An Experiential Learning Framework for EV Charging Systems Using AI-Driven Harmonic Suppression and Advanced Quadrature Signal Generators

Abstract

With the growing use of fast Electric Vehicle (EV) charging systems, a set of complicated issues arises with regard to power quality, such as harmonic distortion, reactive power imbalance, and voltage instability. These obstacles necessitate sophisticated methods of control that in many cases students find hard to conceptualize on the basis of conventional modes of teaching. The research suggest an education-based framework that are incorporate the Model Predictive Control (MPC), the Wavelet Neural Networks (WNN), and the Quadrature Signal Generators (QSG) into a learning environment based on simulation in engineering. The suggested framework allows the students to comprehend the actual real-time problems of power quality with the help of interactive modeling and adaptive AImediated control methods. Two stage architecture is designed whereby the grid-side control is a finite-control-set MPC and adaptive control is a WNN-based controller. The system is instituted as a virtual laboratory module where the learners can see the harmonic suppression, reactive power compensation and power factor improvement. Education results show that students gain a better conceptual knowledge of the nonlinear systems, smart grid control, and AI-based optimization, and the analytical and problem-solving capabilities improve significantly. Multidisciplinary learning is also supported by the framework across the boundaries of power electronics, artificial intelligence, and energy systems. The contribution to engineering education is that this work takes a complex real-life problem of EV charging control and makes it an interactive and student-focused learning experience, in accordance with the contemporary pedagogical principles like experiential learning and outcome-based learning.