First of all you have to be sure that your measured Voltage versus current data exhibits rectification contact properties (current dependency of voltage should be single exponential behavior). Then you will be able to find out diode factor.

Basically you may list your data table as voltage versus current data then you can calculate the natural logarithm for every current values for each data point. Then you will be able to plot Voltage versus Ln I graph. V-Ln I graph generally should exhibit linear behavior. Because it is semi-log plot scale. Using linear region of your plot, you may calculate slope of linear graph this will tell you slope = q / (nkT) coefficient. If you know the operation temperature (T=?) it is straightforward to calculate “n”. Except above explanation. There is a huge collection in the literature however; these techniques are more complicated since they are for multi exponential form I-V behavior.

The FF could be calculated as follow FF = (Jsc*Voc)/(Jmmp*Vmmp) where mmp is the maximum power point

Consequently the efficiency could be calculated as follow: η = FF*Jsc*Voc/Pin where Pin is the incident power of the lamp used as sun simlator

hope this paper helps you to find what you are looking for

http://journals.tubitak.gov.tr/physics/issues/fiz-07-31-1/fiz-31-1-2-0609-3.pdf

http://www.chalcogen.ro/933_Alfaramawi.pdf

please reply with your feed back

 You can calculate diode ideality factor  (A). Jo ,Rs,Rsh of the device by fitting the J-V data with diode equation.  J(V)= Jo [exp(q(V-RS J)/(AkB T))-1] +(V-RS J)/RSh - JPh (V); There are some steps, you can get idea from any of device analysis paper.

These papers could be  helpful to you

S. S. Hegedus and W. N. Shafarman, Prog. Photovoltaics, 2004, 12, 155–176

http://pubs.acs.org/doi/abs/10.1021/acs.jpcc.5b03193

http://pubs.rsc.org/is/content/articlehtml/2016/ra/c6ra02291j

All the best

First of all you have to be sure that your measured Voltage versus current data exhibits rectification contact properties (current dependency of voltage should be single exponential behavior). Then you will be able to find out diode factor.

Basically you may list your data table as voltage versus current data then you can calculate the natural logarithm for every current values for each data point. Then you will be able to plot Voltage versus Ln I graph. V-Ln I graph generally should exhibit linear behavior. Because it is semi-log plot scale. Using linear region of your plot, you may calculate slope of linear graph this will tell you slope = q / (nkT) coefficient. If you know the operation temperature (T=?) it is straightforward to calculate “n”. Except above explanation. There is a huge collection in the literature however; these techniques are more complicated since they are for multi exponential form I-V behavior.

Thank you @Abdulkadir, your answer just solve my problem, but after few experiments ,I found out just use the Isc and Voc from the data to get the efficiency=Isc*Voc/Pinput

which is more easy and straightforward.

I would like to make some correcting and extending remarks:

1.) The fill factor (FF) is, naturally, defined the other way round: FF = (I_mpp * V_mpp) / (I_sc * V_oc); only then its value is below 1 (which makes physically sense).

2.) The efficiency is not calculated from I_sc * V_oc but from I_mpp * V_mpp, since only the latter gives the usable power output of the cell.

Be aware that for determining an "official" value for the cell efficiency, one has to take spectral mismatch into account, since the "official" efficiency values refer to the standard solar spectrum which usually is not matched by the lab equipment.

3.) In case that you deal with a large-area solar cell, for an interpretation of the ideality factor you should have a look at the following paper from MPI Halle (both appeared 2013): "Evaluation of luminescence images of solar cells for injection-level dependent lifetimes" and "A two-diode model regarding the distributed series resistance" (you can download them from http://www-old.mpi-halle.mpg.de/mpi/publi/go/2013/2013.html).

In my previous post I referred to the 2013 publication "A two-diode model regarding the distributed series resistance" (http://www-old.mpi-halle.mpg.de/mpi/publi/go/2013/abs_pdf/11094_13.html). Meanwhile, however, these results have been revised in the follwing 2017 publication, giving a significantly improved interpretation of the measured data: Article Distributed series resistance in a one-dimensional two-diode...