5 January 2022 in IEEE Transactions on Control of Network Systems
This article proposes a data-driven method for distributed frequency control of islanded microgrids based on multiagent quantum deep reinforcement learning (DRL). The proposed method combines the conventional DRL framework with quantum machine learning, and can adaptively obtain the optimal cooperative control strategy. The microgrid secondary frequency control is organized in a distributed manner in which each agent performs the control action only based on the local and neighboring information. To solve the DRL problem, the deep deterministic policy gradient algorithm is derived to tune the agents’ parameters. Simulation tests are performed on an islanded microgrid with four distributed generators and a 13-bus microgrid. The results demonstrate that the proposed method can effectively regulate the frequency with better time-delay tolerance, and displays the quantum advantage in parameter reduction.
23 February 2022 in IEEE Transactions on Smart Grid
Blockchain technology is recognized as a suitable tool to secure the energy trading because it could perfectly match the distributed structure of peer-to-peer (P2P) energy market. But its usage is stuck on the transaction level. Control systems are significant to the microgrid as they ensure a stable power delivery system and regulate the performance of parameters such as active power and frequency. This paper proves that the blockchain technology is also effective in securing the distributed control systems against the false data injection attack. A six-prosumer microgrid is tested with the implementation of the hierarchical blockchain system. The security of both the control system and energy trading system of the microgrid is ensured. Smart contracts are created to calculate the feedback measurements for the control system and execute the energy transactions. According to the hierarchical structure, the private blockchain with static nodes is implemented for the distributed control to match the sampling rate. A Proof-of-Authority based blockchain is utilized to support the energy trading. In addition, a double auction based simple iteration (DA-SI) pricing scheme is designed to improve the social welfare of the microgrid. Finally, case studies are presented to verify the proposed hierarchical blockchain system as an effective method to safeguard the control system and maximize the benefits of prosumers. Numerical results show the effectiveness and feasibility of the proposed approach.
13 January 2022 in IEEE Transactions on Industrial Informatics
Control systems are significant to the microgrid as they regulate performance parameters such as frequency, active power, and voltage. Distributed control systems allow direct communication between the secondary controllers and controls the parameters efficiently. To secure each distributed control process and ensure a good quality of control results, a proof-of-authority private blockchain is applied in this article to defend the distributed control system against various types of cyber-attacks such as false data injection. A four-distributed generation islanded microgrid is tested with the implementation of the blockchain. Smart contracts are created to calculate the control feedback and return the value to corresponding secondary controllers. All of the four nodes are initially assigned as the authority nodes to share the mining burden, but according to the proof-of-authority consensus protocol, the authority role could be excluded if the node behaves illegally and causes damage to the control system. In addition, different attacking scenarios are categorized and analyzed with their respective solutions. Finally, a case study is introduced to verify the corresponding solutions and proves that the proposed method is able to secure the distributed control system while ensuring the control quality. Numerical results show the effectiveness and feasibility of the proposed approach.
15 March 2022 in IEEE Transactions on Industry Applications
The integration of information and communication technologies has introduced cybersecurity challenges to microgrid (MG) apart from bringing more flexibility, reliability, and resilience. Especially in the upcoming quantum era, the existing cryptographic systems are vulnerable to attacks from quantum computers. Motivated by these challenges, this article proposes a quantum-key-distribution (QKD)-based MG distributed control framework for cybersecurity enhancement. To defend the side-channel attacks and make the framework practical enough for industrial applications, measurement-device-independent QKD (MDI-QKD) is introduced. The following contributions are made in this work. 1) A novel QKD-based quantum-secure control architecture is established to assure the data transmission security in MG distributed control. 2) MDI-QKD with asymmetric protocol is used to form a scalable QKD network for MG control. 3) A method for fast parameter optimization based on deep neural network (DNN) is proposed to adjust the parameters in QKD systems in real-time. Simulations are carried on a 13-bus MG to demonstrate the working principles of the proposed control scheme. The test results also prove the feasibility of the DNN-based parameter optimizer.
23 November 2020 in IEEE Transactions on Power Systems
The distributed generators (DGs) and energy storages systems (ESSs) will have different droop characteristics for power sharing in autonomous microgrids (MGs). In previous research, the distributed secondary control is proposed for only one specific type of units, and relies on periodic communications among neighbors. This paper proposes a unified distributed real power sharing control scheme based on the dynamic event-triggered mechanism for droop-governed DGs and ESSs. A unified power sharing strategy is proposed where the power are fairly shared among all DGs and ESSs considering their different power ratings and capacities, while the SoCs among ESSs are balanced. In the meantime, a distributed event-triggered control scheme with a novel dynamic triggering mechanism is designed to address the formulated problem. The proposed controllers are triggered adaptively based on the MG system states. Besides, the proposed event-triggered mechanism is sample-based and only uses the largest and the second smallest eigenvalues of the Laplacian matrix to design the controller gain and eventtriggered parameters. To validate the proposed control design, a MG test system with detailed component models is built in the OPAL-RT real-time simulator. The results demonstrate the effectiveness and performance of the proposed controller.
4 March 2025 in IEEE Transactions on Industrial Cyber-Physical Systems
In Cyber-Physical Systems, maintaining cyber-resilience is critical due to the prevailing asymmetry that benefits attackers, especially concerning Denial of Service attacks that target system availability. In response, this paper introduces a cross-domain approach, utilizing Software-Defined Networking to adapt communication topology and control gain scheduling, thereby enhancing cyber-resilience against attacks. A cyber-resilience validation platform is developed, integrating hardware-in-the-loop with domain-specific simulation tools, to ensure accurate replication of real-world cyber-physical scenarios. An integral aspect of this framework is the real-time exploitation of metrics for agile responses to attacks. Validating our approach with an IEEE 34-bus test feeder for the physical domain and Cisco Modeling Lab for the cyber domain shows a 43% to 66% reduction in average latency compared to scenarios without our proposal during attacks, alongside enhanced rewards in both communication and physical networks. These findings accentuate the efficacy of the proposed cross-domain SDN approach in enhancing the cyber-resilience.