Search

For the droop-controlled distributed generation units (DGs) in microgrids (MGs), the power sharing error associated with voltage difference of DGs can be attributed to the mismatched line impedance. To improve the power sharing accuracy in islanded MGs, a neighbor-prediction based hierarchical control strategy is proposed. In secondary control, a consensus protocol is used to regulate the DGs voltages interacted by communication network to the state of tracking synchronization. Thus, the error-free power sharing is obtained. Network data-loss and delay problems are considered. To stabilize the DGs’ inner-loop control in which there is a delayed input, the H∞ robustness criteria of a dynamic model of DGs is derived. If the DG i ’s voltage in network transmission loses, the prediction system of a neighbor will be activated to forecast the lost voltage for secondary consensus control by extreme learning machine (ELM). Thus, the voltages regulation will be back to normalization. Simulations in MATLAB confirm the effectiveness of proposed control in MGs. [ABSTRACT FROM AUTHOR]

• A voltage regulation optimization problem is formulated considering source-load-storage collaboration. • An attention-based multi-agent deep reinforcement learning approach is designed for voltage control. • A multi-stage parallel training framework and an evolutionary selection mechanism is adopted to improve the scalability of voltage control. With the increasing penetration of renewable energy in active distribution networks (ADNs), voltage regulation problem is becoming more and more challenging. In this article, we focus on providing a scalable data-driven approach to ensure the voltage security of ADNs with high penetration of PVs. To this end, we first formulate an optimization problem for real-time voltage control considering source-load-storage collaboration while minimizing the total active power curtailment of PVs. Then, we reformulate the above problem as a Markov game and propose a novel voltage regulation algorithm based on evolutionary curriculum-based multi-agent deep reinforcement learning (EC-MADRL) to solve it. The key idea of the proposed algorithm is to adopt a multi-stage parallel training framework based on attention multi-agent deep deterministic policy gradient algorithm (Attention-MADDPG) and use an evolutionary mechanism to select voltage regulation models with high fitnesses in the previous stage, which means that good agents with best adaptability could be utilized for learning in the environment with a larger number of PVs. Simulation results show the effectiveness and scalability of the proposed algorithm. [ABSTRACT FROM AUTHOR]