Calculations on heat cost can be done in many ways

Calculations on heat cost can be done in different ways. For example, based on yearly COP figures at location from a vendor or test institute, or based on quarterly or monthly figures. Some do not do them at all, for example, when data is missing. There are handbooks that recommend solutions based on "defining outside temperature" (DUT) and DayDegrees and other similar rough methods.

Whatever your choice, the choice is a compromise. If calculations are too rough, they will miss a lot of factors, for example:

• Be based on other temperatures than exist on your location
• Be based on other needs than those of your house
• Be missing out such things as days below heat pump limits
• Be missing out such things as the differences in COP curves
• Bew missing out such things as days needing additional heating

On the other hand, if the calculations are too complicated, then it will be hard both to find the data and to make them.

The best compromise I have found is using day-to-day values

A lot of data is easily available based on temperatures. And a lot of data, like, for example, temperatures, are available on a day-to-day basis. Also, today's programs, like databases or spreadsheets, easily handle thousands of rows of temperatures in no time.

Thus, I find that with the data available and a simple database, the easiest way to do and feel sure about what affects the calculations is by doing it on a day-to-day basis. Summing all days for the time period desired for the decision (months, quarters, year, decade, century, ...). So using the below:

• Daily average temperatures for my location the last years.
• The COP curve for the heat pumps I am comparing, defined by outside temperature
• The energy needed by the house, defined by outside temperature
• A single summary statement of the above in a database

I could just as well have put the data into Excel, it makes no big difference using Excel or a database.

The important selection is what approach to use, what to leave out and what assumptions to make. The interesting thing is that both a rougher and a more precise method quickly lead to problems. The most common problem is finding data suitable enough to calculate your need and result, instead of some average need. For example, by only being able to compare a few heat pumps tested by some institute, or only being able to compare what would happen in a certain location and time for which data is made available.