How interval data can help optimize building start time
Energy interval data provides a precise view of building behavior over time. While some data patterns require statistical processing to detect, others are easily visible to the naked eye. Building start-up is one of the easiest processes to finetune using little more than a chart of your energy load curve.
We reviewed a year’s worth of start-up data for several hundred commercial buildings in California. The resulting set of benchmark data suggests that building start-up is a solid source of savings for buildings, and perhaps more importantly a diagnostic of how teams respond to tenant complaints.
Pictured is a simple example that shows a typical commercial building starting at 4:45 am. Other buildings have more complex start-up procedures. For example, a lot of buildings exhibit a two-phase start-up as the a chiller plant responds to increasing temperatures. In our simple analysis, we focus on initial start-up only, which yielded some interesting findings.
- Early birds! While the most common building start time is 6:00 am, some buildings start as early as 1:00 am, and a full 25% have started by 4:45 am, well in advance of occupancy. An extra daily hour of 100kW load costs about $1,500 per year in California, indicating the possibility of easy savings for some of these early birds.
- Set and forget. Building start time generally doesn’t vary by day of week or by season. For most buildings it is a fixed item in the schedule. Fewer than 2% of our buildings vary start time by day of week, and fewer than 5% vary start time by season.
- Peaking and start-up. We usually associate peak demand with cooling load, driven by warm temperatures in the mid-afternoon. A significant portion of buildings actually peak first thing in the morning, when systems come online and drive a surge in demand. Given high (and increasing) demand charges, paying attention to start-up sequences with demand charges in mind can pay big dividends.
Building startup is also a good illustration of one of the most important concepts in operational energy efficiency — matching schedules to typical use, not worst-case use. Consider the example below, a Los Angeles building with excellent weekend setback. The building starts earlier on Mondays, because of the additional cooling load required after weekends with the chillers turned off. This start-up sequence makes perfect sense during the summer, but the load curves below are from 70-degree winter day. Does this building really need to start at 3 am on winter Mondays?
Often operators react to complaints on extreme days by appropriately adjusting settings to address comfort issues. What they don’t always consider is the effect on energy consumption on non-extreme days. In this example, you’re paying for a hot August weekend on the other 51 Mondays of the year. How many of them could be started at 6am instead?
Our recent webinar discussed the topic in more detail, including more real world building examples and a deeper discussion of controls strategy.