Dynamic reconfiguration of large-scale PV plant using based on specified switching matrix and genetic algorithm to mitigate partial shading

Thumbnail Image
Aidha Muhammad Ajmal
Journal Title
Journal ISSN
Volume Title
Research Projects
Organizational Units
Journal Issue
Environmental conditions and non-uniform irradiations have a strong influence on the quality of the generated photovoltaic (PV) power. The non-uniform irradiance, also referred to as partial shading reduces the output power of the PV array, causes multiple peaks in its characteristics, reduce the lifetime of the PV array and the effectiveness of the entire system. Several techniques have been proposed to solve the mismatch problem, for example, the reconfiguration technique of PV panels. Most of the reconfiguration techniques are limited for small systems due to the large number of switches that are required for such applications. This thesis proposes a new reconfiguration technique for PV plant to overcome the partial shading effects. The reconfiguration technique relies on rearranging the PV arrays in order to disperse the shading equally over the PV arrays. The proposed technique can be applied to a large scale PV plant and is based on Total-Cross-Tie (TCT) in two reconfigurable stages. The advantages of the proposed new reconfiguration technique include shade dispersion between arrays and cost-effectiveness due to reduced number of switches and sensors as compared to existing techniques. In the first stage, the shade is dispersed over the PV plant by switching specific switches in the arrays. In the second stage, Genetic Algorithm (GA) is applied to optimize the output, via rearranging the columns in PV plants to find the optimal solution of reconfiguration. The major contribution of this thesis is the adaptability of the proposed two-stage technique to an actual 50 MWpk solar PV farm, which has not been covered in any literatures prior to this. To highlight the viability of the concept, a portion of the PV plant rated at 2.5 MWpk was used in this research. Finally, a comparison of the proposed reconfiguration technique with other static and dynamic reconfiguration techniques was carried out under different shading patterns using MATLAB/SIMULINK. A comparative study on performance analysis such as mismatch losses, fill factor, power loss and percentage power enhancement, for TCT configuration, SuDoKu, Competence Square, Two Phase, Particle Swarm Optimization, Genetic Algorithm, Modified Harris Hawks Optimizer and the proposed technique was carried out for each shading case. Also, energy-saving calculations and total income generation are performed and compared with the TCT configuration. From the results, the proposed reconfiguration technique enhances the generated power by 28 %, 3.3 %, 0.85 %, 0.85 % and 0.3 % over the TCT configuration, two-phase, SuDoKu, PSO, and MHHO techniques respectively, for Case A. Meanwhile, Case B yielded a higher Power Enhancement of 7 %, 1.3 %, 2 %, 4 % and 1.6 % for the proposed reconfiguration technique when compared to the TCT configuration, two-phase, SuDoKu, PSO, and MHHO techniques, respectively. The proposed technique also reduced the mismatch of power loss and solved the multiple peaks. In addition, the proposed technique demonstrated superiority over static and dynamic reconfiguration techniques.