Classification of Photovoltaic Failures with Hidden Markov Modeling, an Unsupervised Statistical Approach

Hopwood, Michael; Patel, Lekha; Gunda, Thushara

doi:10.3390/en15145104

Classification of Photovoltaic Failures with Hidden Markov Modeling, an Unsupervised Statistical Approach

Journal Article · Wed Jul 13 00:00:00 EDT 2022 · Energies

DOI:https://doi.org/10.3390/en15145104· OSTI ID:1876226

; ;

Failure detection methods are of significant interest for photovoltaic (PV) site operators to help reduce gaps between expected and observed energy generation. Current approaches for field-based fault detection, however, rely on multiple data inputs and can suffer from interpretability issues. In contrast, this work offers an unsupervised statistical approach that leverages hidden Markov models (HMM) to identify failures occurring at PV sites. Using performance index data from 104 sites across the United States, individual PV-HMM models are trained and evaluated for failure detection and transition probabilities. This analysis indicates that the trained PV-HMM models have the highest probability of remaining in their current state (87.1% to 93.5%), whereas the transition probability from normal to failure (6.5%) is lower than the transition from failure to normal (12.9%) states. A comparison of these patterns using both threshold levels and operations and maintenance (O&M) tickets indicate high precision rates of PV-HMMs (median = 82.4%) across all of the sites. Although additional work is needed to assess sensitivities, the PV-HMM methodology demonstrates significant potential for real-time failure detection as well as extensions into predictive maintenance capabilities for PV.

View Journal Article

Research Organization:: Sandia National Laboratories (SNL-NM), Albuquerque, NM (United States)

Sponsoring Organization:: USDOE; USDOE National Nuclear Security Administration (NNSA); USDOE Office of Energy Efficiency and Renewable Energy (EERE), Solar Energy Technologies Program

Grant/Contract Number:: NA0003525

OSTI ID:: 1876226

Alternate ID(s):: OSTI ID: 1877169

Report Number(s):: SAND2022-9590J; PII: en15145104

Journal Information:: Energies, Journal Name: Energies Journal Issue: 14 Vol. 15; ISSN 1996-1073; ISSN ENERGA

Publisher:: MDPI AGCopyright Statement

Country of Publication:: Switzerland

Language:: English

References (25)

Evaluation of extreme weather impacts on utility-scale photovoltaic plant performance in the United States Jackson, Nicole D.; Gunda, Thushara Applied Energy, Vol. 302 https://doi.org/10.1016/j.apenergy.2021.117508	journal	November 2021
An enhanced machine learning based approach for failures detection and diagnosis of PV systems Garoudja, Elyes; Chouder, Aissa; Kara, Kamel Energy Conversion and Management, Vol. 151 https://doi.org/10.1016/j.enconman.2017.09.019	journal	November 2017
CNN based automatic detection of photovoltaic cell defects in electroluminescence images Akram, M. Waqar; Li, Guiqiang; Jin, Yi Energy, Vol. 189 https://doi.org/10.1016/j.energy.2019.116319	journal	December 2019
Online estimation of satellite lithium-ion battery capacity based on approximate belief rule base and hidden Markov model Zhao, Dao; Zhou, Zhijie; Tang, Shuaiwen Energy, Vol. 256 https://doi.org/10.1016/j.energy.2022.124632	journal	October 2022
Anomaly detection and predictive maintenance for photovoltaic systems De Benedetti, Massimiliano; Leonardi, Fabio; Messina, Fabrizio Neurocomputing, Vol. 310 https://doi.org/10.1016/j.neucom.2018.05.017	journal	October 2018
Hidden Markov model based principal component analysis for intelligent fault diagnosis of wind energy converter systems Kouadri, Abdelmalek; Hajji, Mansour; Harkat, Mohamed-Faouzi Renewable Energy, Vol. 150 https://doi.org/10.1016/j.renene.2020.01.010	journal	May 2020
An unsupervised monitoring procedure for detecting anomalies in photovoltaic systems using a one-class Support Vector Machine Harrou, Fouzi; Dairi, Abdelkader; Taghezouit, Bilal Solar Energy, Vol. 179 https://doi.org/10.1016/j.solener.2018.12.045	journal	February 2019
A tutorial on hidden Markov models and selected applications in speech recognition Rabiner, L. R. Proceedings of the IEEE, Vol. 77, Issue 2 https://doi.org/10.1109/5.18626	journal	January 1989
Neural Network-Based Classification of String-Level IV Curves From Physically-Induced Failures of Photovoltaic Modules Hopwood, Michael W.; Gunda, Thushara; Seigneur, Hubert IEEE Access, Vol. 8 https://doi.org/10.1109/ACCESS.2020.3021577	journal	January 2020
A Machine Learning Evaluation of Maintenance Records for Common Failure Modes in PV Inverters Gunda, Thushara; Hackett, Sean; Kraus, Laura IEEE Access, Vol. 8 https://doi.org/10.1109/ACCESS.2020.3039182	journal	January 2020
Predictive Maintenance of Photovoltaic Panels via deep Learning Huuhtanen, Timo; Jung, Alexander 2018 IEEE Data Science Workshop (DSW) https://doi.org/10.1109/DSW.2018.8439898	conference	June 2018
Markov process reliability model for photovoltaic module encapsulation failures Cristaldi, L.; Khalil, Mohamed; Faifer, M. 2015 International Conference on Renewable Energy Research and Applications (ICRERA) https://doi.org/10.1109/ICRERA.2015.7418696	conference	November 2015
PV Panel and PV Inverter Damages Caused by Combination of Edge Delamination, Water Penetration, and High String Voltage in Moderate Climate Poulek, Vladislav; Safrankova, Jana; Cerna, Ladislava IEEE Journal of Photovoltaics, Vol. 11, Issue 2 https://doi.org/10.1109/JPHOTOV.2021.3050984	journal	March 2021
pvOps: Improving Operational Assessments through Data Fusion Mendoza, Hector; Hopwood, Michael; Gunda, Thushara 2021 IEEE 48th Photovoltaic Specialists Conference (PVSC) https://doi.org/10.1109/PVSC43889.2021.9518439	conference	June 2021
A Density Peak-Based Clustering Approach for Fault Diagnosis of Photovoltaic Arrays Lin, Peijie; Lin, Yaohai; Chen, Zhicong International Journal of Photoenergy, Vol. 2017 https://doi.org/10.1155/2017/4903613	journal	January 2017
Disease surveillance using a hidden Markov model Watkins, Rochelle E.; Eagleson, Serryn; Veenendaal, Bert BMC Medical Informatics and Decision Making, Vol. 9, Issue 1 https://doi.org/10.1186/1472-6947-9-39	journal	August 2009
A hidden Markov model approach to characterizing the photo-switching behavior of fluorophores Patel, Lekha; Gustafsson, Nils; Lin, Yu The Annals of Applied Statistics, Vol. 13, Issue 3 https://doi.org/10.1214/19-AOAS1240	journal	September 2019
The Application of Hidden Markov Models in Speech Recognition Gales, Mark; Young, Steve Foundations and Trends® in Signal Processing, Vol. 1, Issue 3 https://doi.org/10.1561/2000000004	journal	January 2007
Generation of Data-Driven Expected Energy Models for Photovoltaic Systems Hopwood, Michael W.; Gunda, Thushara Applied Sciences, Vol. 12, Issue 4 https://doi.org/10.3390/app12041872	journal	February 2022
A Prediction Mechanism of Energy Consumption in Residential Buildings Using Hidden Markov Model Ullah, Israr; Ahmad, Rashid; Kim, DoHyeun Energies, Vol. 11, Issue 2 https://doi.org/10.3390/en11020358	journal	February 2018
PV System Predictive Maintenance: Challenges, Current Approaches, and Opportunities Bosman, Lisa B.; Leon-Salas, Walter D.; Hutzel, William Energies, Vol. 13, Issue 6 https://doi.org/10.3390/en13061398	journal	March 2020
A Review of the Degradation of Photovoltaic Modules for Life Expectancy Kim, Jaeun; Rabelo, Matheus; Padi, Siva Parvathi Energies, Vol. 14, Issue 14 https://doi.org/10.3390/en14144278	journal	July 2021
Photovoltaic Array Fault Diagnosis Based on Gaussian Kernel Fuzzy C-Means Clustering Algorithm Liu, Shengyang; Dong, Lei; Liao, Xiaozhong Sensors, Vol. 19, Issue 7 https://doi.org/10.3390/s19071520	journal	March 2019
A Monitoring System for Online Fault Detection and Classification in Photovoltaic Plants Lazzaretti, André Eugênio; Costa, Clayton Hilgemberg da; Rodrigues, Marcelo Paludetto Sensors, Vol. 20, Issue 17 https://doi.org/10.3390/s20174688	journal	August 2020
Statistical Methods for Degradation Estimation and Anomaly Detection in Photovoltaic Plants Dimitrievska, Vesna; Pittino, Federico; Muehleisen, Wolfgang Sensors, Vol. 21, Issue 11 https://doi.org/10.3390/s21113733	journal	May 2021

Similar Records

Improving qubit readout with hidden Markov models

Journal Article · Wed Dec 23 19:00:00 EST 2020 · Physical Review A · OSTI ID:1763940

Multiple alignment using hidden Markov models

Technical Report · Sat Dec 30 23:00:00 EST 1995 · OSTI ID:401836

Related Subjects

classification
14 SOLAR ENERGY
failure detection
hidden Markov model
operations and maintenance
photovoltaics
unsupervised statistical learning

Classification of Photovoltaic Failures with Hidden Markov Modeling, an Unsupervised Statistical Approach

Citation Formats

References (25)

Similar Records

Related Subjects