PV Calibration issues

[RangeyRover]

At the moment I have PV calibration turned off.
This is because it is overestimating by a factor of 2.
here are some of my log entries.

50423 | 2025-11-03 10:15:04.629610: Historical days now [7] weight [1]
50422 | 2025-11-03 10:15:04.629129: Historical data totals for days [7] are [28.35] - min 28.35
50421 | 2025-11-03 10:15:04.584388: PV Forecast for day d6 is 22.97 (7.21 10% 26.63 90% 22.97 CL) kWh
50420 | 2025-11-03 10:15:04.545867: PV Forecast for day d5 is 18.27 (3.37 10% 26.38 90% 18.27 CL) kWh
50419 | 2025-11-03 10:15:04.506768: PV Forecast for day d4 is 18.37 (3.74 10% 24.86 90% 18.37 CL) kWh
50418 | 2025-11-03 10:15:04.467478: PV Forecast for day d3 is 9.57 (4.34 10% 15.29 90% 9.57 CL) kWh
50417 | 2025-11-03 10:15:04.428199: PV Forecast for day d2 is 21.91 (12.89 10% 26.67 90% 21.91 CL) kWh
50416 | 2025-11-03 10:15:04.386260: PV Forecast for day tomorrow is 16.5 (4.12 10% 26.62 90% 21.37 CL) kWh
50415 | 2025-11-03 10:15:04.275460: PV Forecast for today is 14.84 (4.21 10% 25.09 90% 19.02 calib) kWh and left today is 11.39 (3.0 10% 20.66 90% 15.05 calib) kWh
50414 | 2025-11-03 10:15:04.200779: PV Calibration: PV production: 51.85 kWh, Total forecast: 39.15 kWh adjustment 1.3244x slot adjustments {0: 1.0, 30: 1.0, 60: 1.0, 90: 1.0, 120: 1.0, 150: 1.0, 180: 1.0, 210: 1.0, 240: 1.0, 270: 1.0, 300: 1.0, 330: 1.0, 360: 1.0, 390: 1.7666666666666666, 420: 1.02197265625, 450: 1.1353868194842407, 480: 1.2061028952301602, 510: 1.2748550884144105, 540: 1.078965438315404, 570: 1.066756315435597, 600: 1.2014179240138156, 630: 1.1669485490475344, 660: 1.2229280226619168, 690: 1.305023349190416, 720: 1.209759292615032, 750: 1.3046949723458228, 780: 1.26130631785928, 810: 1.4335667494917552, 840: 1.4730736178745347, 870: 1.6737937163469079, 900: 1.4817239779347322, 930: 1.316227997572146, 960: 1.4281302978929522, 990: 1.6588898457684176, 1020: 1.5856980703745744, 1050: 0.8448438978240302, 1080: 0.5137495956001294, 1110: 0.5647096983457671, 1140: 0.9954075774971297, 1170: 1.0, 1200: 1.0, 1230: 1.0, 1260: 1.0, 1290: 1.0, 1320: 1.0, 1350: 1.0, 1380: 1.0, 1410: 1.0}
50413 | 2025-11-03 10:15:04.200591: PV slot 1410: production 0.06 kWh, forecast 0.0 kWh, distribution 0.12%, forecast distribution 0.0% slot adjustment 1.0x

Note the PV production: 51.85 kWh which is way beyond what my system will do.

This is the output of the solar power sensor, which is sane and hourly based.
My hypothesis is there is a mismatch with predbat 30 min base for forecasting and the hourly base for the sensor.

This is the last week solar max each day, never getting beyond 34Kwh
I dont want to log as an issue if I've got a configuration wrong.
Any ideas folks?

[gcoan]

This is the output of the solar power sensor, which is sane and hourly based.
My hypothesis is there is a mismatch with predbat 30 min base for forecasting and the hourly base for the sensor.

This is the last week solar max each day, never getting beyond 34Kwh
I dont want to log as an issue if I've got a configuration wrong.

Your solar forecast looks sensible.

Hard to tell if its a configuration issue without details of your configuration. How you are getting your solar forecast, your apps.yaml, etc

I'm not using the predbat calibrated forecast either, I'm using solcast integration auto dampening which is more sophisticated (and I believe accurate), but I get sensible figures from the predbat calibration:

34684	2025-11-03 01:10:02.635493: PV Forecast for today is 9.76 (3.46 10% 19.28 90% 6.77 calib) kWh and left today is 9.76 (3.46 10% 19.28 90% 6.77 calib) kWh
34683	2025-11-03 01:10:02.620805: PV Calibration: PV production: 21.93 kWh, Total forecast: 31.04 kWh adjustment 0.7067x slot adjustments {0: 1.0, 30: 1.0, 60: 1.0, 90: 1.0, 120: 1.0, 150: 1.0, 180: 1.0, 210: 1.0, 240: 1.0, 270: 1.0, 300: 1.0, 330: 1.0, 360: 1.0, 390: 1.0, 420: 0.5, 450: 0.741908827985126, 480: 0.7970503891847603, 510: 0.9229278303500171, 540: 0.8053283477448002, 570: 0.7810685738325621, 600: 0.9170212765957446, 630: 0.8059508408796896, 660: 0.831501677852349, 690: 0.6661151815961553, 720: 0.5787592342521249, 750: 0.5, 780: 0.5682417168674699, 810: 0.5114884555032504, 840: 0.6557560497971744, 870: 0.6194863919460797, 900: 0.748286203941731, 930: 0.6792285916489738, 960: 0.5, 990: 1.4726368159203982, 1020: 1.0, 1050: 1.0, 1080: 1.0, 1110: 1.0, 1140: 1.0, 1170: 1.0, 1200: 1.0, 1230: 1.0, 1260: 1.0, 1290: 1.0, 1320: 1.0, 1350: 1.0, 1380: 1.0, 1410: 1.0}
34682	2025-11-03 01:10:02.620624: PV Calibration: Worst day scaling factor 0.5 best day scaling factor 1.0

[RangeyRover]

I think there is something odd
34683 2025-11-03 01:10:02.620805: PV Calibration: PV production: 21.93 kWh, Total forecast: 31.04 kWh adjustment 0.7067x slot adjustments {0: 1.0, 30: 1.0, 60: 1.0, 90: 1.0, 120: 1.0, 150: 1.0, 180: 1.0, 210: 1.0, 240: 1.0, 270: 1.0, 300: 1.0, 330: 1.0, 360: 1.0, 390: 1.0, 420: 0.5, 450: 0.741908827985126, 480: 0.7970503891847603, 510: 0.9229278303500171, 540: 0.8053283477448002, 570: 0.7810685738325621, 600: 0.9170212765957446, 630: 0.8059508408796896, 660: 0.831501677852349, 690: 0.6661151815961553, 720: 0.5787592342521249, 750: 0.5, 780: 0.5682417168674699, 810: 0.5114884555032504, 840: 0.6557560497971744, 870: 0.6194863919460797, 900: 0.748286203941731, 930: 0.6792285916489738, 960: 0.5, 990: 1.4726368159203982, 1020: 1.0, 1050: 1.0, 1080: 1.0, 1110: 1.0, 1140: 1.0, 1170: 1.0, 1200: 1.0, 1230: 1.0, 1260: 1.0, 1290: 1.0, 1320: 1.0, 1350: 1.0, 1380: 1.0, 1410: 1.0}
50414 | 2025-11-03 10:15:04.200779: PV Calibration: PV production: 51.85 kWh, Total forecast: 39.15 kWh adjustment 1.3244x slot adjustments {0: 1.0, 30: 1.0, 60: 1.0, 90: 1.0, 120: 1.0, 150: 1.0, 180: 1.0, 210: 1.0, 240: 1.0, 270: 1.0, 300: 1.0, 330: 1.0, 360: 1.0, 390: 1.7666666666666666, 420: 1.02197265625, 450: 1.1353868194842407, 480: 1.2061028952301602, 510: 1.2748550884144105, 540: 1.078965438315404, 570: 1.066756315435597, 600: 1.2014179240138156, 630: 1.1669485490475344, 660: 1.2229280226619168, 690: 1.305023349190416, 720: 1.209759292615032, 750: 1.3046949723458228, 780: 1.26130631785928, 810: 1.4335667494917552, 840: 1.4730736178745347, 870: 1.6737937163469079, 900: 1.4817239779347322, 930: 1.316227997572146, 960: 1.4281302978929522, 990: 1.6588898457684176, 1020: 1.5856980703745744, 1050: 0.8448438978240302, 1080: 0.5137495956001294, 1110: 0.5647096983457671, 1140: 0.9954075774971297, 1170: 1.0, 1200: 1.0, 1230: 1.0, 1260: 1.0, 1290: 1.0, 1320: 1.0, 1350: 1.0, 1380: 1.0, 1410: 1.0}

There should not be solar input 24 hours a day. I think the 1.0 values are all errors?

[gcoan]

I think there is something odd

There should not be solar input 24 hours a day. I think the 1.0 values are all errors?

the logfile is not saying that

the PV calibration is calculating adjustment factors to apply to your solar forecast based upon solar generation history. 1.0 of a solar forecast of zero is still zero. Mine shows the same 1.0 factors

The PV calibration is uplifting your solar forecast based upon past history (hence the factors greater than 1.0 - all mine are less than 1), either suggesting your solar forecast is incorrectly too low or your past history is incorrectly too high or predbat can't add up. We should look at both of these entities first