PWM-Geometric Modeling and Centric Control of Basic DC–DC Topologies for Sleek and Reliable Large-Signal Response

PWM-Geometric Modeling and Centric Control of Basic DC–DC Topologies for Sleek and Reliable Large-Signal Response Traditional linear techniques are widely used for the control of basic dc-dc converters due to their simple implementation. However, due to the small-signal validity range of the models employed, the converters usually perform poorly under large transients, and the dynamic response can be improved only to a limited extent in order to ensure a stable behavior. On the other hand, faster dynamic response can be achieved with boundary controllers, which require faster sensors and more powerful processors. A novel control scheme that combines the advantages of fixed-frequency pulsewidth modulation with state-plane geometric analysis is introduced to obtain fast and reliable large-signal response. The natural evolution of the average state variables is described by a large-signal unified model, which provides the basis to develop a reliable nonlinear control scheme. The proposed technique is suitable for implementation in low-cost digital signal processors, using low-bandwidth sensing stages, and it features fast, sleek, and consistent dynamic response with constant switching frequency. Since the model developed accurately predicts the large-signal behavior, reliable and predictable responses can be obtained at any operating point. In this way, the transient response obtained shows reduced, consistent, and well-determined peaks in inductor current and capacitor voltage, avoiding magnetic saturation and system failures even during extremely large transients. Furthermore, the maximum rating specifications for the reactive components can be reduced, which, combined with the low requirements for sensors and processors, lowers the implementation cost and makes the controller a very appealing alternative for high-volume applications. The contributions made to the theoretical and applied field are valid for any combination of reactive components due to the normalized approach adopted. The theoretical concepts are supported by detailed mathematical procedures. The proposed theor- and controller are validated by experimental results.