The integration of variable photovoltaic and wind energy into renewable power systems has created a high demand for high-frequency converters that achieve low switching loss, reduced thermal stress, and stable output regulation. In the case of conventional hard-switching converters, turn-on and turn-off losses are significant at higher switching frequencies, resulting in electromagnetic interference (EMI) and device stress. The objective of this work is to devise a soft-switching power converter structure comprising photovoltaic and wind inputs together with resonant energy transfer, transformer isolation, output filtering and aggregation leveraging coordinated Pulse-Width Modulation (PWM) and frequency. Here, Zero Voltage Switching (ZVS) is attained for the primary switch and Zero Current Switching (ZCS) is reached for the auxiliary switch using a resonant inductor–capacitor network. The 5-kW converter operates from a nominal 300 V input, regulates to a 400 V output and switches at 100 kHz with resonant inductance of ${2 0}~{\mu} \mathrm{H}$ and resonant capacitance of 126 nF. Compared to conventional switching ($94.7\%$ and $93.8\%$, respectively), simulation results demonstrate peak efficiency of $97.3\%$ and rated-load efficiency of $97.0\%$. The result yields a switching loss of 126 W down to a final value of 38 W and a total estimated loss reducing from 297 W to a peak of 190 W in balance-of-systems loss metrics conducive for use in photovoltaic–wind hybrid generation systems, distributed renewable plants, battery interfaces, microgrids, and more effective grid-connected power conversion with enhanced thermal- and electromagnetic-response performance under variable renewable conditions across realistic operating ranges.
Muthukumar Paramasivan· International Conference on...· 0 citations
The adoption rate of electric mobility, renewable energy systems, and smart transportation infrastructures has exacerbated the demand for real-time, high-performance and energy-efficient systems. While high latency, low bandwidth, and low responsiveness are commonly encountered drawbacks in existing cloud-based energy optimization techniques used in these mobility-driven systems. The deployment of Edge AI and IoT technologies will pave the path to efficient, low-latency and real-time distributed renewable energy optimization within the smart sustainable transportation system. This paper presents a detailed survey on the latest trends in AI based renewable energy integration, smart grid, EV, battery management systems, and real-time transportation data analytics. The application of deep learning, reinforcement learning, federated learning, and predictive analytics toward improved renewable energy generation forecasting, smart charging, load balancing, and Vehicle-to-Grid (V2G) co-ordination is also elaborated. Edge computing platforms and IoT devices can support a high-performance real-time monitoring, predictive maintenance and a self-sufficient energy distribution network for smart transportation. The experimental validation on contemporary research works reveal an average 70% reduction in transmission latency, more than 50% increase in the renewable energy consumption, approximately 30% increase in battery lifetime, and nearly 80% reduction in charging infrastructure offline duration for edge systems with AI assistance. Important concerns regarding the edge system’s security, scale-ability, connectivity, privacy, and the direction of future research on enabling smart transportation were also discussed.
Muthukumar Paramasivan, Manikandan Sivasubramanian, K. Alagar et al.· Proceedings of the Instituti...· 0 citations