A purely data-driven modeling method based on subspace system identification for DC-DC converters

A purely data-driven modeling approach based on subspace system identification is proposed for establishing state-space models of DC-DC converters suitable for control system design. Unlike traditional modeling methods, this approach eliminates the need for prior knowledge of circuit topology or analytical equations. Instead, it leverages collected input-output data to accurately estimate the system's structure and parameters. When applied to DC-DC converters, the method achieves an impressive model fitting degree of approximately 97%, highlighting its effectiveness in practical applications.