Development and Analytical Study of a High-Gain Boost Converter for Renewable Energy Systems
Vempada priyanka1, Dadi kalavathi2
1 PG Student , EEE & Sankethika Vidya parishad College of Engineering, Visakhapatnam
2 Assistant professor OF EEE Department, EEE & Sankethika Vidya parishad College of Engineering, Visakhapatnam
ABSTRACT
High-gain DC-DC converters are becoming increasingly common in solar PV systems and renewable energy applications. This article presents a non-isolated, non-coupled inductor-based high-gain DC-DC boost converter that offers high voltage gain at reduced duty ratios while ensuring low voltage stress on controlled power switches. The proposed converter is well-suited for boosting low-input DC voltage from distributed generation sources, such as fuel cells or photovoltaic (PV) systems, to a significantly higher DC voltage. With just two switches controlled by a single PWM signal, the topology simplifies control, reduces weight, minimizes cost, and enhances compactness. To further optimize performance, a machine learning-based predictive algorithm using a paragraph model is integrated into the converter's control system. This model analyzes historical and real-time data to dynamically adjust the duty cycle, improving voltage regulation and efficiency under varying load and input conditions. By leveraging machine learning, the converter can predict optimal switching patterns, reducing losses and enhancing stability in renewable energy applications. A comparative analysis with existing high-gain boost converters demonstrates that the proposed model outperforms previous topologies across multiple performance metrics. A 300 W hardware prototype is developed and tested in a laboratory environment to validate the theoretical claims. The proposed topology achieves a high gain of approximately 12 times the input voltage, with a reduced duty ratio—11.25 at a duty cycle of 0.6 and 17.77 at a duty cycle of 0.7. Efficiency ranges between 92.5% and 94.5%, making it suitable for medium-to-high power applications. The integration of machine learning further enhances system adaptability and operational efficiency, making the converter an ideal solution for next-generation sustainable energy systems requiring high output voltage and improved performance.
