Graph-based semi-supervised learning for fault detection and classification in solar photovoltaic arrays

Zhao, Yi; Lehman, Brad; Ball, Roy; Palma, Jean-François de

doi:10.1109/ecce.2013.6646901

Cited by 41 publications

(63 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Training the model with input-output data helps overcome the limitation of defining thresholds and aids in the detection and classification of faults. Some of the machine-learning techniques used so far are: modified ANN with the extension theory [29], evidence theory and Fuzzy mathematics [30], TSK-FRBS Fuzzy estimator [31], Bayesian belief networks [32], three-layered ANN [33], decision tree-based method [34], and graph-based semisupervised learning [35].…”

Section: Methods 4: Machine-learning Techniques (Mlts) By Learning mentioning

confidence: 99%

“…9. These particular faults have been considered because conventional protection devices may not be able to detect and clear these faults [5], [35].…”

Section: Fault Detection and Classification Using Pnnmentioning

confidence: 99%

“…The observability of a string inverter system is more when compared to the central inverter system, since each string of PV modules has its own inverter. Unlike fault-detection methods such as the method presented in [35], which can only detect open strings rather than where the open circuit is in the string, the proposed fault-detection method gives a good estimate of where the open circuit exists.…”

Section: Fault Detection and Classification Using Pnnmentioning

confidence: 99%

See 2 more Smart Citations

Modeling and Health Monitoring of DC Side of Photovoltaic Array

Akram

Lotfifard

2015

IEEE Trans. Sustain. Energy

138

View full text Add to dashboard Cite

In this paper, a health monitoring method for photovoltaic (PV) systems based on probabilistic neural network (PNN) is proposed that detects and classifies short-and opencircuit faults in real time. To implement and validate the proposed method in computer programs, a new approach for modeling PV systems is proposed that only requires information from manufacturer's datasheet reported under normal-operating cell temperature (NOCT) conditions and standard-operating test conditions (STCs). The proposed model precisely represents characteristics of PV systems at different temperatures, as the temperature dependency of parameters such as ideality factor, series resistance, and thermal voltage is considered in the proposed model. Although this model can be applied to a variety of applications, it is specifically used to test and validate the performance of the proposed fault detection and classification method. Index Terms-Fault detection, monitoring systems, photovoltaic (PV) modeling, probabilistic neural network (PNN).

show abstract

Section: Methods 4: Machine-learning Techniques (Mlts) By Learning mentioning

confidence: 99%

“…9. These particular faults have been considered because conventional protection devices may not be able to detect and clear these faults [5], [35].…”

Section: Fault Detection and Classification Using Pnnmentioning

confidence: 99%

Section: Fault Detection and Classification Using Pnnmentioning

confidence: 99%

See 1 more Smart Citation

Modeling and Health Monitoring of DC Side of Photovoltaic Array

Akram

Lotfifard

2015

IEEE Trans. Sustain. Energy

138

View full text Add to dashboard Cite

show abstract

“…Semi-supervised learning could allow the generation of many realistic faults from a few measured examples [5]. This would mitigate the problem of lopsided data, where very few examples of faults are available.…”

Section: A Machine Learning In Fault Detectionmentioning

confidence: 99%

“…This would mitigate the problem of lopsided data, where very few examples of faults are available. Once generated, clustering approaches such as K-means or graph based semi supervised learning could be used to train a dataset [5,12,21]. We use the K-means algorithm as a starting point since there is an I-V level clustering observed in the data [4].…”

Section: A Machine Learning In Fault Detectionmentioning

confidence: 99%

An 18 kW solar array research facility for fault detection experiments

Rao

Ramirez

Braun

et al. 2016

2016 18th Mediterranean Electrotechnical Conference (MELECON)

View full text Add to dashboard Cite

Monitoring utility-scale solar arrays was shown to minimize cost of maintenance and help optimize the performance of the array under various conditions. In this paper, we describe the design of an 18 kW experimental facility that consists of 104 panels fitted with smart monitoring devices. Each of these devices embeds sensors, wireless transceivers, and relays that enable continuous monitoring, fault detection, and real-time connection topology changes. The facility enables networked data exchanges via the use of wireless data sharing with servers, fusion and control centers, and mobile devices. Research planned at this stage includes developing machine learning methods for fault detection. Preliminary simulation results on fault detection using machine learning are given in this paper.

show abstract

Failure diagnosis and trend‐based performance losses routines for the detection and classification of incidents in large‐scale photovoltaic systems

Livera

Theristis

Micheli

et al. 2022

Progress in Photovoltaics

View full text Add to dashboard Cite

Fault detection and classification in photovoltaic (PV) systems through real-time monitoring is a fundamental task that ensures quality of operation and significantly improves the performance and reliability of operating systems. Different statistical and comparative approaches have already been proposed in the literature for fault detection; however, accurate classification of fault and loss incidents based on PV performance time series remains a key challenge. Failure diagnosis and trend-based performance loss routines were developed in this work for detecting PV underperformance and accurately identifying the different fault types and loss mechanisms. The proposed routines focus mainly on the differentiation of failures (e.g., inverter faults) from irreversible (e.g., degradation) and reversible (e.g., snow and soiling) performance loss factors based on statistical analysis. The proposed routines were benchmarked using historical inverter data obtained from a 1.8 MWp PV power plant. The results demonstrated the effectiveness of the routines for detecting failures and loss mechanisms and the capability of the pipeline for distinguishing underperformance issues using anomaly detection and change-point (CP) models.Finally, a CP model was used to extract significant changes in time series data, to detect soiling and cleaning events and to estimate both the performance loss and degradation rates of fielded PV systems.

show abstract

Graph-based semi-supervised learning for fault detection and classification in solar photovoltaic arrays

Cited by 41 publications

References 7 publications

Modeling and Health Monitoring of DC Side of Photovoltaic Array

Modeling and Health Monitoring of DC Side of Photovoltaic Array

An 18 kW solar array research facility for fault detection experiments

Failure diagnosis and trend‐based performance losses routines for the detection and classification of incidents in large‐scale photovoltaic systems

Contact Info

Product

Resources

About