In recent years, there has been a growing interest in using model-free deep reinforcement learning (DRL)-based controllers as an alternative approach to improve the dynamic behavior, efficiency, and other aspects of DC–DC power electronic converters, which are traditionally controlled based on small signal models. These conventional controllers often fail to self-adapt to various uncertainties and disturbances. This paper presents a design methodology using proximal policy optimization (PPO), a widely recognized and efficient DRL algorithm, to make near-optimal decisions for real buck converters operating in both continuous conduction mode (CCM) and discontinuous conduction mode (DCM) while handling resistive and inductive loads. Challenges associated with delays in real-time systems are identified. Key innovations include a chattering-reduction reward function, engineering of input features, and optimization of neural network architecture, which improve voltage regulation, ensure smoother operation, and optimize the computational cost of the neural network. The experimental and simulation results demonstrate the robustness and efficiency of the controller in real scenarios. The findings are believed to make significant contributions to the application of DRL controllers in real-time scenarios, providing guidelines and a starting point for designing controllers using the same method in this or other power electronic converter topologies.

Losses and maximum temperature are important indicators of the health status of the medium-frequency magnetic components. Both the losses and the maximum temperature are very difficult to obtain/estimate due to the high complexity, change of losses in the core with respect to the average temperature, or the variation of the losses of the windings derived from the construction of the litz wire and the high-frequency effects. This article proposes the use of a real-time estimator, based on artificial neural networks trained from finite element method simulations capable of predicting with an error less than 2% in all cases both the maximum internal temperatures and the losses of a medium frequency transformer, with forced convection, which implies a very complex thermal behavior.

This paper analyzes the possibility to employ Artificial Neural Networks (ANN) for thermal modeling of magnetic components, with special emphasis on Medium Frequency Transformers (MFTs). The objective is to obtain a MFT thermal model that can be easily used in multivariable optimizations and automated design of power electronics converters, when it is necessary to rapidly explore the complete design space, optimizing the system from the point of view of the efficiency and power density, without losing the accuracy. This can be accomplished using low computational cost ANNs that are trained using the results obtained from thermal Finite Element Method (FEM) simulations. Nevertheless, depending on the approximations that are often applied to speed up the simulations (without modifying the mesh) and, consequently, to reduce the dataset generation time, the precision of the simulations and ANN based models vary significantly, especially in the case when the forced air cooling is used. For example, we demonstrate that different approaches can accelerate the dataset generation (460 simulations) by up to 40 hours, but would introduce an error of up to 40 °C. Therefore, a trade-off of the different alternatives to generate the ANN dataset is analyzed in this paper. The proposed design and optimization approach have been verified using the experimental platform, a Medium Frequency Transformer (MFT), designed and optimized for a 7.5 kW Dual Active Bridge (DAB), designed by brute force optimization using the ANN trained with FEM simulations. The experimental results obtained in different operating points show errors lower than 2.6 °C for the estimated average surface temperature. Additionally, the FEM obtained results are compared with the results obtained by state-of-the-art simplified analytical models and simplified FEM simulations, clearly showing the necessity of highly accurate FEM simulations to generate the training dataset. Finally, we discuss the drawback of the proposed ANN approach in the cases when the ANN based model should be scaled, trying to design a MFT with characteristics outside the range that was used for ANN training. We show that one possible solution for this issue may be ANN re-training, allowing to reduce the error from 50 °C to 5 °C by using only 40 new data points.

This paper proposes a new cascaded multilevel converter topology based on three-phase H bridge cells with a common DC-link structure. The proposed multilevel converter topology main advantages, compared with literature renowned multilevel converters topologies, are discussed in the paper, such as modularity, construction, implementation cost, and DC voltage ripple mitigation. Despite presenting an elementary structure and easy implementation, the use of classic PWM switching strategies is not feasible for this topology, causing the appearance of several short-circuit states between its capacitors. Thus, a graph theory algorithm combined with a model predictive control is also proposed in this work to identify and avoid the new cascaded multilevel converter short-circuit switching states and, concomitantly, guaranteeing the converter output power quality. In order to validate the presented topology applicability, a low voltage synchronous static compensators (STATCOM) with an optimal switching vector model predictive control (OSV-MPC) is implemented in a hardware-in-the-loop platform. The real-time experimental results prove the proposed multilevel topology and the OSV-MPC control strategy effectiveness.

This paper proposes a new cascaded multilevel converter topology, presenting some inherent advantages to its construction and implementation cost when compared to the most used multilevel converters in the literature. As a disadvantage, the new topology presents several short-circuit states, which, if used with classic PWM switching strategies, can cause damage to internal components making its implementation unfeasible. Thus, an analysis based on Graph Theory combined with a model predictive control is proposed in order to identify and avoid all possible short-circuit states of the proposed converter The proposed topology is applied to a STATCOM by means of an Optimal Switching Vector Model Predictive Control (OSV-MPC) in order to prove its usability. Simulation results in Simulink MATLAB of such application are also presented.

Heat sinks are a fundamental component of power electronics converters, so it is important to have a reliable method to study and optimize their size. Thermal analysis of heat sinks can be a complex problem as it involves different heat transfer mechanisms, and it is often necessary to use finite element simulations to obtain accurate results. However, these simulations, being very slow, are relegated to the validation process. This paper proposes a thermal model of heat sinks based on artificial neural networks. The model, unlike previous state-of-the-art models that only obtain the average temperature of the heat sink, is able to obtain a thermal map of the heat sink surface, as if it were an image, by using convolutional layers. The main advantage of this approach is that using these convolutional layers, the model is able to efficiently process how the elements are distributed on the heat sink. This model, valid for heat sinks of very different sizes in both laminar and turbulent flow, has an error of less than 1.5% and is 1500 times faster than finite element simulations, so it can be easily used in brute-force optimization processes, where many different designs need to be analyzed.