Very Fast and Accurate Procedure for the Characterization of Photovoltaic Panels from Datasheet Information

Abstract

In recent years several numerical methods have been proposed to identify the five-parameter model of photovoltaic panels from manufacturer datasheets also by introducing simplification or approximation techniques. In this paper we present a fast and accurate procedure for obtaining the parameters of the five-parameter model by starting from its reduced form. The procedure allows characterizing, in few seconds, thousands of photovoltaic panels present on the standard databases. It introduces and takes advantage of further important mathematical considerations without any model simplifications or data approximations. In particular the five parameters are divided in two groups, independent and dependent parameters, in order to reduce the dimensions of the search space. The partitioning of the parameters provides a strong advantage in terms of convergence, computational costs, and execution time of the present approach. Validations on thousands of photovoltaic panels are presented that show how it is possible to make easy and efficient the extraction process of the five parameters, without taking care of choosing a specific solver algorithm but simply by using any deterministic optimization/minimization technique.

Model Identification for Photovoltaic Panels Using Neural Networks

“…Among the different version of the five parameter models, which describe these dependences, the one proposed and validated in (Desoto et al, 2006) is one of the most successfully adopted, and, for this reason, is the one used in this work. (Laudani et al, 2014b). A schematization of this identification problem approach is illustrated in Fig.…”

“…By imposing the non-negativity of I Irr , I 0 and R SH , it is possible to formulate an upper bound for R S itself. Thus the resulting equations with the relative constraints constitutes an optimization problem easier to address and solve than the original one inherent in the five-parameter model identification (Laudani et al, 2014a;Laudani et al, 2014b). Nevertheless, the system of the reduced model still cannot be solved easily outside a suitable computing environment, such as Matlab, Mathcad, etc.…”

“…The five-parameter model needs a high exploration algorithm to isolate candidate solutions, that are later refined by a local search algorithm, to discern local minima from global optimum. The two-parameter reduced form model, even if convex, still requires a local non-linear optimization algorithm to be efficiently identified (Laudani et al, 2014b): this is a serious drawback, which does not allow an easy implementation of a generic solver for the problem outside a suitable computing environment, such as Matlab. This is a serious limitation, since it does not allow the integration of these identification algorithms in low cost embedded system, based on microncontroller architecture, for the monitoring and the management of PV plants (Carrasco et al, 2013).…”

Sensitivity analysis of the reduced forms of the one‐diode model for photovoltaic devices

“…The second category on the other hand, instead of focusing on whole curves, focuses on specific points (OC, open circuit; SC; short circuit; MP; maximum power) that are usually reported on the device datasheet. The identification problem, in this case, consists in solving a system of 5 non‐linear implicit equations in the 5 unknowns that are the parameters . An identification problem posed in this fashion is non‐linear, non‐convex and subject to a semi‐open 5‐dimensional solution space.…”

“…[10][11][12][13][14][15][16][17][18] An identification problem posed in this fashion is non-linear, non-convex and subject to a semi-open 5-dimensional solution space. This problem was simplified considerably when the reduced forms 10,11 were introduced. It was discovered that the 5 parameters defining the model could be divided in 2 sets: 2 dependent parameters (R S and n) and 3 independent parameters (I 0 , I irr , and R SH ).…”