Problem solved. Even very few literature exists for this phenomenon, it 's called anti insulation effect, meaning for some range of cooling setpoints and insulation levels, adding insulation will cause an increase in cooling energy. It has been stated before but not extensively explored.

This is quite interesting - since adding insulation must decrease the heat loss, then some other effect is occurring. Is this due to wider temperature fluctuations due to the temperature control becoming slightly unstable? Increasing R value will increase the house thermal time constant, so if the feedback system is not set correctly (it would need to have several poles, a single dominant lag won't cause instability) then this phenomenon might occur. I would like to see some reference?

Dear prof

It dependes the number of occupation. If occupuation is increased cooloing loads are increased.

Dear Tony,

It's like with the increase of cooling set point as well as the level of insulation, there's an inflection in the curve that adding insulation turns out to increase the cooling energy. The reference is "A new and innovative look at anti-insulation behaviour in building energy consumption" written by O.T. Masoso and L.J. Grobler. Thanks a lot for your help!

Best,

Ray

I looked at the reference (on the web) and, basically, don't believe it is indicating what the authors state.

The set point of 26 deg is going to be close to the external temperature over much of the 24 hours, and the cooler will not be working effectively under these conditions. As you add more insulation the cooler should of course work less - but that also depends on its design and the control algorithm. I would like to be convinced that what you are seeing is not an issue with the cooler.

As with most heat engines, efficiencies will fall under low load - they like in general to be run quite hard to get good efficiency. So as the set point is increased, the energy expended on useful cooling may become less than the losses in the cooler. If this happens then I can imagine that the energy over a year might rise a little because you are basically supplying useless energy to the cooler that doesn't do anything effective with it.

But it is an interesting situation!

Dear Tony,

Thank you for your suggestion! Yes that also explains the phenomenon. I'll do a bit more research on this trying to figure out what is the exact reason causing this situation. It may also related to the site weather. Once I got some result I'll keep updating here.

Thank you!