Power consumption of a heat pump

The peaks are clearly the heating rod, the wide blocks are the heat pump or household electricity. It is highly unlikely that the outside temperature drops every night exactly at 4:00 AM to the point that the heating needs to turn on. During the day, the blocks are distributed very irregularly and the household electricity is typical for the daily pattern. The wide block at night does not fit into this pattern. The heat pump requires 5 kW, the heating rod 6 kW, and there are no other large consumers. There is no night setback. Since the solar system was installed, we only have one electricity tariff without time-based switching.

Then it is a night reduction/night shutdown. The heating rod doesn't seem to always start at 4 either, but the heat pump does. Do you have your user manual as a PDF? Try searching it with keywords like "Nachtabschaltung", "Nachtabsenkung", "Nachtprogramm", "Zeitplaner", "Wochenplaner", "geplante Steuerung", etc...

From 4 o'clock onwards, those are exactly the 5 kWh that the heat pump has, right? Since there is no block overnight, there actually has to be a night shutdown that lasts until around 4 o'clock. After that, the pump starts and first catches up. That's why this is also the largest block. It is interesting that the blocks are always wedge-shaped. The power consumption increases the longer the heating runs, the closer one gets to the end of the cycle.

Once again: There is no night switch-off because it makes little sense for underfloor heating.

Then the heat pump is maybe just normally active. For hot water or for heating. In the graphs, it is also clearly visible that it is not always 4:00, but distributed over the day and sometimes at night as well. What kind of heat pump is installed exactly?

That does not seem plausible, because then these heat pump plateaus would also exist during the period from 10 PM to 4 AM. However, according to your initially shown diagrams, they do not seem to exist (only the short peaks). The fact that the target temperature was not undershot during that period, but several times during the day, is also implausible, since it is usually cooler at night than during the day. But, of course, the sample is small.

It may be that you do not want a night shutdown/reduction, but the data suggests that there is one. Therefore, go through the manual and control menu again. Sometimes there are several options with the same effect. Example: Night setback: scheduler that lowers the desired temperature (possibly flow setpoint) in the window or sets a different temperature. Example: shutdown: blocking time during which the heat pump simply does not run at all in that period, etc.

In den Grafiken ist ja auch deutlich zu erkennen, dass es eben nicht immer 4:00

But this plateau, which according to the OP is the heat pump in heating mode, is always at 4 o’clock. And between 10 PM and 4 AM it never occurs. The short peaks are supposedly hot water.