TY - GEN
T1 - A Novel Approach for PV Cell Fault Detection Using YOLOv8 and Particle Swarm Optimization
AU - Phan, Quoc Bao
AU - Nguyen, Tuy Tan
N1 - Publisher Copyright:
© 2023 IEEE.
PY - 2023
Y1 - 2023
N2 - This paper introduces a novel approach for detecting faults in photovoltaic (PV) cells. The proposed method combines You Only Look Once version 8 (YOLOv8) and the Particle Swarm Optimization (PSO) architecture to enhance detection accuracy. Unlike existing methods, the proposed model leverages PSO to optimize the parameters of YOLOv8. To evaluate the effectiveness of the approach, two study cases are conducted using training sets of 70% and 80%, respectively. The PV system data is utilized as input, with YOLOv8 extracting features to detect faulty cells. The PSO algorithm optimizes the model's parameters to achieve the highest detection accuracy. Experimental results demonstrate that the proposed approach outperforms existing fault detection methods in terms of accuracy and robustness, achieving a mean Average Precision at 50 (mAP@50) of 94%. By harnessing the power of YOLOv8 and PSO, the approach offers a promising solution for reliable and efficient fault detection in PV systems, making it a viable option for enhancing system performance and reducing maintenance costs.
AB - This paper introduces a novel approach for detecting faults in photovoltaic (PV) cells. The proposed method combines You Only Look Once version 8 (YOLOv8) and the Particle Swarm Optimization (PSO) architecture to enhance detection accuracy. Unlike existing methods, the proposed model leverages PSO to optimize the parameters of YOLOv8. To evaluate the effectiveness of the approach, two study cases are conducted using training sets of 70% and 80%, respectively. The PV system data is utilized as input, with YOLOv8 extracting features to detect faulty cells. The PSO algorithm optimizes the model's parameters to achieve the highest detection accuracy. Experimental results demonstrate that the proposed approach outperforms existing fault detection methods in terms of accuracy and robustness, achieving a mean Average Precision at 50 (mAP@50) of 94%. By harnessing the power of YOLOv8 and PSO, the approach offers a promising solution for reliable and efficient fault detection in PV systems, making it a viable option for enhancing system performance and reducing maintenance costs.
KW - deep neural network
KW - optimization
KW - PV cell fault detection
KW - YOLOv8
UR - http://www.scopus.com/inward/record.url?scp=85185384052&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85185384052&partnerID=8YFLogxK
U2 - 10.1109/MWSCAS57524.2023.10406139
DO - 10.1109/MWSCAS57524.2023.10406139
M3 - Conference contribution
AN - SCOPUS:85185384052
T3 - Midwest Symposium on Circuits and Systems
SP - 634
EP - 638
BT - 2023 IEEE 66th International Midwest Symposium on Circuits and Systems, MWSCAS 2023
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2023 IEEE 66th International Midwest Symposium on Circuits and Systems, MWSCAS 2023
Y2 - 6 August 2023 through 9 August 2023
ER -