A Genetic Programming-Machine Learning Based Optimal Power Generation Approach for PV Arrays During Partial Shading
Solar photovoltaic (PV) energy systems are highly influenced to the environmental partial shading that diminishes the power generation to the most. Various mitigation techniques have been presented in the past but, each exhibits limitations in terms of application, cost, adaptability and complexity....
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Other Authors: | |
| Format: | Conference paper |
| Published: |
Institute of Electrical and Electronics Engineers Inc.
2025
|
| Subjects: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Solar photovoltaic (PV) energy systems are highly influenced to the environmental partial shading that diminishes the power generation to the most. Various mitigation techniques have been presented in the past but, each exhibits limitations in terms of application, cost, adaptability and complexity. Array reconfiguration have a wide acceptance as the cost-effective way of enhancing the power generation during partial shading but, the major drawback lies in the effective shade dispersion, faster operation, reliability and implementation. Hence, considering these constraints, this paper suggests an array reconfiguration technique that uses the genetic programming-machine learning (GP-ML) approach for efficient operation of PV arrays during partial shading scenarios. The proposed approach uses a lower switch count to enhance the power generation of the PV arrays and reduces the possibility of non-convex power curves during shading. The validation is carried out in the simulation using a 9?9 PV array under complex shading cases and compared with conventional and existing reconfiguration techniques using power curves and various parameters. From the analysis, it is discovered that the proposed approach enhances the average power generation of the PV array to 38.92% and 19.62% than the conventional and existing reconfiguration techniques. ? 2024 IEEE. |
|---|