Find Y, with X variable with set of data that forms bell curve.

[friendoftheearth68]

Hello All!

I'm creating a PV system design tool (for my own use, but I'd will post it open source afterward for folks to use).

Here is a portion of the data I'm working with:

TABLE 2: Minimum Radiation
					Feb				Mar
	Jan
Rank	Angle	Solar Radiation		Rank	Angle	SR		Rank	Angle	SR
1	0	1.95		1	0	2.07		1	0	3.37
2	22	2.62		2	22	2.43		2	23	3.78
3	37	2.92		3	37	2.55		4	38	3.84
4	52	3.06		5	52	2.54		5	53	3.71
6	90	2.69		6	90	2.01		6	90	2.67
5	57	3.07		4	44	2.56		3	34	3.84
1	0	1.95		1	0	2.07		1	0	3.37
2	22	2.62		2	22	2.43		2	23	3.78
3	37	2.92		3	37	2.55		3	34	3.84
4	52	3.06		4	44	2.56		4	38	3.84
5	57	3.07		5	52	2.54		5	53	3.71
6	90	2.69		6	90	2.01		6	90	2.67

<tbody>
</tbody>

The "Angle" is the angle of the solar panel from the horizontal plain, facing in the direction of the equator, and the "Solar Radiation" numbers are (in very crude terms) how many hours/day I can expect my modules to produce power. (So for example, if I had a 1.5 kWh array set at a 52 degree angle, I can expect to produce about 4.59 kWh/day on average in January at that site.)

What I would like to be able to do is to enter an angle and receive an accurate prediction of what the solar radiation number would be. (BTW, notice the peak--or optimum angle--changes from month to month.)

If I could do this with a formula, it would make everything soooo much easier.

---

I should add that TREND and FORECAST don't seem to work because they are linear (? i think).