On the parameters domain of the single-diode model

TL;DR

This paper introduces a mathematical approach and a computational algorithm to determine the parameter domain of the single-diode model for solar PV systems based on key points of the I-V curve, validated with real data.

Contribution

It provides a new method to explicitly define the parameter domain of the SDM using inequalities and a fast algorithm, enhancing the analysis of PV system characteristics.

Findings

The inequalities $V_{oc}/2 < V_{mp} < V_{oc}$ and $I_{sc}/2 < I_{mp} < I_{sc}$ define the SDM parameter domain.

The algorithm processes 40,000 synthetic points in about one minute on a standard laptop.

The method successfully analyzes real PV data from California Energy Commission and NREL.

Abstract

The current-voltage (I-V) curves of solar photovoltaic (PV) systems have been widely used as a tool to determine their electrical operation. Usually, I-V curves are described employing three cardinal points: the short-circuit point $\left(0,I_\mathrm{sc}\right)$, the open-circuit point $\left(V_\mathrm{oc},0\right)$, and the maximum power point $\left(V_\mathrm{mp},I_\mathrm{mp} \right)$. In this context, to study the I-V curves, the single-diode model (SDM) is often used. In this manuscript, we present a practical approach for determining the parameters domain of the SDM given a set of cardinal points. An in-depth mathematical analysis showed that the SDM must satisfy the following inequalities $V_\mathrm{oc}/2 <V_\mathrm{mp}<V_\mathrm{oc}$ and $I_\mathrm{sc}/2 <I_\mathrm{mp}<I_\mathrm{sc}$. Then, it is possible to implicitly represent the SDM by a single parameter and, consequently, to determine its domain. For this purpose, a fast open-source computational algorithm for the single parameter domain of the SDM is presented. The algorithm has been computer-tested using $40\,000$ synthetically generated maximum power points which are processed in about one minute of the CPU time on a standard laptop computer. Moreover, as practical applications, the databases from the California Energy Commission and the NREL are analyzed using the presented algorithm.