Solar panel analysis with adversary neural networks

When solar panels received the irradiance from the sun, early detection is important to prevent fault or fast degradation. This research article provides a new method using 'Generative Adversary Neural Networks' [1] (GANN), with a deep learning techniques, for evaluation of the degradation in solar panels (SP). The methodology required root cause analysis for SP degradation, it considered four stages for the deep learning: 'preprocessing, segmentation, extraction, and classification' [11]. In this paper, we are determined artificial intelligence methodology and new neural network proposal for panel degradation detection based on root cause analysis [2]. The effectiveness of the results were 97.5%; with minimum information. However, the training process produces 0.105 % false positives.