Optimized Design Parameters for the Bidirectional Isolated Boost DC-DC Converter Using Particle Swarm Optimization

Abstract

This paper describes dual active bidirectional buck–boost converter for battery charging/discharging application. The dual active bridge converter can charge and discharge the battery level with the wide voltage variation. When bidirectional DC-DC converter operates in the boost mode, an alternating path must be provided to reduce the circulating current to minimize the conduction losses. The RCD (a combination of resistance, capacitance, and diode) is the best way to provide the path for the circulating current which creates the voltage spike in the system. RCD snubber circuit provides an alternating path for circulating current; it also creates the I²R losses in the system though it requires the accurate design parameters which can trade between voltage spike versus system losses. The particle swarm optimization (PSO) technique has been widely used for tuning purpose. The novel approach has been used in this paper to obtain the optimized design parameters using this PSO to reduce the copper losses in the system which is caused by RCD parameters. So PSO has been implemented to identify the best possible solution for RCD to the trade-off between losses and voltage spikes. The bidirectional converter system battery voltage has been considered in the wide range 60–320 V, and the output voltage has been considered 650 V with 20 kHz switching frequency. The loss analysis has been carried out with the optimal solution obtained from PSO and without the optimal solution of the system. The simulation result proves that optimal solution obtained from the PSO has lower losses compared to the optimal solution provided by genetic algorithm (GA).