Forecasting Building Energy Usage

Experience vs Analytics vs Simulations

When looking for ways to reduce your operating expenses, there are many technology and engineering companies who can help your buildings become more energy efficient, each with their own process and technique. At Bractlet, we often get asked about our physics-based energy simulations and how these virtual buildings can forecast energy usage with unmatched accuracy. So I thought I'd briefly describe each of the ways building energy efficiency analysis is done and how our way works.

First there are traditional engineering methods, which rely heavily on an individual's experience. There are great engineers out there who have seen numerous buildings and building systems; by just walking through a building, they can often find ways to save some energy. These engineers will traditionally use self-designed excel spreadsheets, as well as vendor-specific calculation tools, to provide the customer with a rough energy savings estimate based on the efficiency ratings of new equipment and/or the savings they've experienced in the past with similar projects. In light of the increasing system complexity in modern buildings and the sheer volume of data available (from Building Management Systems (BMS), hourly interval utility, sub-meters, etc.), these methods are arguably limited in the number of savings measures found, amount of savings achieved, and ability to accurately forecast the savings. 

Second are companies that do building analytics, which relies on gathering lots of historical data (typically from a BMS) and employing statistical techniques to “mine” the data. Regardless of the type of statistical analysis used, the goals are to better understand how the building is currently operating and identify any operations-based savings opportunities. Predictions about the buildings future energy usage are usually nowhere to be seen and, if they are there, are usually calculated using the traditional engineering methods described above. Data analytics can provide some good insight and Bractlet uses these methods to fully understand building operations on every project. In our experience, getting good insight from building analytics requires data from each season (an entire year) and is only useful for understanding current building operation, not what it could be in the future. More importantly, data analytics methods struggle to understanding how different building systems interact as a whole, and how these systems work together to provide comfortable work spaces.

Once developed, a Building Energy Simulation (BES) model becomes a virtual testbed for all types of energy savings strategies.

And finally, there are Building Energy Simulations (BES). A BES is a digital representation of an actual building, with a focus on energy-consuming systems (i.e. HVAC, lighting, domestic hot water) and building utilization (i.e. occupancy, equipment, BMS controls). Once developed, the simulation becomes a virtual testbed for all types of energy savings strategies. Replacement of existing equipment, building controls optimization, strategies for peak-demand reduction - all and more can be rapidly assessed with a BES. Since the BES is built with the fundamental heat and mass transfer equations, it can take into account interactive effects that occur when optimizing or changing systems and equipment. Unlike data analytics, Bractlet’s simulations are truly predictive and, unlike traditional engineering methods, Bractlet’s simulations are comprehensive with respect to system interactions within a data-rich building environment. 

In order to find the maximum amount of savings options with the highest forecast accuracy, Bractlet actually utilizes the strengths from all three of these techniques. We leverage analytics and software to build a virtual building simulation. Equipped with a detailed database of energy savings measures (ESM), the simulation can quickly guide optimization efforts and illustrate how the implementation of ESMs will affect building performance. 

I’ll be going into more details of what makes our BES models so accurate in a future post - please stay tuned and feel free to drop me a line at nick.cardwell@bractlet.com if you have any questions.

Nick Cardwell is the Director of Engineering at Bractlet and is an expert in heat/mass transfer and energy simulations. He received a PhD and MS in Mechanical Engineering from Virginia Tech and a BS in Mechanical Engineering from the University of Texas at Tyler.