Photovoltaic Plant Yield Prediction Using Deep Learning Networks

• Author(s): Gaurang Gupta ; Sanyam Ahuja ; Vrinda Goel ; Deshendra Sihag
• Paper ID: 1706318
• Page: 573-578
• Published Date: 01-10-2024
• Published In: Iconic Research And Engineering Journals
• Publisher: IRE Journals
• e-ISSN: 2456-8880
• Volume/Issue: Volume 8 Issue 3 September-2024

Abstract

One of the most significant sources of renewable energy is solar energy. To maintain the dependability and effectiveness of solar energy conversion for PV systems, defects that arise during operation must be identified and addressed quickly. This study presents a comprehensive analysis of the predictive maintenance technology used for solar photovoltaic (PV) modules. The primary objective is to implement an algorithm based on time-series data that can identify and predict any potential faults related to the power output of PV cells. Using weather predictions as the base data, we perform a comparative analysis with RNN frameworks like Long Short Term Memory (LSTM) and Bidirectional LSTM models and Gated Recurrent Units (GRUs). The performance of the models was evaluated by comparing predicted values with actual values obtained from two test sites located in India. Our results show that GRUs outperform LSTM and Bidirectional LSTM (Bi-LSTM) networks by a margin of 4.45e-01. Our proposed framework gives a mean squared error of the order e-05 on the validation data of the solar power generation dataset.

Keywords

PV Module, Maintenance, LSTM, RNN, GRU

Citations

IRE Journals:
Gaurang Gupta , Sanyam Ahuja , Vrinda Goel , Deshendra Sihag "Photovoltaic Plant Yield Prediction Using Deep Learning Networks" Iconic Research And Engineering Journals Volume 8 Issue 3 2024 Page 573-578

IEEE:
Gaurang Gupta , Sanyam Ahuja , Vrinda Goel , Deshendra Sihag "Photovoltaic Plant Yield Prediction Using Deep Learning Networks" Iconic Research And Engineering Journals, 8(3)