Photovoltaic array reconfiguration method based on fuzzy logic and recursive least squares: An experimental validation. (1st October 2021)

• Record Type: Journal Article
• Authors: Bouselham, Loubna; Rabhi, Abdelhamid; Hajji, Bekkay; Mellit, Adel
• Is Part Of: Energy. Volume 232(2021)
• Publication Date: 2021-10-01
• Subjects: Dynamic PV array -- Partial shading -- Fuzzy logic -- Irradiance equalization -- Irradiance estimator -- Recursive least squares; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2021.121107
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: In this paper, an experimental analysis and validation of a simple reconfigurable photovoltaic (PV) array is carried out. An assessment of a new reconfiguration method based on fuzzy logic (FL) under partial shading conditions is introduced. Furthermore, a recursive least squares based irradiance estimator is proposed aiming to reduce the investment cost of the dynamic PV array. An experimental comparison with other estimators showed the high precision of the proposed estimator. The estimation error has decreased by an average of 10% compared to the first estimator (based on the PV current and voltage measurement) and by 4.28%compared to the second estimator(based on the PV current measurement). On the other hand, the results validated the FL Controller ability to switch to the appropriate configuration under prevailing shading conditions. The method was tested for a simple configuration, however it could be generalized for small-scale configurations as residential house (average power output equal to 5 kWh). To evaluate the performance of the FL method an extended simulation of dynamic PV array of 16 PV modules is also realized. The mismatch loss is mitigated by nearly 50% compared to fixed Total-Cross-Tied and 8% compared to basic Irradiance Equalization techniques. Highlights: Experimental assessment of a reconfigurable small-scale photovoltaic array is performed. Reconfiguration method based on fuzzy logic under partial shading conditions is proposed. A new irradiance estimator based on recursive least squares is proposed. Real time comparison showed the high accuracy of the proposed estimator. Energy loss in the experienced reconfigurable photovoltaic array is significantly reduced.