TY - JOUR
T1 - Intelligent and data-driven fault detection of photovoltaic plants
AU - Yao, Siya
AU - Kang, Qi
AU - Zhou, Mengchu
AU - Abusorrah, Abdullah
AU - Al-Turki, Yusuf
N1 - Funding Information:
Funding: This work was supported in part by the China Scholarship Council Scholarship, in part by the National Natural Science Foundation of China under Grant 51775385, 61703279 and 71371142, the Strategy Research Project of Artificial Intelligence Algorithms of Ministry of Education of China, and in part by Innovation Program of Shanghai Municipal Education Commission under Grant 202101070007E00098, and in part the Deanship of Scientific Research (DSR) at King Abdulaziz University under grant no. RG-22-135-41.
Publisher Copyright:
© 2021 by the authors. Licensee MDPI, Basel, Switzerland.
PY - 2021/10
Y1 - 2021/10
N2 - Most photovoltaic (PV) plants conduct operation and maintenance (O&M) by periodical inspection and cleaning. Such O&M is costly and inefficient. It fails to detect system faults in time, thus causing heavy loss. To ensure their operations are at an ideal state, this work proposes an unsupervised method for intelligent performance evaluation and data-driven fault detection, which enables engineers to check PV panels in time and implement timely maintenance. It classifies monitoring data into three subsets: ideal period A, transition period S, and downturn period B. Based on A and B datasets, we build two non-continuous regression prediction models, which are based on a tree ensemble algorithm and then modified to fit the non-continuous characteristic of PV data. We compare real-time measured power with both upper and lower reference baselines derived from two predictive models. By calculating their threshold ranges, the proposed method achieves the instantaneous performance monitoring of PV power generation and provides failure identification and O&M suggestions to engineers. It has been assessed on a 6.95 MW PV plant. Its evaluation results indicate that it is able to accurately determine different functioning states and detect both direct and indirect faults in a PV system, thereby achieving intelligent data-driven maintenance.
AB - Most photovoltaic (PV) plants conduct operation and maintenance (O&M) by periodical inspection and cleaning. Such O&M is costly and inefficient. It fails to detect system faults in time, thus causing heavy loss. To ensure their operations are at an ideal state, this work proposes an unsupervised method for intelligent performance evaluation and data-driven fault detection, which enables engineers to check PV panels in time and implement timely maintenance. It classifies monitoring data into three subsets: ideal period A, transition period S, and downturn period B. Based on A and B datasets, we build two non-continuous regression prediction models, which are based on a tree ensemble algorithm and then modified to fit the non-continuous characteristic of PV data. We compare real-time measured power with both upper and lower reference baselines derived from two predictive models. By calculating their threshold ranges, the proposed method achieves the instantaneous performance monitoring of PV power generation and provides failure identification and O&M suggestions to engineers. It has been assessed on a 6.95 MW PV plant. Its evaluation results indicate that it is able to accurately determine different functioning states and detect both direct and indirect faults in a PV system, thereby achieving intelligent data-driven maintenance.
KW - Fault detection
KW - PV monitoring system
KW - Performance evaluation
KW - Tree-based regression
KW - Unsupervised learning method
UR - http://www.scopus.com/inward/record.url?scp=85116050703&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85116050703&partnerID=8YFLogxK
U2 - 10.3390/pr9101711
DO - 10.3390/pr9101711
M3 - Article
AN - SCOPUS:85116050703
SN - 2227-9717
VL - 9
JO - Processes
JF - Processes
IS - 10
M1 - 1711
ER -