Learning-Based Model Predictive Control of DC-DC Buck Converters in DC Microgrids: A Multi-Agent Deep Reinforcement Learning Approach

Abstract

This paper proposes a learning-based finite control set model predictive control (FCS-MPC) to improve the performance of DC-DC buck converters interfaced with constant power loads in a DC microgrid (DC-MG). An approach based on deep reinforcement learning (DRL) is presented to address one of the ongoing challenges in FCS-MPC of the converters, i.e., optimal design of the weighting coefficients appearing in the FCS-MPC objective function for each converter. A deep deterministic policy gradient method is employed to learn the optimal weighting coefficient design policy. A Markov decision method formulates the DRL problem. The DRL agent is trained for each converter in the MG, and the weighting coefficients are obtained based on reward computation with the interactions between the MG and agent. The proposed strategy is wholly distributed, wherein agents exchange data with other agents, implying a multi-agent DRL problem. The proposed control scheme offers several advantages, including preventing the dependency of the converter control system on the operating point conditions, plug-and-play capability, and robustness against the MG uncertainties and unknown load dynamics.

Adaptive backstepping controller design for DC/DC buck converter optimised by grey wolf algorithm

“…By replacing Equation (10) in Equation ( 12), the differentiation of Lyapunov Function can be negative semi-definite. This condition verifies the stability of the subsystem as it is shown in Equation (13).…”

“…Therefore, modern and more improved methods are designed to compensate for the limitations addressed by the classical schemes. [8][9][10][11][12][13][14] listed some of the most effective controllers used for power converters, including Sliding mode, Fuzzybased, and Model predictive control techniques. These methods are more desirable for practical usages as they depict better output regulation, lower sensitivity to more comprehensive disturbances, and higher control adaptability.…”

Hybrid Intelligent Control System for Adaptive Microgrid Optimization: Integration of Rule-Based Control and Deep Learning Techniques

“…In recent years, there has been considerable emphasis on building hybrid control systems for MG management by combining deep learning methods with rule-based control [49][50][51][52]. These hybrid approaches aim to integrate rule-based systems' interpretability with deep neural network learning capabilities to provide more flexible and intelligent control techniques.…”

Sliding Mode-based Model Predictive Control of Grid-Forming Power Converters

“…Including WFs, however, cannot always be bypassed, as the number of objectives is so much which is difficult to be unified. To calculate the WFs, techniques that employ artificial intelligence [17]- [20], genetic algorithms [21], and fuzzy control [22] were proposed. On the other hand, the approach presented in [17] employs an offline computation process of the WF and the flux reference, achieving a quick drive start and satisfactory response under various loading scenarios.…”