Calculating Chiller COP when Cycling

I'm calculating hourly COPs for a low-load air-cooled chiller. From what I understand (due to this post) there are two ways to do this:

COP Calc #1: $ COP,hr = Assigned Load / Power$

COp Calc #2: $ COP,hr = PLR / Corrected EIR$

The issue I am having is that when the chiller is operating for a fraction of the hour (meaning the load is lower than the minimum part load of the chiller), these calculations diverge and do not equal each other. When the fraction of the hour the chiller operates is 1, the calculations are equal, as in the last two rows. In the table below, I've used the equations above to do the calculations in the "COP Calc" columns on the right.

Why do these numbers diverge? Is it due to eQuest accounting for zeros in its average across the hour?

image description

This discrepancy is caused by the start-up and shut-down times inputs in the Miscellaneous tab of the chiller properties. When the defaults are left alone, the start-up time is 0.040 h, and the standby time is 0.025 h.

image description

When the chiller cycles there is additional electrical power associated with the startup and standby. This increases the electrical power value and means that COP Calc #2 above is no longer valid.

To verify this, I zeroed out the start-up time and the standby time. Once this is done, the electrical power decreases and COP Calc #1 equals COP Calc #2.

Though it is important not to zero out this value as it is a very real energy usage that affects the effective COP of the chiller plant.

image description