Ok, so part of the problem is that your points aren’t in order
| a | b |
|---|---|
| 0.0095 | 0 |
| 0.49 | 410 |
| 0.63 | 530 |
| 2.67 | 2230 |
| 3.05 | 2620 |
| 3.29 | 2850 |
| 3.81 | 3380 |
| 4.23 | 3770 |
| 4.61 | 4120 |
| 4.9 | 4420 |
| 5.07 | 4610 |
| 5.11 | 4670 |
| 5.19 | 4780 |
| 5.25 | 4850 |
| 5.28 | 4970 |
| 5.31 | 5050 |
| 5.38 | 5210 |
| 5.42 | 5230 |
| 5.43 | 5220 |
| 5.43 | 5250 |
| 5.45 | 5300 |
| 5.48 | 5390 |
| 5.49 | 5330 |
| 5.49 | 5400 |
| 5.5 | 5420 |
| 5.52 | 5560 |
| 5.58 | 5630 |
| 5.65 | 5720 |
| 5.73 | 6050 |
| 6.094 | 8200 |
with 6th order
Thanks, I will give that a try.
Here’s the whole thing
compensation:
estimated_pv_panel_power:
source: sensor.small_solar_panel_voltage
unique_id: estimated_pv_panel_power
precision: 0
unit_of_measurement: W
degree: 6
data_points:
- [0.0095, 0 ]
- [0.49, 410 ]
- [0.63, 530 ]
- [2.67, 2230 ]
- [3.05, 2620 ]
- [3.29, 2850 ]
- [3.81, 3380 ]
- [4.23, 3770 ]
- [4.61, 4120 ]
- [4.9, 4420 ]
- [5.07, 4610 ]
- [5.11, 4670 ]
- [5.19, 4780 ]
- [5.25, 4850 ]
- [5.28, 4970 ]
- [5.31, 5050 ]
- [5.38, 5210 ]
- [5.42, 5230 ]
- [5.43, 5220 ]
- [5.43, 5250 ]
- [5.45, 5300 ]
- [5.48, 5390 ]
- [5.49, 5330 ]
- [5.49, 5400 ]
- [5.5, 5420 ]
- [5.52, 5560 ]
- [5.58, 5630 ]
- [5.65, 5720 ]
- [5.73, 6050 ]
- [6.094, 8200]
How did you select which data points to use, and the order?
I just sorted the points, deleted the ones that were way off from what you wanted, then chose a polynomial order that fit close to what you wanted. I used Excel.
It looks to be behaving. The output isn’t going wild. I will have to check tomorrow when there is sun light to see how well the estimator output tracks the actual PV array output.
Thanks for your help.
you may want to include more points (in order) between 0.63 and 2.67. Try to make them as linear as possible. right now you got a small bump in that area.
Is there a way to reload the compensation integration without restarting HA?
Not at this time, sorry.
1 Like
Capturing those early points should be easy. Thanks for the recommendation.