Maximum Power Point Tracking Algorithm for Advanced Photovoltaic Systems

Abstract

Photovoltaic (PV) systems are the major nonconventional sources for power generation for present power strategy. The power of PV system has rapid increase because of its unpolluted, less noise and limited maintenance. But whole PV system has two main disadvantages drawbacks, that is, the power generation of it is quite low and the output power is nonlinear, which is influenced by climatic conditions, namely environmental temperature and the solar irradiation. The natural limiting factor is that PV potential in respect of temperature and irradiation has nonlinear output behavior. An automated power tracking method, for example, maximum power point tracking (MPPT), is necessarily applied to improve the power generation of PV systems. The MPPT methods undergo serious challenges when the PV system is under partial shade condition because PV shows several peaks in power. Hence, the exploration method might easily be misguided and might trapped to the local maxima. Therefore, a reasonable exploratory method must be constructed, which has to determine the global maxima for PV of shaded partially. The traditional approaches namely constant voltage tracking (CVT), perturb and observe (P&O), hill climbing (HC), Incremental Conductance (INC), and fractional open circuit voltage (FOCV) methods, indeed some of their improved types, are quite incompetent in tracking the global MPP (GMPP). Traditional techniques and soft computing-based bio-inspired and nature-inspired algorithms applied to MPPT were reviewed to explore the possibility for research while optimizing the PV system with global maximum output power under partially shading conditions. This paper is aimed to review, compare, and analyze almost all the techniques that implemented so far. Further this paper provides adequate details about algorithms that focuses to derive improved MPPT under non-uniform irradiation. Each algorithm got merits and demerits of its own with respect to the converging speed, computing time, complexity of coding, hardware suitability, stability and so on.