Improved SMPS modeling for photovoltaic applications by a novel neural paradigm with Hamiltonian-based training algorithm

• Published in: 2015 International Conference on Clean Electrical Power (ICCEP) pp. 723 - 730
• Main Authors: Bonanno, F., Capizzi, G., Lo Sciuto, G.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01-06-2015
• Subjects: Boost converter; Computational modeling; Dynamic modeling; Hamiltonian formulation; Inductors; Integrated circuit modeling; Mathematical model; Photovoltaic; Recurrent neural network; Simulation; SMPS; Switched-mode power supply; Switches; Training; Training algorithm
• DOI: 10.1109/ICCEP.2015.7177571

This paper discuss as the dynamics of a SMPS can be investigated by recurrent neural network (RNN) based models with an Hamiltonian formulation and function used for the training, so leading to a novel paradigm that we call RNNHT model. By using the calculated state variables in a boost converter a RNN is trained by considering also the minimization of the energy stored according to a defined cost function. Simulation results show the improvements in the dynamic performance output prediction versus some well assessed boost converter models in the recent literature.