Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Physics-Based Method for Generating Fully Synthetic IV Curve Training Datasets for Machine Learning Classification of PV Failures

Journal Article · · Energies
DOI:https://doi.org/10.3390/en15145085· OSTI ID:1875971

Classification machine learning models require high-quality labeled datasets for training. Among the most useful datasets for photovoltaic array fault detection and diagnosis are module or string current-voltage (IV) curves. Unfortunately, such datasets are rarely collected due to the cost of high fidelity monitoring, and the data that is available is generally not ideal, often consisting of unbalanced classes, noisy data due to environmental conditions, and few samples. In this paper, we propose an alternate approach that utilizes physics-based simulations of string-level IV curves as a fully synthetic training corpus that is independent of the test dataset. In our example, the training corpus consists of baseline (no fault), partial soiling, and cell crack system modes. The training corpus is used to train a 1D convolutional neural network (CNN) for failure classification. The approach is validated by comparing the model’s ability to classify failures detected on a real, measured IV curve testing corpus obtained from laboratory and field experiments. Results obtained using a fully synthetic training dataset achieve identical accuracy to those obtained with use of a measured training dataset. When evaluating the measured data’s test split, a 100% accuracy was found both when using simulations or measured data as the training corpus. When evaluating all of the measured data, a 96% accuracy was found when using a fully synthetic training dataset. The use of physics-based modeling results as a training corpus for failure detection and classification has many advantages for implementation as each PV system is configured differently, and it would be nearly impossible to train using labeled measured data.

Sponsoring Organization:
USDOE
Grant/Contract Number:
NA0003525
OSTI ID:
1875971
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 (40)

Fault diagnosis of Photovoltaic Modules journal March 2019
Identification of technical risks in the photovoltaic value chain and quantification of the economic impact: Economic impact of technical risks in photovoltaic plants journal January 2017
Photovoltaic defect classification through thermal infrared imaging using a machine learning approach
  • Dunderdale, Christopher; Brettenny, Warren; Clohessy, Chantelle
  • Progress in Photovoltaics: Research and Applications, Vol. 28, Issue 3 https://doi.org/10.1002/pip.3191
journal December 2019
PV field reliability status—Analysis of 100 000 solar systems journal March 2020
Deep learning‐based automatic detection of multitype defects in photovoltaic modules and application in real production line journal January 2021
Deep‐learning‐based pipeline for module power prediction from electroluminescense measurements journal May 2021
Computer vision tool for detection, mapping, and fault classification of photovoltaics modules in aerial IR videos
  • Bommes, Lukas; Pickel, Tobias; Buerhop‐Lutz, Claudia
  • Progress in Photovoltaics: Research and Applications, Vol. 29, Issue 12 https://doi.org/10.1002/pip.3448
journal July 2021
Computer simulation of the effects of electrical mismatches in photovoltaic cell interconnection circuits journal October 1988
An enhanced machine learning based approach for failures detection and diagnosis of PV systems journal November 2017
Fault diagnosis for photovoltaic array based on convolutional neural network and electrical time series graph journal September 2019
A novel fault diagnosis technique for photovoltaic systems based on artificial neural networks journal May 2016
Recent advances in failure diagnosis techniques based on performance data analysis for grid-connected photovoltaic systems journal April 2019
Manufacturing metrology for c-Si photovoltaic module reliability and durability, Part I: Feedstock, crystallization and wafering journal June 2016
A comprehensive review on protection challenges and fault diagnosis in PV systems journal August 2018
Application of Artificial Neural Networks to photovoltaic fault detection and diagnosis: A review journal March 2021
A comprehensive study on different types of faults and detection techniques for solar photovoltaic system journal December 2017
Deep learning based automatic defect identification of photovoltaic module using electroluminescence images journal May 2020
A Comparison of Three Methods for Selecting Values of Input Variables in the Analysis of Output From a Computer Code journal February 2000
Newly-Designed Fault Diagnostic Method for Solar Photovoltaic Generation System Based on IV-Curve Measurement journal January 2019
A Novel Convolutional Neural Network-Based Approach for Fault Classification in Photovoltaic Arrays journal January 2020
A Novel Fault Identification Method for Photovoltaic Array via Convolutional Neural Network and Residual Gated Recurrent Unit journal January 2020
Neural Network-Based Classification of String-Level IV Curves From Physically-Induced Failures of Photovoltaic Modules journal January 2020
Data-Driven $I$–$V$ Feature Extraction for Photovoltaic Modules journal September 2019
Generalized and Mechanistic PV Module Performance Prediction From Computer Vision and Machine Learning on Electroluminescence Images journal May 2020
Analytic $I_{\text{sc}}$–$V_{\text{oc}}$ Method and Power Loss Modes From Outdoor Time-Series $I$–$V$ Curves journal September 2020
Photovoltaic Module Current Mismatch Fault Diagnosis Based on I-V Data journal May 2021
Machine Learning in PV Fault Detection, Diagnostics and Prognostics: A Review
  • Rodrigues, Sandy; Ramos, Helena Geirinhas; Morgado-Dias, F.
  • 2017 IEEE 44th Photovoltaic Specialists Conference (PVSC), 2017 IEEE 44th Photovoltaic Specialist Conference (PVSC) https://doi.org/10.1109/PVSC.2017.8366581
conference June 2017
Experimental Methods to Replicate Power Loss of PV Modules in the Field for the Purpose of Fault Detection Algorithm Development conference June 2019
pvOps: Improving Operational Assessments through Data Fusion conference June 2021
An assessment of the value of principal component analysis for photovoltaic IV trace classification of physically-induced failures conference June 2020
The Impact of Cracked Solar Cells on Solar Panel Energy Delivery conference June 2020
Identifying Degradation Modes of Photovoltaic Modules Using Unsupervised Machine Learning on Electroluminescense Images conference June 2020
Deep Learning Based Module Defect Analysis for Large-Scale Photovoltaic Farms journal March 2019
Modeling and Health Monitoring of DC Side of Photovoltaic Array journal October 2015
Detecting loss mechanisms of c-Si PV modules in-situ I-V measurement conference September 2016
Review and Performance Evaluation of Photovoltaic Array Fault Detection and Diagnosis Techniques journal February 2019
pvlib python: a python package for modeling solar energy systems journal September 2018
Impact of Energy Losses Due to Failures on Photovoltaic Plant Energy Balance journal February 2018
Low-Cost I–V Tracer for PV Modules under Real Operating Conditions journal August 2020
Monitoring and Fault Detection in Photovoltaic Systems Based On Inverter Measured String I-V Curves text January 2015

Similar Records

Characterizing Quantum Classifier Utility in Natural Language Processing Workflows
Conference · Fri Sep 01 00:00:00 EDT 2023 · OSTI ID:2397467

Line Faults Classification Using Machine Learning on Three Phase Voltages Extracted from Large Dataset of PMU Measurements
Conference · Mon Jan 03 23:00:00 EST 2022 · Proceedings of the Annual Hawaii International Conference on System Sciences Proceedings of the 55th Hawaii International Conference on System Sciences · OSTI ID:1874493

Line Faults Classification Using Machine Learning on Three Phase Voltages Extracted from Large Dataset of PMU Measurements
Journal Article · Mon Jan 03 23:00:00 EST 2022 · Proceedings of the Annual Hawaii International Conference on System Sciences · OSTI ID:1891312

Related Subjects