An Effective Salp Swarm Based MPPT for Photovoltaic Systems Under Dynamic and Partial Shading Conditions

Mohd Nasrul Izzani Jamaludin,Syahrul Ashikin Azmi,Hassan Haes Alhelou,Thanikanti Sudhakar Babu,Azralmukmin Azmi,Nur Hafizah Ghazali,Mohammad Faridun Naim Tajuddin,Jubaer Ahmed

doi:10.1109/access.2021.3060431

Abstract

This study proposes a duty cycle-based direct search method that capitalizes on a bioinspired optimization algorithm known as the salp swarm algorithm (SSA). The goal is to improve the tracking capability of the maximum power point (MPP) controller for optimum power extraction from a photovoltaic system under dynamic environmental conditions. The performance of the proposed SSA is tested under a transition between uniform irradiances and a transition between partial shading (PS) conditions with a focus on convergence speed, fast and accurate tracking, reduce high initial exploration oscillation, and low steady-state oscillation at MPP. Simulation results demonstrate the superiority of the proposed SSA algorithm in terms of tracking performance. The performance of the SSA method is better than the conventional (hill-climbing) and among other popular metaheuristic methods. Further validation of the SSA performance is conducted via experimental studies involving a DC-DC buck-boost converter driven by TMS320F28335 DSP on the Texas Instruments Experimenter Kit platform. Hardware results show that the proposed SSA method aligns with the simulation in terms of fast-tracking, convergence speed, and satisfactory accuracy under PS and dynamic conditions. The proposed SSA method tracks maximum power with high efficiency through its superficial structures and concepts, as well as its easy implementation. Moreover, the SSA maintains a steady-state oscillation at a minimum level to improve the overall energy yield. It thus compensates for the shortcomings of other existing methods.

Highlights

Over the past decade, the increasing energy consumption and the inevitable reduction in fossil fuel resources, as well as the rapid environmental deterioration resulting from global warming, have prompted global efforts to study renewable energy sources (RESs), such as wind, solar energy, hydropower, geothermal energy and biomass [1]
This study proposes a duty cycle-based direct search method that capitalizes on a bioinspired optimization algorithm known as the salp swarm algorithm (SSA)
Enable standard SSA algorithm to deal with dynamic maximum power point trackers (MPPTs) problem effectively, the following re-initialization shown in Equation (6) is used to reset the positions of the salps: P(k) i

Summary

INTRODUCTION

The increasing energy consumption and the inevitable reduction in fossil fuel resources (coal, oil, natural gas), as well as the rapid environmental deterioration resulting from global warming, have prompted global efforts to study renewable energy sources (RESs), such as wind, solar energy, hydropower, geothermal energy and biomass [1]. The proposed algorithm introduces a duty cycle boundary concept to direct the searching area towards the probable GMPP region It has a straightforward control structure, simple implementation, low energy loss during the initial oscillation of the MPP tracking process, high convergence speed, and accurate tracking. Even if the optimum shifts slightly in the event of a small change in irradiance, there is a continuous loss in power attainable from the PV system, which can significantly lower efficiency To overcome this drawback, several improvements are proposed to the standard SSA to ensure dynamic tracking of the true optima and avoid severe losses. To enable standard SSA algorithm to deal with dynamic MPPT problem effectively, the following re-initialization shown in Equation (6) is used to reset the positions of the salps: P(k) i

ADVANTAGES OF THE PROPOSED SSA

SIMULATIONS RESULTS

EXPERIMENTAL VALIDATION

VIII. CONCLUSION