Please use this identifier to cite or link to this item:
https://hdl.handle.net/20.500.13091/5737
Title: | Deep Learning Stack LSTM based MPPT Control of Dual Stage 100 kWp Grid-Tied Solar PV System | Authors: | Younas, U. Kulaksiz, A.A. Ali, Z. |
Keywords: | Adaptation models Artificial neural networks Complexity theory Computer architecture Deep learning Deep Learning Deep Neural Network Grid Integration Heuristic algorithms Long short term memory Long Short-Term Memory Network Maximum power point trackers Maximum Power Point Tracking Optimization Photovoltaic systems Power grids Solar Photovoltaic Solar power generation Brain Controllers Electric power system control Energy harvesting Fossil fuels Gas emissions Greenhouse gases Heuristic algorithms Learning algorithms Long short-term memory Maximum power point trackers Memory architecture Network architecture Solar energy Solar power generation Time domain analysis Two term control systems Adaptation models Complexity theory Deep learning Grid integration Heuristics algorithm Long short-term memory network Maximum Power Point Tracking Memory network Optimisations Photovoltaic systems Power grids Solar photovoltaics Optimization |
Publisher: | Institute of Electrical and Electronics Engineers Inc. | Abstract: | Rising global energy demand, predominantly satisfied by fossil fuels, triggers fuel price surges, fuel scarcity, and substantial greenhouse gas emissions. Solar photovoltaics (PV), as an abundant renewable alternative, can potentially address this demand, yet low cell efficiency (15-25%) and fluctuating output power due to intermittent irradiance (<italic>G</italic>) and temperature (<italic>T</italic>) impedes grid integration. This paper presents a novel Deep Learning (DL) based stacked LSTM (Long Short-Term Memory) MPPT controller to maximize power harvesting from a 100 kW grid-tied solar PV system, demonstrating superiority over conventional Perturb & Observe (P&O) and Feed Forward-Deep Neural Network (FF-DNN) MPPT approaches. Subsequently, a Neutral-Point-Clamped (NPC) 3-level inverter with proportional-integral (PI) controllers regulates the DC link voltage and transfers the extracted PV power to the grid. The proposed MPPT methodology includes collection of one million-sample (<italic>G</italic>, <italic>V</italic>, <italic>Vmp</italic>) datasets; preprocessing via z-score normalization; visualizing distributions through histograms and correlation matrix plots; an 80/20 split rule-based training and test sets; a two-hidden layer stacked LSTM (64 and 32 neurons) architecture; hyperparameters including the Adam optimizer, 0.05 learning rate, 32 batch size, and 50 epochs. Model efficacy quantification uses MSE, RMSE, MAE, loss, and R2 metrics. For 100 kW generated PV power, the stacked LSTM extracts 98.2 kW, versus 96.1 kW and 94.3 kW for the DNN and P&O MPPTs respectively. By integrating the optimized proposed stack LSTM MPPT with a streamlined inverter topology, the proposed approach advances the state-of-the-art in DL based solar PV energy harvesting optimization and grid integration. Authors | URI: | https://doi.org/10.1109/ACCESS.2024.3407605 https://hdl.handle.net/20.500.13091/5737 |
ISSN: | 2169-3536 |
Appears in Collections: | Scopus İndeksli Yayınlar Koleksiyonu / Scopus Indexed Publications Collections WoS İndeksli Yayınlar Koleksiyonu / WoS Indexed Publications Collections |
Show full item record
CORE Recommender
Items in GCRIS Repository are protected by copyright, with all rights reserved, unless otherwise indicated.