Photovoltaic cell module classification

The present study focuses on automatic defects classification of PV cells in electroluminescence images. Two machine learning approaches, features extraction-based support vector machine (SVM) and convolutional neural network (CNN), are used for the solar cell defect classifications.

Automatic defect classification in PV cells is presumed to be possible using CNN architecture and other feature extraction techniques such as histograms of oriented gradients (HOG), KAZE, SIFT, and speeded-up-robust features (SURF).

Photovoltaic (PV) power is generated when PV cell (i.e. solar cell) converts sunlight into electricity. As the industrial-level of PV cell, mono- and multi-crystalline silicon solar cells are taking the highest market share (over 97%) . In producing solar cells, invisible microcracks or defects in the Si wafer are common during process steps.

EL image classification for Photovoltaic cells is accomplished by training a model with EL images using a radial-based kernel SVM. This sub-section introduces various features extraction techniques used for this purpose.

Actually, in the proposed system, binary and multi-class classifications were performed to classify PV module defects. In binary classification, PV modules are classified as normal or abnormal using NB. Then, the abnormal is classified into defective, various surface abnormalities, and minor defects using Enhanced Multi-Layer Perceptron (EMLP).

In the present research, Support Vector Machines (SVM) are proposed to address Photovoltaic (PV) cells Defect Classification. It is an intelligent supervised machine learning algorithm. In SVM, a high-dimensional feature space is used to map the input vectors non-linearly into a very concise form .