Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 5
Encontrada(s) 1 página(s)

e_Buah Biblioteca Digital Universidad de Alcalá
Publicación Artículo científico (JournalArticle)

A deep reinforcement learning approach to DC-DC power electronic converter control with practical considerations

  • Mazaheri Tehrani, Nafiseh
  • Santamargarita Mayor, Daniel|||0000-0003-3848-5886
  • Bueno Peña, Emilio José|||0000-0003-0401-4576
  • Pizarro Pérez, Daniel|||0000-0003-0622-4884
  • Cóbreces Álvarez, Santiago|||0000-0002-3308-4613
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., Agencia Estatal de Investigación, Junta de Comunidades de Castilla-La Mancha, Universidad de Alcalá
e_Buah Biblioteca Digital Universidad de Alcalá
Publicación Contribución de congreso (ConferenceOutput)

Control H∞ de bajo orden basado en pasividad para Dual Active Bridge

  • Pérez Ruiz, Juan José
  • Santamargarita Mayor, Daniel|||0000-0003-3848-5886
  • Molinero Blanco, David
  • Huerta Sánchez, Francisco|||0000-0002-4901-752X
  • Pizarro Pérez, Daniel|||0000-0003-0622-4884
  • Cóbreces Álvarez, Santiago|||0000-0002-3308-4613
  • Griñó Cubero, Robert
XXXI Seminario Anual de Automática, Electrónica Industrial e Instrumentación (SAAEI2024), 03/07/2024-05/07/2024, Granada, España., Este trabajo presenta una metodología de diseño para el lazo de control de tensión de salida de un Dual Active Bridge (DAB). Se basa en la síntesis de controladores lineales (de bajo orden) con un método de optimización para problemas no convexos y no suaves combinada con especificaciones de pasividad sobre la función de transferencia de impedancia de lazo cerrado, la cual tiene un impacto crítico en la estabilidad del sistema. La propuesta se basa en las ventajas de las técnicas modernas, haciendo uso del diseño basado en pesos frecuenciales, así como del modelado basado en cajas negras, sin elevar la complejidad del controlador., Junta de Comunidades de Castilla-La Mancha, Universidad de Alcalá
e_Buah Biblioteca Digital Universidad de Alcalá
Publicación Contribución de congreso (ConferenceOutput)

Non-smooth H♾ impedance shaping for DC catenary stability enhancement in railway converter

  • Huerta Sánchez, Francisco|||0000-0002-4901-752X
  • Cóbreces Álvarez, Santiago|||0000-0002-3308-4613
  • Bueno Peña, Emilio José|||0000-0003-0401-4576
  • Toro Zarza, José Manuel del
2024 IEEE Applied Power Electronics Conference and Exposition (APEC), 25/02/2024-29/02/2024, Long Beach, Estados Unidos., In railway systems, the constant power load behavior of both traction and auxiliary power system converters interacts negatively with the variable impedance of the catenary and can lead to voltage fluctuations and compromise the stability of the railway line. This paper proposes the use of a structured H? synthesis to design an AC voltage controller for the APS converter that models the converter?s DC input admittance. The resulting low-order controller guarantees catenary stability, ensures tracking of the AC output voltage reference, and provides robustness against non-modeled uncertainties. Experimental evaluation demonstrates its compliance with railway standards and modeling of the specified impedance., Junta de Comunidades de Castilla-La Mancha, Universidad de Alcalá, Agencia Estatal de Investigación
e_Buah Biblioteca Digital Universidad de Alcalá
Publicación Contribución de congreso (ConferenceOutput)

Artificial neural network based thermal model for a three-phase medium frequency transformer

  • Molinero Blanco, David
  • Santamargarita Mayor, Daniel|||0000-0003-3848-5886
  • Bueno Peña, Emilio José|||0000-0003-0401-4576
  • Marrón Romera, Marta|||0000-0001-7723-2262
  • Vasic Matic, Miroslav
32nd IEEE International Symposium on Industrial Electronics, 19/06/2023-21/06/2023, Helsinki, Finland., This paper proposes an Artificial Neural Network (ANN) as a method to thermally analyze a three-phase medium frequency transformer (MFT). This transformer is part of a 50kW three-phase Dual Active Bridge (DAB). After choosing the most suitable architecture for the ANN, it is trained with the results of finite element method (FEM) simulations that model its behavior. The ANN is thus able to calculate temperature increments of the core and of each winding with high reliability. This performance of the ANN is studied by comparing its results with some other from widely used state-of-the-art theoretical thermal models that emulate the transformer?s thermal behavior. The neural network-based solution is proved to be as accurate as the FEM simulations ground truth, and as fast to apply as the theoretical thermal model, therefore fusing the advantage of using any of these two in power converter design optimization., Junta de Comunidades de Castilla-La Mancha, Universidad de Alcalá
e_Buah Biblioteca Digital Universidad de Alcalá
Publicación Contribución de congreso (ConferenceOutput)

Dual Active Bridge simultaneous input admittance passivity shaping and reference tracking using low order H-infinity control

  • Pérez Ruiz, Juan José
  • Santamargarita Mayor, Daniel|||0000-0003-3848-5886
  • Molinero Blanco, David
  • Huerta Sánchez, Francisco|||0000-0002-4901-752X
  • Pizarro Pérez, Daniel|||0000-0003-0622-4884
  • Cóbreces Álvarez, Santiago|||0000-0002-3308-4613
2024 IEEE Applied Power Electronics Conference and Exposition (APEC), 25/02/2024-29/02/2024, Long Beach, Estados Unidos., This paper presents a design methodology for the output voltage control loop of a dual active bridge (DAB). It is based on fixed-order H? non-smooth synthesis combined with passivity restrictions over closed-loop impedance transfer function, known to have a critical impact on system voltage stability. The proposal inherits the advantages of modern techniques, making use of simple frequential weighting-based design and plant dynamical complexity black-boxing, while keeping controller complexity low., Junta de Comunidades de Castilla-La Mancha, Universidad de Alcalá, Agencia Estatal de Investigación