Using IoT to measure and verify energy consumption in buildings
An Internet of Things (IoT) platform can revolutionize the traditional measurement and verification process approach.
When it comes to the Internet of Things (IoT), a building’s potential energy efficiency relies on the network created by sensors and building devices paired with data storage and computing capabilities. In this regard, energy management information system (EMIS) software tools, designed to store, analyze, and visualize data generated by the IoT, have continuously shown potential for better system controls and a deeper understanding of how a building is functioning.
The IoT and EMIS have a promising role in reshaping the approach to evaluating energy-conservation measures (ECM). Whether in planning, measuring, collecting, or analyzing data, the enhanced visibility and understanding of building systems provided by an IoT platform can significantly revamp the industry’s traditional approach of the measurement and verification (M&V) process.
Optimizing building management with EMIS
With the ability to collect near real-time data and analyze it with higher temporal and spatial resolution, an EMIS offers new prospects for the M&V process to become more cost-effective, quicker, and more immediate. For instance, having access to 15-minute or hourly data provides more immediate insight into the performance of an energy efficiency project. This continuous-monitoring capability can improve operation and maintenance (O&M) by facilitating a more proactive building-management approach. In addition, increased resolution in the data (hourly versus monthly or annual) and live trending of weather data enable continuing and automatic energy-savings calculation with improved accuracy.
Due to enhanced granularity in the data—in space and time—an EMIS can, for example, enable the end user to zoom in and review trends from a particular point mapped from a specific device variable (e.g., zone temperature for a specific variable air volume box). Similarly, users can zoom out and compare energy data across a portfolio of buildings. In addition, cross-referencing different data streams from various sources within or outside of the building (e.g., HVAC subsystem data, occupancy or operational trends, indoor environmental data, weather data, etc.) can support O&M while providing additional insight into the facility by identifying correlations between the building subsystems. This holistic approach also can be used to investigate chains of causation for suspect energy-consumption patterns, system behavior, or malfunctions, which provides a good support system for fault detection and diagnostics and energy analytics.
In the context of whole building M&V, estimating energy savings from ECMs usually involves using baseline models. A baseline model is a mathematical model that represents typical energy performance relative to independent variables. For whole building M&V, this baseline model is traditionally linear or piecewise linear. Assuming building energy use is dependent on demand for cooling and heating, the model draws a relationship between energy-consumption data from utility bills and independent variables, such as weather data (i.e., outside-air temperature). This baseline model is then projected into the reporting period (post-ECM) as a projected baseline that quantifies the energy that would have been consumed without the ECM.
By using higher-resolution data, cloud-based computing capabilities, and a data-analytics toolkit, an EMIS allows the use of linear models employing additional independent variables—such as weather data, occupancy schedules, operating schedules, etc.—or more complex nonlinear models created with machine learning algorithms.
While EMIS offerings foster innovation for the M&V process, they also raise some challenges. First, there is currently no standardized way for the baseline modeling process using EMIS. In addition, since EMIS tools offer various features and M&V processes differ due to specific details in projects, the industry could benefit from additional guidance in determining the suitability of a tool for a given project relative, for example, to cost and technical adequacy. Although an EMIS introduces a wide range of new analytical methods to quantify savings, there is no definite method to assess or to compare these EMIS tools. Lawrence Berkeley National Laboratory’s Energy Technologies Area group has been conducting research on this issue and has developed a framework to evaluate and compare proprietary EMIS tools for M&V, focusing on performance metrics and baseline model accuracy.
When it comes to leveraging IoT for energy efficiency projects, one of the biggest challenges is to figure out how to use the data that is available, and one of the best assets is to be able to turn this information into actionable events for better facility management.
Soazig Kaam is a data analyst in the operations department at BuildingIQ—a company that is bringing a suite of cloud-based energy-intelligence service to building markets. Soazig graduated with a master’s degree in architecture and building science from the University of California Berkeley. Prior to working for BuildingIQ, Soazig conducted research at the Center for the Built Environment where she developed her interest in the use of data-driven approaches to improve building management for energy savings and enhanced occupant comfort. Soazig also holds a masters’ degree from Arts et Métiers ParisTech with an emphasis on mechanical engineering and sustainability.