NEW CURVE FITTING ERROR CRITERION FOR SOLAR CELL I-V CHARACTERISTICS.

Abstract

A commonly used method of parameter recovery by curve fitting minimizes sigma , which is defined as the rms of the difference between the experimental and theoretical current values. This method is demonstrated to be unreliable when it is used with characteristics which have been collected by linear A/D systems, or which have certain data point distributions. A more reliable minimization criterion, epsilon , is proposed. epsilon is based on the area between the experimental and theoretical curves. Computation experiments show that the use of epsilon results in much more accurate parameter recovery for both dark and illuminated characteristics, and that its accuracy is almost independent of data point distribution. epsilon also provides a good basis for comparing the quality of fit of theoretical models to experimental characteristics.