Ir arriba
Información del artículo

Energy management of a microgrid considering nonlinear losses in batteries through Deep Reinforcement Learning

D. Domínguez-Barbero, J. García-González, M.A. Sanz-Bobi, A. García-Cerrada

Applied Energy Vol. 368, pp. 123435-1 - 123435-12


The massive deployment of microgrids could play a significant role in achieving decarbonization of the electric sector amid the ongoing energy transition. The effective operation of these microgrids requires an Energy Management System (EMS), which establishes control set-points for all dispatchable components. EMSs can be formulated as classical optimization problems or as Partially-Observable Markov Decision Processes (POMDPs). Recently, Deep Reinforcement Learning (DRL) algorithms have been employed to solve the latter, gaining popularity in recent years. Since DRL methods promise to deal effectively with nonlinear dynamics, this paper examines the Twin-Delayed Deep Deterministic Policy Gradient (TD3) performance – a state-of-the-art method in DRL – for the EMS of a microgrid that includes nonlinear battery losses. Furthermore, the classical EMS-microgrid interaction is improved by refining the behavior of the underlying control system to obtain reliable results. The performance of this novel approach has been tested on two distinct microgrids – a residential one and a larger-scale grid – with a satisfactory outcome beyond reducing operational costs. Findings demonstrate the intrinsic potential of DRL-based algorithms for enhancing energy management and driving more efficient power systems.

Resumen divulgativo:

Modelar adecuadamente las baterías debería ayudar a operar eficientemente el sistema eléctrico. Las técnicas de aprendizaje por refuerzo profundo (DRL) operan sin dificultad con modelos no lineales. Este trabajo estudia el impacto del DRL en una micro-red aislada operando con un modelo no lineal de baterías.

Palabras Clave: Deep Reinforcement Learning; Energy management system; Energy savings; Isolated microgrid; Nonlinear battery model

Índice de impacto JCR y cuartil WoS: 11,200 - Q1 (2022)

Referencia DOI: DOI icon

Publicado en papel: Agosto 2024.

Publicado on-line: Mayo 2024.

D. Domínguez-Barbero, J. García-González, M.A. Sanz-Bobi, A. García-Cerrada, Energy management of a microgrid considering nonlinear losses in batteries through Deep Reinforcement Learning. Applied Energy. Vol. 368, pp. 123435-1 - 123435-12, Agosto 2024. [Online: Mayo 2024]

    Líneas de investigación:
  • Despacho óptimo de generación con red para sistemas híbridos AC/DC
  • Redes inteligentes
  • Integración de energía renovable
  • Planificación y operación de recursos energéticos distribuidos
  • Electrónica de potencia