Improving primary frequency response in networked microgrid operations using multilayer perceptron‐driven reinforcement learning

Abstract

Individual microgrids can improve the reliability of power systems during extreme events, and networked microgrids can further improve efficiency through resource sharing and increase the resilience of critical end-use loads. However, networked microgrid operations can be subject to large transients due to switching and end-use loads, which can cause dynamic instability and lead to system collapse. These transients are especially prevalent in microgrids with high penetrations of grid-following inverter-connected renewable energy resources, which do not provide the system inertia or fast frequency response needed to mitigate the transients. One potential mitigation is to engage the existing generator controls to reduce system voltage in response to a frequency deviation, thereby reducing load and improving primary frequency response. This study investigates the use of a reinforcement-learning-based controller trained over several switching transient scenarios to modify generator controls during large frequency deviations. Compared to previously used proportional-integral controllers, the proposed controller can improve primary frequency response while adapting to changes in system topologies and conditions.

Control and optimisation of networked microgrids: A review

“…The characteristics of conventional droop control, i.e. P − 𝜔 and • Hierarchical control • Large signal model • Uncertainty of RES is not considered [38] • Power exchange control • Model predictive control • Generation uncertainties has ignored [22] • Distributed control • Adaptive neural network • ESS is not counted [19] • Distributed control • Droop control • RES and ESS are not considered [41,42] • Distributed control • Cluster-oriented control • Double-layer communication network [43] • Primary frequency control • Reinforcement learning • No experimental validation [44] Voltage stabilisation and generation cost minimisation…”

“…In [68], a genetic algorithm‐based control method was used to enhance system stability in photovoltaics (PV) MG clusters by addressing oscillation due to the dynamic nature of PV. The effect of a large transient due to inverter switching and end‐user load on system dynamic stability was analysed in [44] for an RES‐dominant NMG. A reinforcement learning‐based trained controller was adopted to avoid considerable frequency deviation by reducing system voltage set point.…”

A Brief Review on Identification, Categorization and Elimination of Power Quality Issues in a Microgrid Using Artificial Intelligent Techniques

“…In (Prabaakaran et al 2019;Mahmoud, Abouheaf, and Sharaf 2019;Bagheri et al 2018;Younesi, Shayeghi, and Siano 2020;Radhakrishnan et al 2020), the authors have made use of the Reinforcement Learning algorithm to control different power quality issues occurring in a microgrid.…”

“…From the above observations, the authors concluded that the RL-tuned PID technique performed better in mitigating voltage and frequency oscillations in the microgrid system. In order to establish a quicker frequency response to mitigate large frequency disturbances due to switching actions in a multiple interconnected microgrid network, the authors Nikita Radhakrishnan et al (Radhakrishnan et al 2020), formulated a control mechanism based on the RL algorithm for generator control refinement when subjected to large disturbances. They designed and simulated the network by using GridLAB-D simulation software and used the HELICS platform for controller communication.…”

Deep Reinforcement Learning for Control of Microgrids: A Review

“…A DRL-based optimal control scheme with additional safety features for voltage control problems has been developed as a continuous action space [27]. Furthermore, a Policy Approximation based on a multilayer perceptron neural network and feed-forward algorithm was proposed to synchronize frequencies and solve control problems in interconnected microgrids [28]. Frequency regulation in an islanded mode of AC microgrids can be improved using a controller based on a deep deterministic policy gradient algorithm.…”