Analysing the Neo-Fuzzy Neural Network for Capacitor and Supercapacitor Health

• Published in: International Conference on System Modeling & Advancement in Research Trends (Online) pp. 779 - 782
• Main Authors: Laudya, Ravi, Dharuman, Narrain Prithvi, Mall, Rakshita, Bikshapathi, Mahaveer Siddagoni, Jain, Arpit, Kshirsagar, Rajas Paresh
• Format: Conference Proceeding
• Language: English
• Published: IEEE 06.12.2024
• Subjects: Artificial Neural Network; Artificial neural networks; Capacitors; Fluctuation; Fluctuations; Levenberg-Marquart; Monitoring; Power electronics; Predictive models; Supercapacitors; Testing; Transportation; Validation
• ISBN: 9798350380569
• ISSN: 2767-7362
• DOI: 10.1109/SMART63812.2024.10882529

Evaluation of Neo-Fuzzy Neural Networks for Health Monitoring of Capacitors and Supercapacitors This study presents supercapacitor models using Artificial Neural Networks (ANN s). A black box nonlinear multiple input., single output system serves as the foundation for the theory. The supercapacitor voltage is the output of the system., while the temperature and current are its inputs. Establishing the link between inputs and outputs is done via the ANN model learning and validation using experimental charges and discharges of supercapacitors. The testing results of the 2700F., 3700F., and supercapactior packages are used in the learning and validation of an ANN model. The ANN model can forecast the behaviour of a supercapacitor with temperature fluctuations after the network has been trained. The Levenberg-Marquardt approach is used to update the ANN model's parameters to reduce the error between the system's anticipated output and actual outcome. The results produced using the supercapacitor ANN model and the experimental ones agree rather well.