Artificial neural network based control strategy for single input dual output converter for futurestic solar DC smart homes

Solar-based Direct Current (DC) home loads are available with different voltage ratings. The multiport DC-DC converter is suitable for integrating different DC sources and loads. This article presents the design and implementation of an Artificial Neural Network (ANN) controller for the Single-Input Dual-Output (SIDO) converter. ANN performance is compared with the conventional Proportional-Integral (PI) controller for the SIDO converter. A SIDO converter has a 48 V input and regulated 12V and 24V output ports, with a 300W power rating. The ANN controller has resulted in 3.55% less voltage ripple, quicker rise time, and a faster settling time, compared to the PI controller at the 24V port. The ANN controller resulted in 1% lower voltage ripple, a faster rise time, and a faster settling time compared to the PI controller at the 12V port. The validation of the ANN Controller has shown that the design is appropriate for controlling the converter.