The monitoring and control of DC-DC converters have become key issues since DC-DC converters are gradually playing increasingly crucial roles in power electronics applications such as electric vehicles and renewable energy systems. As an emerging and transforming technology, the digital twin, which is a dynamic virtual replica of a physical system, can potentially provide solutions for the monitoring and control of DC-DC converters. This work discusses the design and implementation of the digital twin DC-DC converter in detail. The key features of the physical and twin systems are outlined, and the control architecture is provided. To verify the effectiveness of the proposed digital twin method, four possible cases that may occur during the practical control scenarios of DC-DC converter applications are discussed. Simulations and experimental verification are conducted, showing that the digital twin can dynamically track the physical DC-DC converter, detect the failure of the physical controller and replace it in real time.

With the rapid development of cyber-physical microgrids (MGs) and distributed generations (DGs), the distributed secondary control scheme becomes a key component of the MGs' control framework. However, the introduction of communication networks inevitably brings about time delays. Against this backdrop, this article proposes a distributed networked predictive control (DNPC) strategy for DC MGs suffering from time-varying communication delays. The proposed strategy can regulate the DG's voltage to the desired value and optimize each DG's generation cost simultaneously. Explicitly, the multistep predictor generates predictions for both local nodes and neighbor DGs, using the pseudo-prediction-enabled method. Thus, the delayed information in control protocols can be replaced by the latest predictions, so as to realize the active compensation for delays. Furthermore, the analysis of the stability of the closed-loop MG system is presented to ensure the generality of the DNPC strategy. Finally, the efficiency of the proposed control system against the time-varying delay effect is verified by comprehensive experiments in a DC MG containing a wind turbine generator.

Microgrids have been identified as a viable solution to the integration of renewable distributed generations (DGs) into power systems, while the coordination of DGs is frequently hindered by nonideal communication networks. To eliminate the negative consequences of these communication constraints, this article proposes a distributed predictive-proportional-integral control strategy for dc microgrids subject to communication delays and packet dropouts. The proposed control method can achieve voltage regulation and proportional power sharing simultaneously in a distributed manner. To address the issues of time delay and packet loss, an active prediction method is recommended, in combination with the PID scheme, which makes use of the model information and historical data for predictive compensation as well as retain the benefits of PID controllers effectively. The closed-loop system of the dc microgrid under the proposed control scheme is presented for stability and consensus analysis. Finally, experimental results from a wind turbine-based dc microgrid system are provided to demonstrate the superiority and effectiveness of the proposed method.

To improve the poor dynamic consensus in multiple dc-dc converter systems caused by parameter differences, this article proposes a global optimal cooperative control strategy. The proposed control strategy achieves dynamic consensus of output currents by minimizing a global performance index, which consists of local errors and neighbor synchronization errors. In this system, each converter is regarded as an agent in the communication digraph, so as to construct a cyber-physical model. Accordingly, a cooperative control method based on the linear quadratic regulator (LQR) is proposed, with the help of nonlinear coordinate transformation techniques. Different from conventional LQR, integral action is introduced to eliminate steady-state errors and improve coordinated performance. Meanwhile, the proof of the stability and consensus of the closed-loop system is given using Lyapunov approach. Moreover, the design guidelines to tune the dynamic response are provided through simulations. The experimental results further demonstrate the effectiveness and superiority of the proposed control method in a multiconverter prototype.