Model-driven Per-panel Solar Anomaly Detection for Residential Arrays

There has been significant growth in both utility-scale and residential-scale solar installations in recent years, driven by rapid technology improvements and falling prices. Unlike utility-scale solar farms that are professionally managed and maintained, smaller residential-scale installations often lack sensing and instrumentation for performance monitoring and fault detection. As a result, faults may go undetected for long periods of time, resulting in generation and revenue losses for the homeowner. In this article, we present SunDown, a sensorless approach designed to detect per-panel faults in residential solar arrays. SunDown does not require any new sensors for its fault detection and instead uses a model-driven approach that leverages correlations between the power produced by adjacent panels to detect deviations from expected behavior. SunDown can handle concurrent faults in multiple panels and perform anomaly classification to determine probable causes. Using two years of solar generation data from a real home and a manually generated dataset of multiple solar faults, we show that SunDown has a Mean Absolute Percentage Error of 2.98% when predicting per-panel output. Our results show that SunDown is able to detect and classify faults, including from snow cover, leaves and debris, and electrical failures with 99.13% accuracy, and can detect multiple concurrent faults with 97.2% accuracy.

[1]  Prashant Shenoy,et al.  Solar-TK: A Data-Driven Toolkit for Solar PV Performance Modeling and Forecasting , 2019, 2019 IEEE 16th International Conference on Mobile Ad Hoc and Sensor Systems (MASS).

[2]  Dongsheng Li,et al.  Hierarchical Anomaly Detection and Multimodal Classification in Large-Scale Photovoltaic Systems , 2019, IEEE Transactions on Sustainable Energy.

[3]  Ying Sun,et al.  An unsupervised monitoring procedure for detecting anomalies in photovoltaic systems using a one-class Support Vector Machine , 2019, Solar Energy.

[4]  Yang Hu,et al.  Fault diagnosis approach for photovoltaic arrays based on unsupervised sample clustering and probabilistic neural network model , 2018, Solar Energy.

[5]  Margarida Silveira,et al.  Unsupervised Anomaly Detection in Energy Time Series Data Using Variational Recurrent Autoencoders with Attention , 2018, 2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA).

[6]  Ling Zhu,et al.  Condition classification and performance of mismatched photovoltaic arrays via a pre-filtered Elman neural network decision making tool , 2018, Solar Energy.

[7]  Athanasios V. Vasilakos,et al.  Anomaly detection and predictive maintenance for photovoltaic systems , 2018, Neurocomputing.

[8]  Bradley J. Bazuin,et al.  Unsupervised Fault Detection and Analysis for Large Photovoltaic Systems Using Drones and Machine Vision , 2018, Energies.

[9]  Prashant J. Shenoy,et al.  SolarClique: Detecting Anomalies in Residential Solar Arrays , 2018, COMPASS.

[10]  Silvano Vergura,et al.  Hypothesis Tests-Based Analysis for Anomaly Detection in Photovoltaic Systems in the Absence of Environmental Parameters , 2018 .

[11]  T. Hoff,et al.  A New Version of the SUNY Solar Forecast Model: A Scalable Approach to Site-Specific Model Training , 2018 .

[12]  Violeta Holmes,et al.  Parallel fault detection algorithm for grid-connected photovoltaic plants , 2017 .

[13]  Guangyu Liu,et al.  A fault detection and diagnosis technique for solar system based on Elman neural network , 2017, 2017 IEEE 2nd Information Technology, Networking, Electronic and Automation Control Conference (ITNEC).

[14]  David E. Irwin,et al.  Black-box Solar Performance Modeling , 2017, SIGMETRICS Perform. Evaluation Rev..

[15]  David E. Irwin,et al.  A Cloud-Based Black-Box Solar Predictor for Smart Homes , 2017, ACM Trans. Cyber Phys. Syst..

[16]  Alessia Saggese,et al.  Real Time Fault Detection in Photovoltaic Cells by Cameras on Drones , 2017, ICIAR.

[17]  Ying Sun,et al.  Statistical fault detection in photovoltaic systems , 2017 .

[18]  David Irwin,et al.  SunDance: Black-box Behind-the-Meter Solar Disaggregation , 2017, e-Energy.

[19]  H. Mekki,et al.  Artificial neural network-based modelling and fault detection of partial shaded photovoltaic modules , 2016, Simul. Model. Pract. Theory.

[20]  Bernhard Schölkopf,et al.  Modeling confounding by half-sibling regression , 2016, Proceedings of the National Academy of Sciences.

[21]  Giorgio Sulligoi,et al.  A novel fault diagnosis technique for photovoltaic systems based on artificial neural networks , 2016 .

[22]  Philip T. Krein,et al.  Photovoltaic Hot-Spot Detection for Solar Panel Substrings Using AC Parameter Characterization , 2016, IEEE Transactions on Power Electronics.

[23]  Yuji Kasai,et al.  Anomaly Detection of Solar Power Generation Systems Based on the Normalization of the Amount of Generated Electricity , 2015, 2015 IEEE 29th International Conference on Advanced Information Networking and Applications.

[24]  Lukasz Golab,et al.  What's Wrong with my Solar Panels: a Data-Driven Approach , 2015, EDBT/ICDT Workshops.

[25]  Nicholas A. Engerer,et al.  KPV: A clear-sky index for photovoltaics , 2014 .

[26]  Joshua S. Stein,et al.  Introduction to the open source PV LIB for python Photovoltaic system modelling package , 2014, 2014 IEEE 40th Photovoltaic Specialist Conference (PVSC).

[27]  Xueguan Song,et al.  Photovoltaic fault detection using a parameter based model , 2013 .

[28]  Jung-Wook Park,et al.  Diagnosis of Output Power Lowering in a PV Array by Using the Kalman-Filter Algorithm , 2012, IEEE Transactions on Energy Conversion.

[29]  Bo Hu.  Solar Panel Anomaly Detection and Classification , 2012 .

[30]  Prashant J. Shenoy,et al.  Predicting solar generation from weather forecasts using machine learning , 2011, 2011 IEEE International Conference on Smart Grid Communications (SmartGridComm).

[31]  Hans-Georg Beyer,et al.  Irradiance Forecasting for the Power Prediction of Grid-Connected Photovoltaic Systems , 2009, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[32]  Detlev Heinemann,et al.  FORECAST OF ENSEMBLE POWER PRODUCTION BY GRID-CONNECTED PV SYSTEMS , 2007 .

[33]  M. Ratner.  The Year in Review , 1990, Bio/Technology.