26 July 2021 in IEEE Transactions on Smart Grid
This paper aims to crack the individual EV charging scheduling problem considering the dynamic user behaviors and the electricity price. The uncertainty of the EV charging demand is described by several factors, including the driver’s experience, the charging preference and the charging locations for realistic scenarios. An aggregate anxiety concept is introduced to characterize both the driver’s anxiety on the EV’s range and uncertain events. A mathematical model is also provided to describe the anxiety quantitatively. The problem is formulated as a Markov Decision Process (MDP) with an unknown state transition function. The objective is to find the optimal sequential charging decisions that can balance the charging cost and driver’s anxiety. A model-free deep reinforcement learning (DRL) based approach is developed to learn the optimal charging control strategy by interacting with the dynamic environment. The continuous soft actor-critic (SAC) framework is applied to design the learning method, which contains a supervised learning (SL) stage and a reinforcement learning (RL) stage. Finally, simulation studies verify the effectiveness of the proposed approach under dynamic user behaviors at different charging locations.
7 March 2019 in IEEE Transactions on Industrial Electronics
Hybrid energy storage system (HESS) consisting of battery and supercapacitor (SC) is an effective approach to alleviate voltage stability problems brought by fluctuation of renewable resources in a DC microgrid. This paper proposes a novel distributed control scheme for multiple HESSs based on a leaderless consensus protocol to realize the power splitting between batteries and SCs and thus regulate the DC bus voltage. Without assigning leaders among the batteries and SCs, each SC or battery has equal priority in participating in voltage or state of charge (SOC) regulation using only neighboring information exchange instead of global communication. This distributed approach makes the system robust against any physical failure or communication fault only if the remaining communication network is connected. Consensus for the state variable associated with the battery SOC, the SC terminal voltage, and the DC average bus voltage are all achieved in this scheme, and thus the voltage quality and the energy storage lifetime can be improved. In addition, the small-signal stability of a four-HESS system is investigated, and the impact of the control parameters on the system is analyzed. Simulations and experimental studies are conducted to demonstrate the effectiveness of the proposed control scheme in an islanded DC microgrid.
18 February 2022 in IEEE Transactions on Industrial Informatics
The growth of electric vehicles (EVs) significantly increases the residential electricity demand and potentially leads to the overload of the transformer in the distribution grid. Aiming to coordinate the charging control of EVs, this article formulates the EVs charging problem as a Markov game with an unknown transition function and proposes a cooperative charging control strategy based on the multiagent deep reinforcement learning. The uncertainties from the dynamic electricity price, non-EV residential load consumption and drivers’ individual behaviors are considered to construct the dynamic charging environment. Each agent contains a collective-policy model and an independent learner. The collective-policy model is introduced to model other agent's behaviors by approximating their power consumption. The independent learner is used to learn the optimal charging strategy by interacting with the environment. The soft-actor-critic framework is adopted to train the independent learner, enabling the proposed method to address the continuous state and action. Agents are trained with only the local observation and approximation, indicating that the proposed approach is fully decentralized and scalable to the problem with multiple agents. Finally, several numerical studies constructed based on the real-world data demonstrate the effectiveness and scalability of the proposed approach.
31 October 2019 in IEEE Transactions on Power Systems
The conventional voltage observer based distributed control can make a tradeoff between voltage restoration and reactive power sharing. But it suffers from estimation error when there exists input disturbances, such as initial values and communication delays. To address this issue, this paper proposes a novel distributed secondary control based on PI consensus algorithm. A voltage controller and a frequency controller are designed separately to realize the restoration of voltage/frequency and also accurate reactive/active power allocation. Moreover, a detailed small-signal model with PI consensus algorithm is established to show the impact of the proposed control strategy on system stability. Compared with the conventional distributed control, the proposed control scheme is simpler with higher control accuracy under input disturbances. Simulations are conducted to verify the effectiveness of the proposed control strategy in an isolated ac microgrid.
1 January 2022 in Renewable and Sustainable Energy Reviews
20 November 2018 in IEEE Transactions on Industrial Electronics
In this paper, a distributed cooperative control scheme, considering the State of Charge (SoC) balance and power limits of Battery Storage Units (BSUs) in the consensus protocol, is proposed to achieve average bus voltage consensus in a DC microgrid with photovoltaics (PVs). The state variable defined in the consensus protocol enables all BSUs to charge or discharge together without introducing circulating currents among BSUs. Furthermore, by incorporating power limits of BSUs in the defined state variable, power violation of the BSUs can be prevented. The consensus-based distributed control only utilizes neighbor-to-neighbor communication to realize the global consensus and thus avoids the single point of failure compared to the conventional centralized control. Simulation and experimental results demonstrate the efficacy of the proposed method in an islanded DC microgrid under different testing scenarios.