Successive large perturbation method for the elimination of initial value dependence in I-V curve fitting

A successive large perturbation method (SLPM) is proposed to resolve the problem of initial value dependence in the numerical least-square fitting for the extraction of I-V parameters in solar cells. In this method a large perturbation is applied onto certain a parameter before next turn of Newton-like iteration is proceeded and an improved result is usually obtained if the I-V parameters suffer from a latent initial value dependence problem. The numerical insensitivity of mean square of deviation to the variation of large shunt resistance is a critical factor to cause the initial value dependence. An application example for a dye-sensitized solar cell shows that about a 60% change of reverse saturation current I(0) occurs after SLPM is applied to the result obtained by Newton-like method. Our result demonstrates that SLPM is a powerful tool to eliminate the initial value dependence in I-V curve fitting.