Modeling and Simulation of a Reconfigurable PV Preregulator for Efficiency Improvement Under Shading
Available at: https://digitalcommons.calpoly.edu/theses/3227
Date of Award
2-2026
Degree Name
MS in Electrical Engineering
Department/Program
Electrical Engineering
College
College of Engineering
Advisor
Taufik
Advisor Department
Electrical Engineering
Advisor College
College of Engineering
Abstract
In this work, a solar array with a single reconfigurable preregulator capable of switching two strings of panels between series and parallel connections to improve efficiency over various shading conditions was proposed. To demonstrate the operation and evaluate the performance of the proposed preregulator, a MATLAB model consisting of a 2400W array, a buck DC-DC converter controlled by a P&O MPPT, a full-bridge unipolar PWM inverter, an LCL filter, and a grid-emulating load was constructed. The array consisted of two strings of 3 400W panels, with the strings connected in series, parallel, or with the preregulator in between the strings. Each of these connections were simulated at different temperatures, irradiances, and shading conditions to determine system efficiency, and the efficiencies between connection types were compared. Simulation results show that for the preregulator, the efficiency was increased up to 50% under light shading compared to a series connection, and by up to 50% under heavy shading conditions over parallel. This was under best conditions, with average being closer to 20% for each. It also performed 2.5% better than parallel and 0.5% better than series under ideal conditions. The device cost was estimated to be minimal compared to system costs, and significantly less than the grid reconfiguration technique currently being developed in various literature. Because of these factors, further studies into control development and various sized systems is warranted, as this proves to be an effective avenue to improving solar generation, especially for systems that operate under variable shading conditions.