In hopes of brightening your bleak mid-winter, we offer a few nuggets of weather wisdom from our latest Energy Leader webinar. Here are Steve Heinz's written responses to chat questions submitted by attendees during the session,
which has been recorded and is available online:
Why not throw another Yule log on the fire and settle back in your chair for an interesting read about energy and weather?
1. QUESTION: How does night and holiday thermostat setback affect Heating Degree Day (HDD) and Cooling Degree Day (CDD) calculations?
ANSWER (SH): The simple answer is that HDD and CDD are an expression of outdoor drybulb temperature hence nothing related to the building structure or internal operations has any impact. For example, NOAA has been calculating and posting DD data for 80+ years and the data is independent of any consideration of building type, age, internal equipment, etc.
However a deeper answer is a bit different. The HDD/CDD that a building “feels” at the meter is related to thermostat settings and building envelope/equipment. Consider a building that due to a combination of factors needs no heat until the outside temperature is 55F. We should use a balance point of 55F for that building. If we use NOAA 65F degree day data, that data suggests that on a 56F day the building has used heat because there were 9 HDD, but in reality the building used no heat, so this “mismatch” (predicted vs. actual) will affect the accuracy of the statistical model.
Thermostat setbacks, in the same manner as improvements to a building’s windows, roof/wall insulation, and HVAC equipment will have the net effect of potentially lowering the HDD balance point and raising the CDD balance point. In fact you could say that the purpose of any energy conservation measure is to weaken the correlation between extreme weather and additional energy use.
2. QUESTION: Is there validity in modeling hourly interval data rather than monthly utility bills?
ANSWER (SH): Sure, the more precise and voluminous the data, the more confidence you will probably have in the statistical model. Using hourly interval data would allow you to create multiple models; for example one model for occupied days and another for unoccupied days. But there is a price to pay for the added accuracy—and that is cost. Hourly weather data will probably have a cost, whether purchased from a provider or via your own weather station hardware. And the analysis will take additional high-skill-level time—time that could be spent on other energy management activities. So you have to weigh accuracy and cost with benefit.
3. QUESTION: Is it better to use average unit cost (“blended” price per KWH) or the actual rate schedule (separate energy and demand charges) when calculating the cost avoidance attributable to an energy conservation measure?
ANSWER (SH): Valuation methodologies were not discussed in the webinar, but this is a good question nonetheless. As M&V analysts quickly learn when addressing this issue, once you step off of average unit cost (“AUC”), you’re in the deep end of the pool. M&V cost and complexity can increase 10x, so you have to consider the cost and benefits. Let’s consider a typical time of use electric rate.
Option 1 is simple AUC on the entire bill. Option 2 is to value the savings in each rate component separately. To do so you need to treat on-peak, mid-peak, off-peak KWH separately because each will have different use vs. weather statistical models. Your level of effort has increased 3x already, keeping in mind that any adjustments for floor area changes, occupancy, facility scheduling, other retrofit projects, etc now require three separate adjustments.
And since KW demand is a cost component, you now have to also value KW reductions. This is very tricky for a number of reasons, not the least is weather adjustment which can’t be done using DD or average temperature, it requires correlation with max daily temp, taking unoccupied days into account. You may have on-peak, mid-peak and off-peak KW charges. And since this question was from a K-12 Energy Manager, assuming energy conservation measures in many school buildings, you may have multiple rates and dozens of electric meters, all requiring ongoing analysis.
You can see why the IPMVP organization cautions us to not invest in “extra” M&V when the cost to do so is not outweighed by additional savings benefit. In EnergyCAP, we use AUC but we caution our users that when an energy conservation measure affects an electric meter on a rate that has a substantial KW cost component, be sure that KW reduction is commensurate with KWH reduction. If “balanced” then AUC is a fair approximation. If unbalanced, then adjust the AUC (i.e. stipulate your own adjusted AUC value) to a value that is appropriate for the situation.
And remember that many ECMs will save more KWH during expensive on-peak periods than during off-peak, so the use of AUC may actually undervalue the cost avoidance. Don’t assume that AUC will always overvalue cost avoidance; that’s becoming less common as we see more TOU rates with extremely high on-peak charges and as ECMs become more sophisticated in reducing electricity usage at those times.
This is our last scheduled blog of the year. Thanks to all our loyal subscribers for your time, support, comments and suggestions in 2015. Merry Christmas and Happy New Year!