Local outlier factor-based fault detection and evaluation of photovoltaic system

Ding, Hanxiang; Ding, Kun; Zhang, Jingwei; Wang, Yue; Gao, Lingang; Li, Yuanliang; Chen, Fudong; Shao, Zhi-Xiong; Lai, Wanbin

doi:10.1016/j.solener.2018.01.049

Cited by 46 publications

(19 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this case, a modified LOF algorithm is introduced and claimed to have better performance than the conventional one. The improved performance was verified based upon several simulations and testing on a real PV system [43]. A time series sliding window of photovoltaic string current can also be utilized to identify fault types and locations [44].…”

Section: Other Simulation Methodsmentioning

confidence: 96%

Review on Methods of Fault Diagnosis in Photovoltaic System Applications

Syafaruddin¹,

Zinger²

2019

JESTR

View full text Add to dashboard Cite

Recently, detection and identification of faults in photovoltaic (PV) system applications have been attracting researchers worldwide. Some of them have investigated the causes of potential faults, have worked on how to monitor and identify the faults through fault diagnosis, and have elaborated on how to develop problem-solving methods for the early detection of faults in PV system operations. This paper attempts to provide information comprehensively from the previous research. It summarizes types of faults, methods used in early fault identification and monitoring, and PV fault prevention and protection systems. The paper also investigates future trends in PV system fault diagnosis which include real-time implementation. The paper is expected to bring new perspective about the literature associate with fault studies in PV systems. This survey will be beneficial for the future discussion on how to provide comprehensive solutions for PV system fault problems. It will help avoid research repetition on similar topics and focus on the improvement and performance development of PV fault diagnosis methods.

show abstract

Section: Other Simulation Methodsmentioning

confidence: 96%

Review on Methods of Fault Diagnosis in Photovoltaic System Applications

Syafaruddin¹,

Zinger²

2019

JESTR

View full text Add to dashboard Cite

show abstract

“…Outlier detection can be mainly grouped into five categories: distribution-based, depth-based, distance-based, clusteringbased and density-based outlier detection [28]. Among them, the outlier detection based on distribution, depth or distance adopts the overall criteria, which is not accurate for some special dataset [26]. Moreover, the deviation between different points is large enough to be considered, so the outlier detection algorithm based on density performs better than others [37].…”

Section: A Outlier Detection For High-dimensional Datamentioning

confidence: 99%

“…Traditionally, when calculating the Euclidean distance, all data features, also called attributes, get the same treatment. The calculation formula is presented: (26) However, in a real dataset there is always the possibility that different features may have different degrees of relevance, which should be taken into account through the feature-weighting method [37]. In this paper, feature weights are obtained based on the Spearman correlation, which is a kind of typical nonlinear correlation measurement methods to measure the monotonic relationship between data.…”

Section: A Outlier Detection For High-dimensional Datamentioning

confidence: 99%

“…And the weight value is calculated by an improved local outlier factor (LOF) algorithm. This density-based LOF can be used to give a detailed mathematical value to represent the outlier level of the observation point, instead of using a binary attribute tag to represent whether it is an outlier [26], [27]. At the same time, traditional LOF algorithm is improved by feature weighting technology in Euclidean space.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Novel GPR-Based Prediction Model for Strip Crown in Hot Rolling by Using the Improved Local Outlier Factor

Feng

et al. 2021

IEEE Access

View full text Add to dashboard Cite

“…The 3-sigma rule was applied to determine each cluster center using the normalized voltage and current at the MPPs. Similarly, the PV local outlier factor (PVLOF) was computed from the current of the PV array to identify the degree of faults [45]. A single diode model-based prediction was implemented, enabling the generation of the residual, which was applied to the one-class SVM by quantifying the dissimilarity between the normal and faulty features [46].…”

Section: Introductionmentioning

confidence: 99%

Clustering-Based Self-Imputation of Unlabeled Fault Data in a Fleet of Photovoltaic Generation Systems

Park

Kim

et al. 2020

Energies

View full text Add to dashboard Cite

This work proposes a fault detection and imputation scheme for a fleet of small-scale photovoltaic (PV) systems, where the captured data includes unlabeled faults. On-site meteorological information, such as solar irradiance, is helpful for monitoring PV systems. However, collecting this type of weather data at every station is not feasible for a fleet owing to the limitation of installation costs. In this study, to monitor a PV fleet efficiently, neighboring PV generation profiles were utilized for fault detection and imputation, as well as solar irradiance. For fault detection from unlabeled raw PV data, K-means clustering was employed to detect abnormal patterns based on customized input features, which were extracted from the fleet PVs and weather data. When a profile was determined to have an abnormal pattern, imputation for the corresponding data was implemented using the subset of neighboring PV data clustered as normal. For evaluation, the effectiveness of neighboring PV information was investigated using the actual rooftop PV power generation data measured at several locations in the Gwangju Institute of Science and Technology (GIST) campus. The results indicate that neighboring PV profiles improve the fault detection capability and the imputation accuracy. For fault detection, clustering-based schemes provided error rates of 0.0126 and 0.0223, respectively, with and without neighboring PV data, whereas the conventional prediction-based approach showed an error rate of 0.0753. For imputation, estimation accuracy was significantly improved by leveraging the labels of fault detection in the proposed scheme, as much as 18.32% reduction in normalized root mean square error (NRMSE) compared with the conventional scheme without fault consideration.

show abstract

Local outlier factor-based fault detection and evaluation of photovoltaic system

Cited by 46 publications

References 29 publications

Review on Methods of Fault Diagnosis in Photovoltaic System Applications

Review on Methods of Fault Diagnosis in Photovoltaic System Applications

A Novel GPR-Based Prediction Model for Strip Crown in Hot Rolling by Using the Improved Local Outlier Factor

Clustering-Based Self-Imputation of Unlabeled Fault Data in a Fleet of Photovoltaic Generation Systems

Contact Info

Product

Resources

About