Binary Classification of Defective Solar PV Modules Using Thermography

Niazi, Kamran Ali Khan; Akhtar, W.; Khan, Hassan Abbas; Sohaib, Sarmad; Nasir, Ahmad Kamal

doi:10.1109/pvsc.2018.8548138

Cited by 18 publications

(14 citation statements)

References 23 publications

(22 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where RC(f sw ) is the effective series resistance of C at f sw . Consequently, the overall power losses can be found by summing up all the losses in ( 9)-( 12), which are given as P t = 2P on + 2P swloss + P L_loss + P C_loss (13) where P t is the total power loss. The overall loss evaluation for the proposed SCL-based DPP topology has been performed using ( 9)-( 13), and the findings have been discussed in the subsequent sections.…”

Section: Power Loss Analysismentioning

confidence: 99%

“…Therefore, the extra-power generated by non-shaded panels starts to dissipate across the shaded cell, which is acting as a load to non-shaded cells. Hence, the dissipation of power due to the shaded cells also increases the temperature of these cells known as hotspots [9], which affects the long-term reliability of such cells [10][11][12][13][14].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Simple Mismatch Mitigating Partial Power Processing Converter for Solar PV Modules

et al. 2021

View full text Add to dashboard Cite

Partial shading affects the energy harvested from photovoltaic (PV) modules, leading to a mismatch in PV systems and causing energy losses. For this purpose, differential power processing (DPP) converters are the emerging power electronic-based topologies used to address the mismatch issues. Normally, PV modules are connected in series and DPP converters are used to extract the power from these PV modules by only processing the fraction of power called mismatched power. In this work, a switched-capacitor-inductor (SCL)-based DPP converter is presented, which mitigates the non-ideal conditions in solar PV systems. A proposed SCL-based DPP technique utilizes a simple control strategy to extract the maximum power from the partially shaded PV modules by only processing a fraction of the power. Furthermore, an operational principle and loss analysis for the proposed converter is presented. The proposed topology is examined and compared with the traditional bypass diode technique through simulations and experimental tests. The efficiency of the proposed DPP is validated by the experiment and simulation. The results demonstrate the performance in terms of higher energy yield without bypassing the low-producing PV module by using a simple control. The results indicate that achieved efficiency is higher than 98% under severe mismatch (higher than 50%).

show abstract

Section: Power Loss Analysismentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A Simple Mismatch Mitigating Partial Power Processing Converter for Solar PV Modules

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Different image processing-based machine learning and deep learning approaches are utilised to identify and classify the PV system defects based on images. Niazi et al [ 26 ] used the naïve-Bayes approach to classify PV panels into two classes, defective and non-defective, with IR thermograph texture features with a 98.4% mean recognition rate. Meanwhile, Ali et al [ 1 ] used a support vector machine on IR images after extracting image features and classified PV panels into three classes based on health with 92% accuracy.…”

Section: Introductionmentioning

confidence: 99%

Photovoltaic Panels Classification Using Isolated and Transfer Learned Deep Neural Models Using Infrared Thermographic Images

Ahmed

Hanif

Kallu

et al. 2021

Sensors

View full text Add to dashboard Cite

Defective PV panels reduce the efficiency of the whole PV string, causing loss of investment by decreasing its efficiency and lifetime. In this study, firstly, an isolated convolution neural model (ICNM) was prepared from scratch to classify the infrared images of PV panels based on their health, i.e., healthy, hotspot, and faulty. The ICNM occupies the least memory, and it also has the simplest architecture, lowest execution time, and an accuracy of 96% compared to transfer learned pre-trained ShuffleNet, GoogleNet, and SqueezeNet models. Afterward, ICNM, based on its advantages, is reused through transfer learning to classify the defects of PV panels into five classes, i.e., bird drop, single, patchwork, horizontally aligned string, and block with 97.62% testing accuracy. This proposed approach can identify and classify the PV panels based on their health and defects faster with high accuracy and occupies the least amount of the system’s memory, resulting in savings in the PV investment.

show abstract

“…Solar panels are categorized into (defective and nondefective panels) using texture features extraction (TFE) which includes contrast, correlation, energy, entropy, and homogeneity [29]. The classification using n Bayes a binary class density-based classifier of c-Si PV module using thermography assessment showed a mean recognition rate of 98.4% for a set of 260 test samples.…”

Section: Introductionmentioning

confidence: 99%

Novel Features Extraction for Fault Detection Using Thermography Characteristics and IV Measurements of CIGS Thin-Film Module

Eltuhamy¹,

Rady²,

Ibrahim³

et al. 2020

I2M

View full text Add to dashboard Cite

Regarding the fault diagnosis of Copper Indium Gallium Selenide (CIGS) PV modules, previously published articles focused on employing statistical analysis of thermography images. This approach failed in many cases to distinguish among fault types. This article presents a novel methodology to diagnose and predict faults of thin-film CIGS PV modules using infrared thermography analysis combined with measurements of I-V characteristics. The proposed methodology encompasses a comprehensive site work to capture images that cover many fault types of the PV module under study. The novelty of the technique depends on utilizing processing and analysis of the captured images using new proposed mathematical parameters to extract different faults’ features. Using I-V measurements combined with thermography analysis, the differences between different types of faults are detected. Then, a general classification matrix of CIGS fault detection and diagnosis, using features based on mathematical parameters and IV measurements has been established. Results show that the analysis of the temperature distribution is proved to be insufficient to identify specific modes of different faults. In addition, the proposed procedure for fault detection and classification, which depends on the pattern of faults, can be used for any type of PV module. This results in more reliance on the proposed technique to increase the confidence level of fault detection.

show abstract

Binary Classification of Defective Solar PV Modules Using Thermography

Cited by 18 publications

References 23 publications

A Simple Mismatch Mitigating Partial Power Processing Converter for Solar PV Modules

A Simple Mismatch Mitigating Partial Power Processing Converter for Solar PV Modules

Photovoltaic Panels Classification Using Isolated and Transfer Learned Deep Neural Models Using Infrared Thermographic Images

Novel Features Extraction for Fault Detection Using Thermography Characteristics and IV Measurements of CIGS Thin-Film Module

Contact Info

Product

Resources

About