Change n_cycles for shower

[bramvdh91]

Checking Jordan & Vajens, I noticed the number of cycles for a shower should not be 73 cycles per year, but 2/day * 365 days/year i.e. 730 cycles/year.

[bramvdh91]

I've compared the results with the changed n_cycles, but the average water consumption stays around 128 l/day, so no real difference. Not sure why the average water demand is not influenced.

[cprotopa]

I believe the parameter cal is used in the simulation of these loads, and not cycle_n directly. cal in turn is defined based on a calibration process using __calibrate__.py. The latter simulates many households, and then updates cal based on the ratio of resulting number of cycles and cycle_n, per load. I would imagine if the number of cycles changes so much, many calibration rounds would be needed to have some kind of conversion. Perhaps it also influences other water demand loads, but I'm not sure. I'm afraid to see the results after this change...

[cprotopa]

I'm running the calibration to fix this cal value. I had to change the __calibrate__.py model actually, cause it didn't include the flow types of use, for some reason... I'll include a commit about that also.
However I notice during calibration with this new number of shower cycles (730/year), that DHW demand increases too much. It's risen to around 130 l per person per day, and that's hot water only. This raises some issues, of course, as it's hard to pinpoint the exact cause of this high demand... The parameters for the other flows are following Jordan & Vajens correctly, for a supposed 200l/day household. Has anybody looked deeper into this? Any ideas, @bramvdh91 @SilkeVerbruggen ?

[SilkeVerbruggen]

When I checked the domestic hot water data, it seemed to be overestimating the hot water flows, especially for the bath.

[SilkeVerbruggen]

I've no idea about the calibration. I've never checked that.

[cprotopa]

After a long time, coming back to this starts to make more sense. Based on an average 200 l/day household, and given that Strobe produces more than 40% single-person households, it makes sense that the average l/day/person is around 130. This assumes the water is the one asked at the tap, thus some 38 degrees C, not 60C (as I believe was wrongly mentioned).
Of course, what would be better is to change the implementation to vary per number of occupants. I'm not sure how this can be done though.

[SilkeVerbruggen]

If the hot water is at 38°C instead of 60°C the values seem more logical indeed. In my adaptation different calibration values are used for households with 1,2,3 or 4+ persons. So instead of having one calibration factor based on the number of cycles of an average household, I have 4 different ones depending on the size of the household.