<p>The electric vehicle demand keeps rising, concerning utility as environmental concerns, underscoring the critical need for a reliable charging infrastructure to encourage mass adoption. It emphasizes the need for renewable energy sources with solar power for electric vehicle charging as part of a sustainable transport system. Some challenges include intermittency of renewables, the fundamental role of battery storage, and intelligent energy management between power grids and electric vehicles. This research proposes a soft-switched high step-up quasi-Z-source converter, which has low voltage stress during switching, to improve solar photovoltaic and electric vehicle integration. The 3 degree-of-freedom fractional order tilt proportional integral derivative-tilt integral differentiator with filter controller values are fine-tuned using extended random neural networks. The prediction accuracy of extended random neural networks is enhanced by the adaptive fitness-distance balance-based geyser-inspired algorithm. The research evaluates the performance of the proposed system using the Python platform and compares it with various existing methods. A high voltage gain of 34.5 and a low switch voltage stress of 0.38 V for the duty cycle of 0.45 and a minimum error of 0.00055 was achieved during verification of converter operation with prototype validation.</p>