifications. New operating strategies can be verified using real-time time series charting capabilities, and tuned based on energy saving or occupant comfort needs.

Learning Objectives

  • Identify common challenges to maintaining building system performance as systems become more complex and more dynamic.
  • Understand how building data analytics tools are deployed.
  • Demonstrate 3 use cases for automated data analytics and/or fault detection diagnostics
  • Demonstrate work flow integration to promote lasting energy savings


Matt Gudorf has led UC Irvine's energy management group and Smart Labs initiative for the last 6 years. The culmination of his leadership has been record breaking energy efficiency project completion under the UC/CSU/IOU Energy Efficiency Partnership.

Christopher Hartley is a Senior Analyst with six years of experience in energy and operational analysis of commercial and institutional buildings. A graduate of the University of California, Irvine with a Master's degree in Environmental Engineering with an Emphasis in Energy, Christopher currently works with customers to install, integrate, and tailor the data collection and analysis tools discussed in this presentation. He is also working concurrently on obtaining his California PE license.


Note: I2SL did not edit or revise abstract or biography text. Abstracts and biographies are displayed as submitted by the author(s).