A Nonlinear Control Allocation Strategy for Dual Half Bridge Power Converters

This work focuses on designing nonlinear control algorithms for dual half-bridge converters (DHBs). We propose a two-layer controller to regulate the current and voltage of the DHB. The first layer utilizes a change in the control variable to obtain a quasi-linear representation of the DHB, allowing for the application of simple linear controllers to regulate current and power flow. The second layer employs a nonlinear control allocation algorithm to select control actions that fulfill (pseudo) power setpoints specified by the first control layer; it also minimizes peak-to-peak currents in the DHB and enforces voltage balance constraints. We apply the DHB and this new control strategy to manage power flow in a hybrid energy storage system comprising of a battery and supercapacitors. Numerical simulation results demonstrate that, in comparison with state-of-the-art approaches, our control algorithm is capable of maintaining good transient behavior over a wide operating range, while reducing peak-to-peak current by up to 80%. Note to Practitioners—Dual half bridges (DHB) are an important class of converters that is used in a wide range of practical applications, including automotive and renewable energies. Despite their benefits—such as galvanic isolation, zero-voltage switching, and bidirectional power flow—the design of control algorithms for this converter is challenging due to its highly nonlinear response. In this work, we provide a new control framework that allows the designer to obtain a linear input-to-state representation of the converter and to “cancel” the dominant nonlinearities. The paper also presents the application of the parameter space tuning technique, which allows the designer to synthesize DHB controllers that fulfill transient, disturbance rejection, and robustness requirements. The tuning approach provides a set of possible controller gains that fulfill the control specifications (and not just a single value). This enables the application engineer to perform fine-tuning of the controller gains within a pre-certified parameter space that satisfies the control requirements, providing another important practical benefit.