Modeling and evaluation of main maximum power point tracking algorithms for photovoltaics systems

Abstract

This paper presents modeling and evaluation of more widely used Maximum power Point tracking (MPPT) algorithms. These algorithms are simulated in Matlab/Simulink environment in order to provide a comparison in terms of sensors required, ease of implementation, efficiency, and the dynamic response of the Photovoltaics (PV) systems to variations in temperature and irradiance. This simulation based evaluation can be useful in specifying the appropriateness of the MPPT algorithms for the different PV system applications. It can be used as a reference modeling for future research related to the PV power generation. Furthermore, a novel artificial intelligence technique based on Adaptive Neuro-Fuzzy Inference System (ANFIS) is presented in this work. The solar irradiance and cell temperature are used as input to predict the duty cycle of the electronic switch of the DC–DC converter adopted in the system. The proposed technique provides high accuracy, stability, very fast tracking algorithm.

Introduction

The key solution to overcome energy related problems of the world is the Renewable energy technologies [1]. However, only about 14% of total world energy demand are supplied by renewable energy resources [2]. Solar energy provides an unlimited, clean and environmentally friendly energy, So, it is regarded as one of the most promising renewable energy technologies [3], [4], [5], [6], [7], [8].

The Photovoltaics systems appear to be one of the most promising technologies due to many reasons such as the electrical energy production close to where it is required, so that transport expenditure and energy losses can be avoided, no greenhouse gas emission, low maintenance cost and no noise because it do not require any moving parts [9]. The main applications of PV systems are in solar vehicles, water pumping, street lighting, remote areas electrification and space stations [10]. However, there are many drawbacks of PV systems: nonlinear behavior, the variation of the maximum power point with the climatic conditions which complicates the tracking task, low conversion efficiency which is less than 17% [11]; and high installation cost. So, it becomes necessary to use MPPT system in order to ensure the efficient operation of the solar array. MPPT is a switch-mode power converter can be used for extracting the maximum power despite the continuous changes in irradiance and temperature by the control technique which adjusts the power and achieves its greatest possible value [12]. Several MPPT algorithms have been proposed in the literature such as Fractional open circuit voltage algorithm [13], Perturb and Observe (P&O) algorithm [14], Incremental Conductance (IncCond) algorithm [15], Fuzzy Logic Control (FLC) based algorithm [16], Temperature measurement based algorithm [17] and ANFIS-based algorithm [18].

Both fractional open circuit voltage and fractional short circuit current algorithms are modified by Sayal [19] in order to overcome their limitations.

A modified P&O MPPT algorithm with variable perturbation step size is presented by Zhang et al. [20] in order to reduce the power oscillation around the MPP. A similar technique is developed by Piegari et al. [21].

Elgendy et al. [22] presented an experimental evaluation of the operating characteristics of the IncCond algorithm when employed by a PV pumping system.

In [23], Farhat et al. presented a single input fuzzy logic controller (SIFLC) based on the constant voltage algorithm. It is used as a MPPT control scheme for a stand-alone PV system.

In [24], Bin-Halabi et al. proposed an ANFIS-based MPPT algorithm.

These algorithms vary in complexity, sensors required, convergence speed, ease of implementation, popularity, and in other respects. The objective of this paper is to analyze, simulate and evaluate PV system employing different maximum power point tracking algorithms under varying operating conditions. The simulation tool SIMULINK® allows the direct evaluation of more widely used MPPT algorithms under the same operating conditions. Furthermore, this paper introduce a novel MPPT algorithm based on Adaptive Neuro-Fuzzy Inference system which can predict and track the maximum power point of PV system under rabidly changing environmental conditions in short time with minimum error and low oscillations. To be able to properly simulate the PV system, detailed mathematical model for the PV panel is presented and used to study the effect of different influences of the environmental conditions on its output characteristics.