Solar PV Fault Classification using Back Propagation Neural Network 
Poonam Shinde1, S. R. Deore2
1Poonam Shinde, Student, Department of Electrical Engineering, A.C. Patil College of Engineering, Kharghar, Mumbai University, India.
2Dr. S. R. Deore, Professor, Department of Electrical Engineering, A.C. Patil College of Engineering, Kharghar, Mumbai University, India.
Manuscript received on February 28, 2020. | Revised Manuscript received on March 22, 2020. | Manuscript published on March 30, 2020. | PP: 5568-5574 | Volume-8 Issue-6, March 2020. | Retrieval Number: F9790038620/2020©BEIESP | DOI: 10.35940/ijrte.F9790.038620
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Solar energy is that the foremost abundant, inexhaustible and clean of all renewable energy resources. Interest in electrical solar PV power generation has accumulated in recent years due to its benefits. This wide distribution of physical phenomenon panel production wasn’t followed by watching, fault detection and designation functions to verify higher gain. In this paper, real time fault analysis and fault detection is done by using Back propagation. By simulating various fault conditions, the performances of a faulty electrical solar photovoltaic module have been compared with respect to its faultless model by quantifying the precise differential residue which can be associated with it. The deformations and faults induced on the I-V curves and P-V curves have been studied to generate data for neural network analysis for different types of faults. Five different fault cases like module to module fault, module – ground faults, short circuit fault, and different shading patterns of modules and solar cells are considered. The MATLAB simulation model’s results show the respective results for various fault conditions along with variation of different solar irradiation which commonly occur in the photovoltaic systems. The projected technique is often generalized and extended to additional sorts of faults. This faults condition was analyzed by using Backpropagation Based Neural Network (BP-ANN). Back propagation technique ensures fine tuning the weights of neural network to get lower error rates making the model more reliable, therefore the BP-ANN technique contributes in improving the overall accuracy for fault detection in the system using Artificial Neural Network.
Keywords: Neural Network, Solar PV system faults, fault detection, Back propagation.
Scope of the Article: Classification.