In the last post, we saw how valuable utility data can be in terms of documenting improvements we make to our facilities and in ensuring they persist. But utility consumption patterns also can provide clues regarding opportunities, especially when combined with some of the other clues we have been talking about.
The General Case
Ultimately, the consumption patterns associated with the dysfunction of a system (and any improvements that are made) will show up at the bottom line; i.e the utility meter. Facilities with simultaneous heating and cooling loads will tend to show high baseline consumption patterns (the area below the “valley” in the consumption curve) when compared to other facilities, all other things being equal.
Note that for this Oregon facility (read “mild climate”), the minimum gas consumption in the summer on an average day is 50% of the consumption on an average day during the coldest month of the year.
If you look at the electrical consumption pattern for this facility, you see something like this.
The electrical graph implies that if you could totally shut down the chiller plant (the presumed cause for the peak in consumption during the warmer months), you would only have a modest impact on the electrical costs for the facility.
Hmmm; I wonder if there are things running round the clock that could be scheduled?
If you were to use a data logger to spot check plug loads and use equipment schedules and operating hours to estimate how much power things like lights and fans and pumps use, you might discover that you can’t account for where a significant portion of the electricity goes (the red band in the graph below).
If you convert the summertime therms in the gas graph (i.e. the base load gas consumption) to kWh, the number you get is very similar to the unknown kWh in the electric graph.
Hmmm; I wonder if there is some simultaneous heating and cooling going on?
For the facility that this data came from, it turned out that there were a lot of things running round the clock that could be scheduled. And it was being ventilated for 6,000 – 8,000 people but the actual census was more like 2,000 people. If you want to know a bit more about this particular project, you can read about it in Tudi Haasl’s paper titled Retrocommissioning’s Greatest Hits (it’s the high tech facility).
One caveat here; the peaks and valley’s need to be taken in context. For instance, if your only load was 1,000 watt light bulb that ran 24/7, then you would get a straight line at 24 kWh per average day across your graph. If you full scale on the graph was 25 kWh, it would like like a huge base load. But if you plotted it as a separate line on the graphs above, you might not even see it.
Our Target Facility’s Case
For our Golden Colorado hotel, the electrical consumption shows a significant baseline and a fairly consistent pattern, including a seasonal peak that is about 20% above the base consumption pattern.
Patterns like this tend to make me want to look for scheduling opportunities, and oversized pumps more than they make me want to look at ways to optimize the central plant. That said, the facility likely operates the central plant on a year round basis, so a part of the baseline could be related to the efficiency of the central plant at low loads.
The gas consumption patterns show an interesting trend towards an increasing baseline consumption.
There could be a lot of things that could cause this, like higher occupancy during the summer months, colder summers, improved guest satisfaction scores (maybe they shut down reheat during the summer in 2010, and then realized they needed reheat to keep folks comfortable so they started to keep the reheat system on line in 2011 and 2012), or loss of calibration in the control system. Plotting things like average occupancy and heating and cooling degree days on the graph may provide some insight into what is driving the consumption pattern, both for electricity and for gas.
More Information and a Few Tools
The bottom line on this post is that utility data can point you towards targets for your retrocommissioning efforts, especially energy intensive targets like simultaneous heating and cooling processes. If you want to know more about how to apply the utility assessment techniques behind the graphs, there is a paper about that too, along with some tools to help you do it on the California Commissioning Collaborative web site. The graphs for our Golden Colorado hotel were generated in a matter of minutes using the UCA tool from the CACx website. We can thank Bill Koran for that and a lot of the other spreadsheet tools like ECAM that you see if you follow the link associated with Bill’s name.
Senior Engineer – Facility Dynamics Engineering