What is the best way to measure the maximum power generated by a solar panel?
Assuming the current/voltage relationship is linear (it's not, but this gives you a crude lower bound), you could measure the short-circuit current and the open-cell voltage and do
1/4 * I * V to obtain the maximum theoretical power given a worst-case 0.25 fill factor. However a more reasonable value might be obtained by using a different factor
Typical fill factors range from 0.5 to 0.82. (source)
Because the I-V curve is non-linear you either need to find the parameters of your cell and apply equations (such as found here), or simply do it numerically by taking measurements at various I/V and finding the largest
The Solar Cell is just another diode which generates free electrons when light falls on it. You've got to have a light source whose light intensity (irradiance) can be varied. Direct the light on the cell, vary the intensity and start measuring the voltage and current at that voltage. Plot them on a graph paper.You are bound to get the I-V characteristics of the cell. At a certain point Imp and Vmp co-inside, this point is the max Power (Pmp).
Since I'm new here I'm not able post pics. You can just Google you will get the I-V curve of solar cell.
If you are too lazy to do this then buy a flasher (will cost you few thousand dollars) :) and you will get all parameters of the solar cell on a single sheet of paper.
First: the solar panel has a V/I curve which is shaped like this:
As you can see, for low currents the voltage varies slightly, and for low voltages the current is almost constant. So you will have the maximum current when the panel is short-circuited, and the maximum voltage when the panel is open-circuited. But, since the power is V*I, you have to find the point in which their product is maximum.
As you can see in the second picture, the power has a peak at abour the 80% of the open circuit voltage, and some Max Power Point Tracking (MPPT) systems will just bias the panel at that voltage. But smarter ones, will use tricks like measuring the slope of the power curve, which is zero when the MPP is reached.
The picture shows also another interesting fact: changing the light, the curve scales vertically, which means that the current will be proportional to light, but the voltage will remain about the same (with rough approximation).
Increasing the temperature, instead, will lower the open circuit voltage, shifting the curve leftwards and thus reducing the total amount of power that it's possible to obtain. That's why it's important to keep the panel cool.
So, to find the maximum power point, you can do this way:
First, fix the light level to a stable (and possibly known) value;
Then, connect the panel to a voltage source capable of sinking current and measuring both voltage and current accurately;
Starting from 0 V, increase the voltage at which you bias the panel until you reach the open circuit voltage Voc, while measuring the current for each point.
To find the MPP, you have several ways:
measuring the open circuit voltage Voc or the short circuit current Isc. The MPP is usually an almost constant fraction of these parameters. This is the roughest method, because it's not accurate and will need adjustments when the light changes.
derivative: when the panel is biased to a point, introduce a small change in the voltage of the panel, and measure the change in the current. Depending on this ratio (dI/dV) it's possible to adjust again until reaching the MPP. There are several different implementations of this technique, which is the most common.
other solutions may use an external light meter, and derive from it the right biasing voltage for the panel.