Showing papers in "IEEE Transactions on Smart Grid in 2022"

Fast Grid Frequency and Voltage Control of Battery Energy Storage System Based on the Amplitude-Phase-Locked-Loop

Abstract: This paper presents a novel fast frequency and voltage regulation method for battery energy storage system (BESS) based on the amplitude-phase-locked-loop (APLL). In the proposed method, the primary frequency regulation and inertia emulating control are designed based on grid frequency deviation ( ${\Delta }f$ ) and its differential $(df/dt)$ signals detected by APLL, avoiding complicated and sensitive differential operation to get better dynamic support performance. Moreover, unlike the traditional voltage control only using grid voltage deviation ( ${\Delta }V$ ), the proposed method also uses the voltage differential control ( $dV/dt$ -control) to improve dynamic performance of traditional voltage control and behaviors of grid voltage based on the $dV/dt$ signal from APLL. Finally, simulations and experiments are implemented to validate the effectiveness of the proposed APLL-based BESS on improving the dynamic behaviors of grid frequency and voltage.

Reinforcement Learning for Selective Key Applications in Power Systems: Recent Advances and Future Challenges

Abstract: With large-scale integration of renewable generation and distributed energy resources, modern power systems are confronted with new operational challenges, such as growing complexity, increasing uncertainty, and aggravating volatility. Meanwhile, more and more data are becoming available owing to the widespread deployment of smart meters, smart sensors, and upgraded communication networks. As a result, data-driven control techniques, especially reinforcement learning (RL), have attracted surging attention in recent years. This paper provides a comprehensive review of various RL techniques and how they can be applied to decision-making and control in power systems. In particular, we select three key applications, i.e., frequency regulation, voltage control, and energy management, as examples to illustrate RL-based models and solutions. We then present the critical issues in the application of RL, i.e., safety, robustness, scalability, and data. Several potential future directions are discussed as well.

Detection of False Data Injection Attacks in Smart Grid: A Secure Federated Deep Learning Approach

Abstract: As an important cyber-physical system (CPS), smart grid is highly vulnerable to cyber attacks. Amongst various types of attacks, false data injection attack (FDIA) proves to be one of the top-priority cyber-related issues and has received increasing attention in recent years. However, so far little attention has been paid to privacy preservation issues in the detection of FDIAs in smart grids. Inspired by federated learning, a FDIA detection method based on secure federated deep learning is proposed in this paper by combining Transformer, federated learning and Paillier cryptosystem. The Transformer, as a detector deployed in edge nodes, delves deep into the connection between individual electrical quantities by using its multi-head self-attention mechanism. By using federated learning framework, our approach utilizes the data from all nodes to collaboratively train a detection model while preserving data privacy by keeping the data locally during training. To improve the security of federated learning, a secure federated learning scheme is designed by combing Paillier cryptosystem with federated learning. Through extensive experiments on the IEEE 14-bus and 118-bus test systems, the effectiveness and superiority of the proposed method are verified.

Detection of False Data Injection Attacks in Smart Grid: A Secure Federated Deep Learning Approach

Abstract: As an important cyber-physical system (CPS), smart grid is highly vulnerable to cyber attacks. Amongst various types of attacks, false data injection attack (FDIA) proves to be one of the top-priority cyber-related issues and has received increasing attention in recent years. However, so far little attention has been paid to privacy preservation issues in the detection of FDIAs in smart grids. Inspired by federated learning, a FDIA detection method based on secure federated deep learning is proposed in this paper by combining Transformer, federated learning and Paillier cryptosystem. The Transformer, as a detector deployed in edge nodes, delves deep into the connection between individual electrical quantities by using its multi-head self-attention mechanism. By using federated learning framework, our approach utilizes the data from all nodes to collaboratively train a detection model while preserving data privacy by keeping the data locally during training. To improve the security of federated learning, a secure federated learning scheme is designed by combing Paillier cryptosystem with federated learning. Through extensive experiments on the IEEE 14-bus and 118-bus test systems, the effectiveness and superiority of the proposed method are verified.

Multi-Stage Real-Time Operation of a Multi-Energy Microgrid With Electrical and Thermal Energy Storage Assets: A Data-Driven MPC-ADP Approach

Abstract: This paper studies the multi-stage real-time stochastic operation of grid-tied multi-energy microgrids (MEMGs) via the hybrid model predictive control (MPC) and approximate dynamic programming (ADP) approach. In the MEMG, practical power and thermal network constraints, heterogeneous energy storage devices, and distributed generations are involved. Given the relatively large thermal inertia and slow thermal energy fluctuation, only uncertainties of renewable energy sources and active/reactive power loads are considered. Then, historical data are adopted as training scenarios for the MPC-ADP method to acquire empirical knowledge for dealing with all the diverse uncertainties. Further, piecewise linear functions are used to approximate value functions with respect to the operation status of energy storage assets, which enables sequentially solving the Bellman’s equation forward through time to minimize MEMG operation cost. Finally, numerical case studies are conducted to illustrate the effectiveness and superiority of the proposed MPC-ADP approach. Simulation results indicate that with sufficient information embedded, the MPC-ADP approach could obtain good-enough real-time operation solutions with the successively updated forecast. Further, it outperforms alternative real-time operation benchmarks in terms of optimality and convergence for various application scenarios.

Distributed Robust Model Predictive Control-Based Energy Management Strategy for Islanded Multi-Microgrids Considering Uncertainty

Abstract: A microgrid is considered to be a smart power system that can integrate local renewable energy effectively. However, the intermittent nature of renewable energy causes operating pressure and additional expense in maintaining the stable operation by the energy management system in a microgrid. The structure of multi-microgrids provides the possibility to construct flexible and various energy trading framework. In this paper, in order to reduce the adverse effects of uncertain renewable energy output, a distributed robust model predictive control (DRMPC)-based energy management strategy is proposed for islanded multi-microgrids. This strategy balances the robustness and economy of single-microgrid system operation by combining the advantages of robust optimization and model predictive control, while coping with the uncertainty of renewable energy sources. Furthermore, a dynamic energy trading market is formed among microgrids, which can enhance the overall economy of the multi-microgrids system. Simulation results verify the feasibility of the proposed DRMPC strategy.

A Distributed MPC to Exploit Reactive Power V2G for Real-Time Voltage Regulation in Distribution Networks

Abstract: It has been demonstrated theoretically and experimentally that the Vehicle-to-Grid (V2G) enabled electric vehicle (EV) charger is of a reactive power compensation ability with a battery or capacitor connected. To exploit the aggregated reactive power V2G abilities of massively dispersed EV chargers, a distributed model predictive control (DMPC) strategy applying to both balanced and unbalanced distribution networks (DNs) is proposed to integrate them into real-time DN voltage regulation. Firstly, based on the instantaneous power theory and voltage sensitivity matrices, a voltage regulation model considering the reactive response of EV chargers is established without violating the normal EV active charging demands. Then, a completely distributed framework is achieved by DMPC, in which prediction information and calculation results are shared in a Peer-to-Peer (P2P) way to realize the asynchronous broadcast. The proposed model and techniques are validated by numerical results obtained from the IEEE European low voltage test feeder system. The case studies indicate that the proposed DMPC is robust to communication latency (CML) and works effectively in both balanced and unbalanced DNs without any control center, which is a significant advantage for the promotion of real-time reactive power V2G in DNs with irregular user integration and relatively poor communication infrastructure.

An Overview of Soft Open Points in Electricity Distribution Networks

Abstract: Soft open points (SOPs) are power electronic devices that are usually placed at normally open points of electricity distribution networks to provide flexible power control to the networks. This paper gives a comprehensive overview of both academic research and industrial practice on SOPs in electricity distribution networks. The topologies of SOPs as multi-functional power electronic devices are identified and compared, which include back-to-back voltage source converters, multi-terminal voltage source converters, unified power flow controllers, and direct AC-to-AC modular multilevel converters. The academic research is reviewed in three aspects, i.e., benefit quantification, control, and optimal siting and sizing of SOPs. The benefit quantification indices are categorized into feeder load balancing, voltage profile improvement, power losses reduction, three-phase balancing and DG hosting capacity enhancement. The control of SOPs is summarized as a three-level control structure, where the system-level and converter-level control are further discussed. For optimal siting and sizing of SOPs, problem formulation and solution methods are analyzed. Besides the academic research, practical industrial projects of SOPs worldwide are also summarized. Finally, opportunities of research and industrial application of SOPs are discussed.

Multi-Objective Optimization for Smart Integrated Energy System Considering Demand Responses and Dynamic Prices

Abstract: The coordinated implementation of demand response technology and dynamic energy prices facilitates the interaction among multiple stakeholders in the smart integrated energy system. To achieve the optimal operation between different entities in the system, this paper proposes a multi-objective optimization model for smart integrated energy system considering demand responses and dynamic prices that reflects the preferences of multiple stakeholders. Based on the tightly coupled characteristics of a multi-energy system, a flexible two-dimensional demand response model with spatio-temporal coupling characteristics is established. By analyzing the characteristics of multi-entities joint pricing, the dynamic energy price formulated with the participation of both supplier and demander is optimized, and the dynamic price control strategy of different stakeholders under different benefit weights is obtained. A case study verifies the effectiveness of the proposed method. According to the interest preferences of different entities, different strategies and operating mechanisms can be derived, which is conducive to improving the economy and reliability of the operation of the smart integrated energy system and promoting the interaction between multiple entities.

Resilience Assessment and Improvement for Cyber-Physical Power Systems Under Typhoon Disasters

Abstract: The cyber-physical deep coupling makes power systems face more risks under small-probability and high-risk typhoon disasters. Resilience describes the ability of cyber-physical power system (CPPS) withstanding extreme disasters and resuming normal operation. To improve the resilience assessment and analysis method of CPPS, first, a CPPS resilience assessment framework that considers the space-time metrics of disasters and the interactions of information systems and power grids is proposed, including fault scenarios extraction, response and recovery analysis, quantitative assessment of resilience. Second, from the perspective of the geographical coupling between OPGW and transmission lines and the control coupling between automatic generation control system (AGC), substation automation system (SAS) and power system, the interaction of information flow and energy flow during the failure period is analyzed. The network flow theory is used to establish an information network traffic model to describe the operating status of the information system at each stage. On this basis, a mixed integer linear programming model for DC optimal power flow considering the information network constraints and a multi-stage bi-level model for cyber-physical collaborative recovery are established. Finally, we take the IEEERTS-79 system as an example to show that the proposed method can improve the quantization accuracy comparing with the assessment method of the conventional power system, and evaluate the enhancement of typical measures at different stages.

Peer-to-Peer Energy Trading and Energy Conversion in Interconnected Multi-Energy Microgrids Using Multi-Agent Deep Reinforcement Learning

Abstract: A key aspect of multi-energy microgrids (MEMGs) is the capability to efficiently convert and store energy in order to reduce the costs and environmental impact. Peer-to-peer (P2P) energy trading is a novel paradigm for decentralised energy market designs. In this paper, we investigate the external P2P energy trading problem and internal energy conversion problem within interconnected residential, commercial and industrial MEMGs. These two problems are complex decision-making problems with enormous high-dimensional data and uncertainty, so a multi-agent deep reinforcement learning approach combining the multi-agent actor-critic algorithm with the twin delayed deep deterministic policy gradient algorithm is proposed. The proposed approach can handle the high-dimensional continuous action space and aligns with the nature of P2P energy trading with multiple MEMGs. Simulation results based on three real-world MG datasets show that the proposed approach significantly reduces each MG’s average hourly operation cost. The impact of carbon tax pricing is also considered.

A High-Impedance Fault Detection Method for Distribution Systems Based on Empirical Wavelet Transform and Differential Faulty Energy

Abstract: High-impedance faults (HIFs) pose the greatest challenge for distribution system protection, especially for microgrids and distribution networks with distributed generators (DGs) that have flexible operation modes. This paper analyzes the faulty features of HIFs and proposes a HIF detection method that uses empirical wavelet transform (EWT) and differential faulty energy. The proposed method is as follows. First, the various time-frequency components are obtained by utilizing the EWT to decompose the differential faulty energy and adaptively select the feature component with the largest permutation entropy. Second, the permutation variance index is constructed based on the sample point number and feature component energy, and then it is employed to detect HIFs. Finally, low voltage microgrid simulation tests, medium voltage distribution system integrated by DG simulation tests, and field tests show that the proposed method can correctly distinguish HIFs from normal disturbances, including operation mode switches, load switches, capacitor switches, and DG switches. The advantages of the proposed method are also elaborated in detail, from signal preprocessing and feature extraction.

Decentralized Coordination and Stabilization of Hybrid Energy Storage Systems in DC Microgrids

Abstract: Hybrid energy storage system (HESS) is an attractive solution to compensate power balance issues caused by intermittent renewable generations and pulsed power load in DC microgrids. The purpose of HESS is to ensure optimal usage of heterogeneous storage systems with different characteristics. In this context, power allocation for different energy storage units is a major concern. At the same time, the wide integration of power electronic converters in DC microgrids would possibly cause the constant power load instability issue. This paper proposes a composite model predictive control based decentralized dynamic power sharing strategy for HESS. First, a composite model predictive controller (MPC) is proposed for a system with a single ESS and constant power loads (CPLs). It consists of a baseline MPC for optimized transient performance and a sliding mode observer to estimate system disturbances. Then, a coordinated scheme is developed for HESS by using the proposed composite MPC with a virtual resistance droop controller for the battery system and with a virtual capacitance droop controller for the supercapacitor (SC) system. With the proposed scheme, the battery only supplies smooth power at steady state, while the SC compensates all the fast fluctuations. The proposed scheme achieves a decentralized dynamic power sharing and optimized transient performance under large variation of sources and loads. The proposed approach is verified by simulations and experiments.

Grid-Forming Inverter Enabled Virtual Power Plants With Inertia Support Capability

Abstract: Modern energy systems are experiencing the transition towards renewable-powered ones. Some conventional thermal units based on synchronous machines are gradually decommissioned and replaced by power electronics interfaced renewables. Thus, the lack of natural inertia and governor damping, which are the features of synchronous machines, raises significant concern about system frequency stability, including the faster rate of change and lower nadir point of frequency. Meanwhile, with the rapid development of communication and Internet of Things technologies, distributed energy resources can be aggregated as a virtual power plant to help balance real-time electricity demand and supply. However, the capability of utilizing the whole virtual power plant to provide adjustable inertia support has not been explored yet. In this paper, we propose a framework of the synchronous virtual power plant based on grid-forming inverter interfaced distributed energy resources. By coordinating the parameter settings of grid-forming inverters, the virtual power plant provides inertia support. Also, we design an online learning-based parameter settings method that makes the inertia of the virtual power plant adjustable. A case study in IEEE 34 nodes system illustrates the effectiveness of the proposed method.

Optimal Energy Management for Multi-Microgrid Under a Transactive Energy Framework With Distributionally Robust Optimization

Abstract: The increasing penetrations of renewable energy and electric vehicles bring more uncertainties and challenges to the existing power grid. The coordinated networked microgrids (MGs) contain renewable distributed generations (DGs) and nonrenewable DGs, which will be an important component in the future. We formulate an optimization problem based on a transactive energy (TE) framework for the energy schedule of upstream network and networked MGs to minimize the operation cost. The energy management between MGs and upstream network is operated by the distribution system operator (DSO), which is different from the direct control signal and fixed pricing mechanism in the traditional power system. We develop a distributionally robust optimization algorithm with ambiguity set based on Wasserstein distance (DROW) to solve the optimization problem with the uncertainties from real-time electricity price, renewable energy, loads, and electric vehicles. We carry out case studies about the energy schedule of the modified IEEE 33-bus and IEEE 118-bus power system with networked MGs. Numerical results indicate that the TE framework is conducive to schedule the energy of upstream network and networked MGs efficiently with the dynamic pricing scheme and the proposed DROW algorithm can seek a robust energy schedule of DSO and networked MGs with uncertainties.

A Stacked Machine and Deep Learning-Based Approach for Analysing Electricity Theft in Smart Grids

Abstract: The role of electricity theft detection (ETD) is critical to maintain cost-efficiency in smart grids. However, existing methods for theft detection can struggle to handle large electricity consumption datasets because of missing values, data variance and nonlinear data relationship problems, and there is a lack of integrated infrastructure for coordinating electricity load data analysis procedures. To help address these problems, a simple yet effective ETD model is developed. Three modules are combined into the proposed model. The first module deploys a combination of data imputation, outlier handling, normalization and class balancing algorithms, to enhance the time series characteristics and generate better quality data for improved training and learning by the classifiers. Three different machine learning (ML) methods, which are uncorrelated and skillful on the problem in different ways, are employed as the base learning model. Finally, a recently developed deep learning approach, namely a temporal convolutional network (TCN), is used to ensemble the outputs of the ML algorithms for improved classification accuracy. Experimental results confirm that the proposed framework yields a highly-accurate, robust classification performance, in comparison to other well-established machine and deep learning models and thus can be a practical tool for electricity theft detection in industrial applications.

Grid-Forming Inverter Enabled Virtual Power Plants With Inertia Support Capability

Abstract: Modern energy systems are experiencing the transition towards renewable-powered ones. Some conventional thermal units based on synchronous machines are gradually decommissioned and replaced by power electronics interfaced renewables. Thus, the lack of natural inertia and governor damping, which are the features of synchronous machines, raises significant concern about system frequency stability, including the faster rate of change and lower nadir point of frequency. Meanwhile, with the rapid development of communication and Internet of Things technologies, distributed energy resources can be aggregated as a virtual power plant to help balance real-time electricity demand and supply. However, the capability of utilizing the whole virtual power plant to provide adjustable inertia support has not been explored yet. In this paper, we propose a framework of the synchronous virtual power plant based on grid-forming inverter interfaced distributed energy resources. By coordinating the parameter settings of grid-forming inverters, the virtual power plant provides inertia support. Also, we design an online learning-based parameter settings method that makes the inertia of the virtual power plant adjustable. A case study in IEEE 34 nodes system illustrates the effectiveness of the proposed method.

Federated Clustering for Electricity Consumption Pattern Extraction

Abstract: The wide popularity of smart meters enables the collection of massive amounts of fine-grained electricity consumption data. Extracting typical electricity consumption patterns from these data supports the retailers in their understanding of consumer behaviors. In this way, diversified services such as personalized price design and demand response targeting can be provided. Various clustering algorithms have been studied for electricity consumption pattern extraction. These methods have to be implemented in a centralized way, assuming that all smart meter data can be accessed. However, smart meter data may belong to different retailers or even consumers themselves who are not willing to share their data. In order to better protect the privacy of the smart meter data owners, this paper proposes two federated learning approaches for electricity consumption pattern extraction, where the k-means clustering algorithm can be trained in a distributed way based on two frequently used strategies, namely model-averaging and gradient-sharing. Numerical experiments on two real-world smart meter datasets are conducted to verify the effectiveness of the proposed method.

MPC-Based Decentralized Voltage Control in Power Distribution Systems With EV and PV Coordination

Abstract: Distribution systems are growing rapidly in size and complexity with increased penetrations of distributed energy resources (DER) and electric vehicles (EVs), leading to operational challenges. Reactive power compensation from photovoltaic (PV) inverters and active power curtailment of PV output are commonly used to mitigate voltage-related problems. However, the charging/discharging flexibility of EVs can be used to avoid PV output curtailment and save energy motivating methods to coordinate EV charging with PVs. Unfortunately, a large number of control variables makes the centralized voltage control computationally expensive. In this work, a decentralized voltage control algorithm which considers the active and reactive power compensation from PV inverters and EVs is presented. The proposed approach helps to solve the voltage issues in a more effective way. The approach involves clustering of the distribution network based on modified modularity, an index that considers EV flexibility and prediction time period. A model predictive control (MPC)-based algorithm is proposed for each cluster to solve the voltage problem using the respective PVs and EVs, while ensuring that the EV charging demand is satisfied while contributing to the voltage regulation. The proposed algorithm is validated using the IEEE 123 node test systems under two PV and EV penetration levels and shown to be effective in solving the voltage regulation problem.

Optimal Distributed Control of AC Microgrids With Coordinated Voltage Regulation and Reactive Power Sharing

Abstract: In this paper, we propose an optimal distributed voltage control for grid-forming (GFM) inverters in islanded AC microgrids. An optimization problem is formulated where the distributed generator (DG) output voltage is considered as the control variable with technical constraints on voltage and reactive power output capacity and an objective function that makes a trade-off between voltage regulation and reactive power sharing. A distributed primal-dual gradient based algorithm is developed to solve the formulated optimization problem to address the challenges due to non-separable objective function, unavailable global average voltage, and globally coupled reactive power constraints. The effectiveness of the proposed optimal distributed control is validated through simulations on the 4-DG test microgrid and the modified IEEE 34-bus distribution test system, and the advantages of the proposed control over existing controls are demonstrated.

Home Energy Recommendation System (HERS): A Deep Reinforcement Learning Method Based on Residents’ Feedback and Activity

Abstract: Smart home appliances can take command and act intelligently, making them suitable for implementing optimization techniques. Artificial intelligence (AI) based control of these smart devices enables demand-side management (DSM) of electricity consumption. By integrating human feedback and activity in the decision process, this work proposes a deep Reinforcement Learning (RL) method for managing smart devices to optimize electricity cost and comfort residents. Our contributions are twofold. Firstly, we incorporate human feedback in the objective function of our DSM technique that we name Home Energy Recommendation System (HERS). Secondly, we include human activity data in the RL state definition to enhance the energy optimization performance. We perform comprehensive experimental analyses to compare the proposed deep RL approach with existing approaches that lack the aforementioned critical decision-making features. The proposed model is robust to varying resident activities and preferences and applicable to a broad spectrum of homes with different resident profiles.

Hierarchical Blockchain Design for Distributed Control and Energy Trading Within Microgrids

Abstract: 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.

A Transformer-Based Method of Multienergy Load Forecasting in Integrated Energy System

Abstract: Multienergy load forecasting technique is the basis for the operation and scheduling of integrated energy system. Different types of loads in an integrated energy system, i.e., electricity load, heat load, cold load, might have complex and strong coupling relationships among them. If the internal relationship of multienergy load can be considered to realize joint prediction, the accuracy of multienergy load forecasting could be improved. This paper proposes a multi-task model, MultiDeT (Multiple-Decoder Transformer), which firstly adopts the one-encoder multi-decoder structure to realize the multi-task architecture and perform joint prediction of multienergy load. Based on the encoder-decoder architecture, the proposed model encodes all the input data by a uniform encoder and performs each forecasting task by multiple decoders. All tasks share the same encoder parameters, but have dedicated decoders for subtask learning. Therefore, the proposed multi-decoder structure can achieve different levels of attention to the output representation of the encoder by multi-head attention. The entire model is jointly trained end-to-end with losses from each task. Finally, the proposed model is tested on the publicly available datasets. Compared with other forecasting models, the results show that the proposed model has more accurate load forecasting results and has higher generalization capability.

Converter-Based Moving Target Defense Against Deception Attacks in DC Microgrids

Abstract: With the rapid development of information and communications technology in DC microgrids (DCmGs), the deception attacks, which typically include false data injection and replay attacks, have been widely recognized as a significant threat. However, existing literature ignores the possibility of the intelligent attacker, who could launch deception attacks once obtaining necessary information by exploiting zero-day vulnerabilities or bribing insiders, to affect the system in an unforeseeable manner. In this paper, based on the observation that the primary control law of the power converter device in DCmGs is usually programmable, we propose a novel converter-based moving target defense (CMTD) strategy by proactively perturbing the primary control gains to defend against deception attacks. First, we study the impact of perturbing the primary control gains on the voltage stability in DCmGs and provide explicit conditions for the perturbation magnitude and frequency under which the voltage stability can be ensured. Then, we investigate the improved detectability against deception attacks under CMTD and present sufficient conditions under which these attacks can be detected. Finally, we conduct extensive MATLAB SIMULINK/PLECS based simulations and systematic hardware-in-the-loop based experiments to validate the effectiveness of CMTD.

A Cooperative Hierarchical Multi-Agent System for EV Charging Scheduling in Presence of Multiple Charging Stations

Abstract: The increasing penetration of plug-in electric vehicles (EVs) to the electrical grid raises concerns over secure and economic operation of the system. A coordination mechanism between system operator and EV aggregators is necessary to ensure that the system is operated within the security limits, and to reduce the charging costs while satisfying EV users’ energy needs. In this work, we present a cooperative hierarchical multi-agent system and propose an EV charging scheduling strategy in order to minimize the demand and energy charges while meeting the EV users’ energy requirements and satisfying the system security constraints. Within the designed framework, the higher-level agents calculate a set of proposed control signals by solving the designated optimization problems, and send them to the lower-level agents to facilitate an optimal scheduling in line with the aforementioned objectives. Through this hierarchically distributed approach, it is possible to effectively coordinate multiple EV charging stations without the need of direct communication or any prior information related to EV arrivals. The computational complexity of the problem is reduced by distributing the work among agents, and the privacy of sensitive data, such as system topology, load profiles, and EV parameters, is preserved. Moreover, unlike the traditional distributed solution methods that converge iteratively, the proposed approach calculates the optimal charging schedule after a single round of communication. The efficacy of the proposed methodology is demonstrated by a series of case studies on 33-bus and 118-bus distribution test feeders.

A Methodology for Modeling Interoperability of Smart Sensors in Smart Grids

Abstract: Smart sensors in smart grids provide real-time data and status of bidirectional flows of energy for monitoring, protection, and control of grid operations to improve reliability and resilience. Smart sensor data interoperability is a major challenge for smart grids. This paper proposes a methodology for modeling interoperability of smart sensors in terms of interactions using labeled transition systems and finite state processes in order to quantitatively and automatically measure and assess the interoperability, identify and resolve interoperability issues, and improve interoperability. A generic interoperability model of synchronous message passing from a sender to a receiver is built based on the proposed methodology. A case study is provided to apply this methodology for modeling interoperability between the Institute of Electrical and Electronics Engineers C37.118 phasor measurement unit-based smart sensors and phasor data concentrators. The interoperability model can be used for the quantitative and automated measurement and assessment of the interoperability of phasor measurement unit-based smart sensors and phasor data concentrators to address interoperability issues. This methodology can also be applied to modeling interoperability of smart sensors based on other standard communication protocols in order to achieve and assure sensor data interoperability in smart grids.

Learning to Operate Distribution Networks With Safe Deep Reinforcement Learning

Abstract: In this paper, we propose a safe deep reinforcement learning (SDRL) based method to solve the problem of optimal operation of distribution networks (OODN). We formulate OODN as a constrained Markov decision process (CMDP). The objective is to achieve adaptive voltage regulation and energy cost minimization considering the uncertainty of renewable resources (RSs), nodal loads and energy prices. The control actions include the number of in-operation units of the switchable capacitor banks (SCBs), the tap position of the on-load tap-changers (OLTCs) and voltage regulators (VRs), the active and reactive power of distributed generators (DGs), and the charging and discharging power of battery storage systems (BSSs). To optimize the discrete and continuous actions simultaneously, a stochastic policy built upon a joint distribution of mixed random variables is designed and learned through a neural network approximator. To guarantee that safety constraints are satisfied, constrained policy optimization (CPO) is employed to train the neural network. The proposed approach enables the agent to learn a cost-effective operating strategy through exploring safe scheduling actions. Compared to traditional deep reinforcement learning (DRL) methods that allow agents to freely explore any behaviors during training, the proposed approach is more practical to be applied in a real system. Simulation results on a modified IEEE-34 node system and a modified IEEE-123 node system demonstrate the effectiveness of the proposed method.

Generalized Multivariable Grid-Forming Control Design for Power Converters

Abstract: The grid-forming converter is an important unit in the future power system with more inverter-interfaced generators. However, improving its performance is still a key challenge. This paper proposes a generalized architecture of the grid-forming converter from the view of multivariable feedback control. As a result, many of the existing popular control strategies, i.e., droop control, power synchronization control, virtual synchronous generator control, matching control, dispatchable virtual oscillator control, and their improved forms are unified into a multivariable feedback control transfer matrix working on several linear and nonlinear error signals. Meanwhile, unlike the traditional assumptions of decoupling between AC and DC control, active power and reactive power control, the proposed configuration simultaneously takes all of them into consideration, which therefore can provide better performance. As an example, a new multi-input-multi-output-based grid-forming (MIMO-GFM) control is proposed based on the generalized configuration. To cope with the multivariable feedback, an optimal and structured $H_{\infty}$ synthesis is used to design the control parameters. At last, simulation and experimental results show superior performance and robustness of the proposed configuration and control.

SAMNet: Toward Latency-Free Non-Intrusive Load Monitoring via Multi-Task Deep Learning

Abstract: Non-intrusive load monitoring (NILM), including state detection and energy disaggregation, aims to identify the on/off state and energy consumption from the aggregate load of a building. By monitoring the electrical behavior of consumers, smart grid applications such as demand response and recommendation services can also be realized for saving energy bills, environmental effectiveness, assisted living, and fault diagnosis. To achieve latency-free monitoring, this paper presents a Scale- and Attention-experts based Multi-task neural network (SAMNet) with a large enough context of the view to make full use of the correlation between the tasks of the NILM. A shared expert learner is designed to learn a good summary of the features. Self-attention mechanism is creatively adopted to realize the weighted fusion of different experts. To address the problem of manually setting different threshold values for different appliances to decide the on/off states, we designed the network automatically labeling the on/off states. Extensive experimental results with the public datasets demonstrate the effectiveness and superiority of our proposed model.

Credit-Based Peer-to-Peer Electricity Trading in Energy Blockchain Environment

Abstract: Peer-to-peer (P2P) electricity trading promotes the local consumption of renewable energy. However, it suffers from high transaction costs and mutual distrust among users. To address these issues, we propose a credit-based P2P electricity trading model in a blockchain environment. First, the P2P electricity trading process is introduced, which involves six stages: order generation, default query, order picking, trading execution, trading verification, and payment. Credit management is also introduced in this model to manage the default behavior of users. In particular, the default users are provided a waiting time in the default query stage or penalized in the payment stage. Finally, the model is simulated on the Hyperledger Fabric platform using Docker and Go. Experimental results show that the proposed model can facilitate cost reduction for users in the blockchain and realize credit management in P2P electricity trading, thereby enhancing trading stability and efficiency.