Artificial Neural Networks-Based Energy Storage Predictor of (Ba0.85Ca0.15) (Ti0.9Zr0.1) O3 under Temperature-Induced Variation | ||||
| SVU-International Journal of Engineering Sciences and Applications | ||||
| Volume 6, Issue 1, June 2025, Page 1-8 PDF (4.32 MB) | ||||
| Document Type: Original research articles | ||||
| DOI: 10.21608/svusrc.2024.297015.1220 | ||||
 View on SCiNiTO | ||||
| Authors | ||||
Dina A. Naser  1; Mohamed Hassan Essai2; Abd El-Razek Mahmoud3; G. A. Gamal3
| ||||
| 1basics science, High Institute of Engineering &Technology Luxor- Tod, Luxor, Egypt | ||||
| 2communication, faculty of engineering, Al-Azhar University, Qena | ||||
| 3Physics Department, Faculty of Science, South Valley University, Qena, Egypt | ||||
| Abstract | ||||
| Comprehending and predicting the fluctuations in the energy storage functionality of ferroelectric-based apparatuses throughout a broad range of temperatures is crucial. To achieve this, we developed and simulated a Function Fitting ANN model using MATLAB. The model was trained using the back-propagation algorithm, effectively capturing the relationship between the applied electric field, and resulting polarization through experimental data. The model demonstrates excellent performance with two hidden layers consisting of 37 neurons in each and three input layers. Extensive experimentation confirms the model's impressive accuracy in predicting energy storage performance, particularly at different temperature conditions around Curie temperature Tc. The experimental part of the study was done in the temperature range (43-95 ᵒC) which seems to be limited. However, it justifies temperature-induced changes around the Curie temperature (TC). Above curie temperature (T > TC), the material becomes paraelectric and loses its spontaneous polarization resulting in more decrease in recoverable energy-storage density and efficiency. The remarkable predictive performance of the model is attributed to its remarkably low mean square error of 3.68×10-5. This result emphasizes the model's precision and reliability in accurately forecasting energy storage parameters. Finally, BCZT ceramic samples were selected for the present work for being a very famous ferroelectric material and has well-known ferroelectric properties. | ||||
| Keywords | ||||
| Artificial Neural Networks; Energy Storage Parameters; Ceramic Capacitors; Temperature Variation; Ferroelectric Hysteresis Loops | ||||
|
Statistics Article View: 198 PDF Download: 184 |
||||
