A Fast-Converging Radial Basis Function Neural Network-Based MPPT Controller for Static and Dynamic Variations in Solar Irradiation

The use of maximum power point tracking techniques, often known as MPPT algorithms, is required to improve the performance of PV systems. In rapidly varying atmospheric conditions, the traditional MPPT approaches do not work as intended. In the paper, a perturb and observe technique based MPPT algor...

Full description

Saved in:
Bibliographic Details
Main Authors: Venkateswararao, Chepuri, Naik, Kanasottu Anil
Format: Conference Proceeding
Language:English
Subjects:
convergence
Encryption
maximum power point
Maximum power point trackers
neural network
partial shading condition
radial basis function
Radial basis function networks
Radiation effects
Steady-state
Training
Voltage
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The use of maximum power point tracking techniques, often known as MPPT algorithms, is required to improve the performance of PV systems. In rapidly varying atmospheric conditions, the traditional MPPT approaches do not work as intended. In the paper, a perturb and observe technique based MPPT algorithm is developed together with a radial basis function neural network (RBFNN). To specify and track the maximum power point (MPP), the proposed framework is implemented. Employing the RBFNN as the input-output training information set, the optimal duty cycle is computed while considering varied PV array current and voltage values. Further, an intelligent reconfiguration strategy is developed to enhance the MPP and array characteristics. The proposed hybrid RBFNN and intelligent reconfiguration methodology enhance the performance by 43.05%, 12.22%, 6.81%, 5.6% with the reduced convergence time of 0.06 sec under different shading conditions.
ISSN:2768-0576
DOI:10.1109/ICCECE51049.2023.10085281