Neural Estimator for Inductor Losses in Buck DC-DC Converters Operating in CCM
Gabriele Maria Lozito, Vittorio Bertolini, Francesco Riganti Fulginei, Elisa Belloni, Michele Quercio
Abstract
In this work, a methodology to assess the losses related to the main inductor in a Buck DC-DC converter is proposed. The losses are related to the current waveform and the magnetic response of the inductor core. An Artificial Neural Network is used to estimate the losses for given operating conditions of the DC-DC converter. The neural estimator is trained and validated using real data from an experimental workbench, producing as output both the per-period energy loss and an equivalent circuit model useful for inclusion in transfer functions and small signal circuit analysis.
Topics & Concepts
InductorConvertersBuck converterWaveformPulse-width modulationComputer scienceEstimatorElectronic engineeringArtificial neural networkControl theory (sociology)Boost converterEngineeringElectrical engineeringVoltageMathematicsArtificial intelligenceStatisticsControl (management)Advanced DC-DC ConvertersMultilevel Inverters and ConvertersSilicon Carbide Semiconductor Technologies