| Work Detail |
Researchers in India have used the Adaptive Network-Based Fuzzy Inference System (ANFIS) and the Gray Wolf Optimization (GWO) algorithm to develop an MPPT technique that reportedly increases the PV source voltage of an array in partially shaded conditions.
Researchers at the Vellore Institute of Technology, India, have devised a new hybrid maximum power point tracking (MPPT) method to improve the energy performance of photovoltaic systems operating under partial shading conditions.
The proposed technique uses an adaptive network-based fuzzy inference system (ANFIS), which is a special type of artificial neural network that uses the Takagi-Sugeno fuzzy inference system. ANFIS is mainly intended to limit the disadvantages of fuzzy and neural network controllers, and in photovoltaic systems its objective is to reduce failures.
“The ANFIS rules approximate the photovoltaic function from a non-linear to a linear way,” the scientists explain. “In addition, this network is used in energy management systems for the proper functioning of batteries powered by photovoltaic energy.”
The new method is also based on the use of controllers based on the gray wolf optimization (GWO) algorithm, which mimics the social behavior of gray wolves when hunting.
“In this algorithm there are four main parameters: alpha, beta, delta and omega,” explains the research team. “The alpha variable (a) is illustrated as an optimal solution among the different possible solutions or leading wolf. Similarly, the constants beta (ß) and delta (d) are selected as the second and third best optimal solutions from the search space. The normal wolf solution is given by omega (?)”.
The proposed configuration, according to its creators, allows optimizing the exploration convergence ratio of wolves, while balancing their population and search space. Furthermore, it ensures the supply of PV voltage to the DC-DC power converter to enhance the voltage of the PV source to meet future microgrid voltages.
Scientists tested the performance of the MPPT technique and the associated solar-powered converter circuit in a series of simulations performed on a 3-diode PV array model using Matlab-Simulink software. A supply-side capacitor filtered out PV power ripples and protected the switches from rapid variations in supply voltages. A supply inductor loaded the PV supply voltage under the forward bias conditions of the MOSFET and smoothed the supply voltage.
“The MOSFET is used because its properties are the lowest gate-source voltage necessary for conduction, high performance at low nominal voltages, high source resistance to limit the gate current, fast response and the least possibility of damage,” the researchers explain.
The simulation took into account parameters such as oscillations through the MPP, output power extraction, MPP stabilization time, dependence on photovoltaic modeling, operational duty value of the converter, accuracy of detection of errors of the MPPT, the complexity of the algorithm, the tracking speed, the periodic adjustment required and the number of detection parameters used.
The research group claims that the new hybrid MPPT technique can offer “good results” compared to other MPPT methods, with the voltage of the photovoltaic source increasing from one level to another. “The merits of the converter used are its easy understanding, its high reliability and its low implementation price,” he added. |
