Parameter identification of solar cell model using Levenberg-Marquardt algorithm combined with simulated annealing

In recent years, several researchers have been interested in looking for accurate parameter identification techniques able to deal with the high non-linearity of the solar cell model. In this regard, we proposed in this paper, the optimization method of Levenberg-Marquardt (LM) combined with simulated annealing (SA). The aspect of combination between the two methods is based on the optimal setting of the damping factor of LM at each iteration by SA. The performance of the proposed approach is tested by identifying the model parameters of a 57 mm diameter commercial (R.T.C. France) silicon solar cell. As a result, the developed approach shows its high accuracy when comparing the obtained outcomes with those reported by the other methods in the literature. This effectiveness is proved by the minimal values of statistical errors and the good agreement between the experimental data and the fitted curve. On the other hand, in an attempt to understand the synergy between LM and SA, we firstly studied in detail the behavior of each one and then the impact of their combination on the parameters evolution during the convergence procedure.