Search

Summary: The integration of multi‐energy within distribution networks has escalated the need for efficient operation and control of integrated energy systems (IES). Addressing the complexities of real‐time scheduling and low‐carbon optimization, we propose a novel artificial intelligence driven multi‐agent system (MAS) approach for modeling the interactions and operations within the multi‐agent integrated energy systems (MA‐IES) framework. In this framework, distinct components such as electric, gas, and heat networks are conceptualized as autonomous agents, each responsible for managing its domain while interacting with other agents to achieve system‐wide efficiency and economical goals. The agents communicate and coordinate through a distributed online optimization framework, utilizing the alternating direction multiplier method (ADMM) to ensure effective consensus despite the inherent nontransparency of information exchange. This MAS based approach allows for dynamic adaptation of strategies based on local data and global objectives, significantly enhancing the responsiveness and adaptability of MA‐IES. We further integrate an objective function reliant on a tiered carbon pricing mechanism to assess and minimize the environmental impact of operations. Enhanced by adaptive penalty coefficients within the ADMM, our MA‐IES framework demonstrates improved convergence rates and robustness in operational scenarios. Empirical validation through detailed case studies confirms the superior performance of our MAS‐based model, demonstrating its potential to realize an efficient and economical low‐carbon operation of MA‐IES. [ABSTRACT FROM AUTHOR]

For input-delayed mechanical system with input amplitude constraints, the consensus control problem of multi-agent system (MAS) based on directed graph topology is studied. For scenarios where only some agents can receive the desired signal, an observer is proposed to enable each agent to observe the desired signal, with exponential convergence of the observation error being proven. In the controller design, the system's input integral in fed back to the controller to eliminate the effects of input delay. By combining the input integral method with the hyperbolic tangent function, the input amplitude in limited while addressing the input delay issue. The FMINCON function in used to optimize the control parameters. Finally, the Lyapunov method in utilized to prove the achievement of control objectives and the asymptotic stability of the closed-loop system. Simulation results demonstrate the effectiveness of the proposed observer and controller n achieving system consensus control. [ABSTRACT FROM AUTHOR]

In this paper, three sorts of network indices for the weighted three-layered graph are studied through the methods of graph spectra theory combined with analysis methods. The concept of union of graphs are applied to design two sorts of weighted layered multi-fan composed graphs, and the accurate mathematical expressions of the network indices are obtained through the derivations of Laplacian spectra; furthermore, the asymptotic properties are also derived. We find that when the cardinalities of the vertices on a sector-edge-link tend to infinity, the indices of FONC and EMFPT are irrelevant with the number of copies of the fan-substructure based on the considered graph framework. [ABSTRACT FROM AUTHOR]

This paper presents a novel approach to analyzing trends in federated learning (FL) using automatic semantic keyword clustering. The authors collected a dataset of FL research papers from the Scopus database and extracted keywords to form a collection representing the FL research landscape. They employed natural language processing (NLP) techniques, specifically a pre-trained transformer model, to convert keywords into vector embeddings. Agglomerative clustering was then used to identify major thematic trends and sub-areas within FL. The study provides a granular view of the thematic landscape and captures the broader dynamics of research activity in FL. The key focus areas are divided into theoretical areas and practical applications of FL. The authors make their FL paper dataset and keyword clustering results publicly available. This data-driven approach moves beyond manual literature reviews and offers a comprehensive overview of the current evolution of FL. [ABSTRACT FROM AUTHOR]

Copyright of Systems Engineering & Electronics is the property of Journal of Systems Engineering & Electronics Editorial Department and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The impact of communication disturbances on microgrids (MGs) needs robust and scalable Information Communication Technology (ICT) infrastructure for efficient MG control. This work builds on advances in the Internet of Things (IoT) to provide a practical platform for testing the impact of various cyber-attacks on a distributed control scheme for a Multi-Agent System (MAS). This paper presents a Controller Hardware-in-the-Loop (CHIL) testbed to investigate the impact of various cyber-attacks and communication disruptions on MGs. A distributed consensus secondary control scheme for a MAS within an MG cyber-physical system (CPS) is proposed. The proposed cyber-physical testbed integrates a real-time islanded AC microgrid on RT-Lab, secondary controllers implemented on single-board computers, and an attacker agent on another single-board computer. Communication occurs via a UDP/IP network between OPAL-RT and controller agents, as well as between the agents. Through meticulous experimentation, the efficacy of the proposed control strategy using the developed platform is validated. Various attacks were modeled and launched including deception attacks on sensors, actuators, and their combinations, as well as disruption attacks. The ramifications of both deception and disruption cyber-attacks on system performance are analyzed. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Donghua University (English Edition) is the property of Journal of Donghua University Editorial Board and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In this article, a robust index named first-order network coherence (FONC) for the multi-agent systems (MASs) with layered lattice-like structure is studied via the angle of the graph spectra theory. The union operation of graphs is utilized to construct two pairs of non-isomorphic layered lattice-like structures, and the expression of the index is acquired by the approach of Laplacian spectra, then the corresponding asymptotic results are obtained. It is found that when the cardinality of the node sets of coronary substructures with better connectedness tends to infinity, the FONC of the whole network will have the same asymptotic behavior with the central lattice-like structure in the considered classic graph frameworks. The indices of the networks were simulated to illustrate the the asymptotic results, as described in the last section. [ABSTRACT FROM AUTHOR]

This article presents a comprehensive approach to integrate formation tracking control and optimal control for a fleet of multiple surface vehicles (SVs), accounting for both kinematic and dynamic models of each SV agent. The proposed control framework comprises two core components: a high‐level displacement‐based formation controller and a low‐level reinforcement learning (RL)‐based optimal control strategy for individual SV agents. The high‐level formation control law, employing a modified gradient method, is introduced to guide the SVs in achieving desired formations. Meanwhile, the low‐level control structure, featuring time‐varying references, incorporates the RL algorithm by transforming the time‐varying closed agent system into an equivalent autonomous system. The application of Lyapunov's direct approach, along with the existence of the Bellman function, guarantees the stability and optimality of the proposed design. Through extensive numerical simulations, encompassing various comparisons and scenarios, this study demonstrates the efficacy of the novel formation control strategy for multiple SV agent systems, showcasing its potential for real‐world applications. [ABSTRACT FROM AUTHOR]

In this paper, three sorts of network indices for the weighted three-layered graph are studied through the methods of graph spectra theory combined with analysis methods. The concept of union of graphs are applied to design two sorts of weighted layered multi-fan composed graphs, and the accurate mathematical expressions of the network indices are obtained through the derivations of Laplacian spectra; furthermore, the asymptotic properties are also derived. We find that when the cardinalities of the vertices on a sector-edge-link tend to infinity, the indices of FONC and EMFPT are irrelevant with the number of copies of the fan-substructure based on the considered graph framework.

This paper presents a novel approach to analyzing trends in federated learning (FL) using automatic semantic keyword clustering. The authors collected a dataset of FL research papers from the Scopus database and extracted keywords to form a collection representing the FL research landscape. They employed natural language processing (NLP) techniques, specifically a pre-trained transformer model, to convert keywords into vector embeddings. Agglomerative clustering was then used to identify major thematic trends and sub-areas within FL. The study provides a granular view of the thematic landscape and captures the broader dynamics of research activity in FL. The key focus areas are divided into theoretical areas and practical applications of FL. The authors make their FL paper dataset and keyword clustering results publicly available. This data-driven approach moves beyond manual literature reviews and offers a comprehensive overview of the current evolution of FL.

A major challenge in multi‐agent formation is the issue of delay and uncertainty. This article investigates a robust formation control problem for linear multi‐agent systems with input delay and model uncertainty. The strategy adopted by an agent in the system aims to predict the delayed state using output information from neighboring agents over a fixed communication network and generate the control input from the predictor output. The predictor is employed using the finite spectrum assignment (FSA) technique. The overall strategy leads to a unified framework representation, and the multi‐agent formation control problem is simplified to a closed‐loop stability problem for an agent. The overall problem comes down to deciding two feedback gains: predictor gain and controller gain. For a perfect plant model, any choice of stabilizing gains can achieve formation. But when agent models are uncertain, the gains need to be derived from the solutions of linear matrix inequalities (LMIs) containing network information and delay. LMIs are obtained from the bounded real lemma and the gains obtained from the solution guarantee to maximize the H∞$$ {H}_{\infty } $$ norm bound of allowable perturbation, modeled as additive uncertainty, for a known delay size. Further, since the FSA technique is sensitive to discretization, a digital implementation of the overall scheme is elaborated, which may help to debug stability issues in the implementation process. Finally, numerical examples are presented to demonstrate the effectiveness of the proposed ideas. [ABSTRACT FROM AUTHOR]

In this paper, a sustainable, intelligent energy management system for a microgrid based on a multi-agent system (MAS) is studied. The system is designed to address the challenges posed by the intermittence of renewable energy sources. Also, the system optimizes the use of available AC–DC renewable energy sources by utilizing load flexibility and the complementarity of renewable sources. To evaluate the effectiveness of this proposed multi-agent framework, a co-simulation using MATLAB and JADE platforms is conducted for a microgrid connected to the main grid. The results show that the proposed energy sharing system achieved over 82.34% of energy savings. This innovative solution has the potential to reduce the need for energy storage and improve energy efficiency while also reducing CO2 emissions. It offers a promising sustainable development solution for managing and controlling MG. The proposed system contributes to the development of sustainable energy systems based on artificial intelligence to meet the global goals. [ABSTRACT FROM AUTHOR]

Food waste is an alarming issue pertaining to the rising global hunger, huge environmental footprint, and high monetary value. In developing and developed nations, it occurs primarily due to inefficiencies upstream and downstream of the supply chain respectively. A common factor in both developed and developing nations is product flow within the supply chain from farms to retailers. This study aims to identify the root causes of waste generated across the product flow of the beef supply chain from farm to retailer. A workshop involving twenty practitioners of the beef industry was conducted and the collected information was transcribed and coded to generate a current reality tree, which assisted in identifying root causes of waste in the entire beef supply chain. A multi-agent architecture framework spanning the entire beef supply chain from farm to retailer is proposed, which is composed of autonomous agents capable of bringing all segments of the beef industry on a single platform and collaboratively assist them in mitigating root causes of waste. The proposed framework will aid the practitioners in the beef industry to reduce waste, improve their operational efficiency thereby raising food security, economic development whilst curbing their carbon footprint. [ABSTRACT FROM AUTHOR]

The high penetration of renewable energy sources in modern distribution networks poses challenges for grid voltage regulation. In this study, a multi-agent distributed voltage control strategy based on the proximal Jacobian alternating direction method of multipliers (PJ-ADMM) is proposed for distribution power systems with a high penetration of photovoltaic (PV) resources coordinated with battery energy storage systems (BESS). In this context, the active and reactive power outputs of the PV are locally optimized through smart inverters to improve the grid voltage with minimum power loss. Uncertainties associated with solar energy generation and load demands are considered in the defined scenarios. This study consists of two phases. In the first phase, the voltage control problem is formulated as an optimization problem to regulate the voltages within an acceptable limit with fast convergence. In the second phase, a coordinated voltage control strategy for smart PV inverters and BESS is proposed to allocate the power capacity of the battery energy storage systems and the active power loss reduction. Finally, the proposed method is tested on modified IEEE 13-bus, 33-bus and 141-bus distribution systems using MATLAB/Simulink and MATPOWER. A comparison of the results of the voltage profiles with and without the control algorithm demonstrated the efficacy, robustness, and scalability of the distributed scheme for voltage improvement and optimal utilization of PV power under different scenarios.

In multi-agent systems (MAS), it is a basic mission to coordinate and control each agent to complete tasks on mutually exclusive resources due to the high cost of some hardware and software in reality. In supervisory control theory (SCT) of discrete-event systems (DES), a supremal supervisor is synthesized with respect to the system and all the specifications to make the system act as desired. However, the size of the plant model increases exponentially with the number of agents in MAS. To alleviate the state explosion problem when applying SCT to MAS, a synchronous product with constraints using bread-first search strategy is proposed to synthesize a supervisor for the system with respect to the mutex resources. And a framework for computing the supremal supervisor for the entire MAS with respect to all the specifications is presented. Experiments are studied to estimate the effectiveness of the proposed approach. Contributions of the research are two-fold. Firstly, the state explosion problem associated with supervisory control of MAS is alleviated in a certain extent. Secondly, compared with decentralized or distributed control, it is not necessary to design a coordinator additionally.

There may exist system operational risks and the risks of losing the revenue because of the generating uncertainty of the stochastic units during the operational process of the virtual power plant (VPP),which contains multiple large scale stochastic units. To solve these problems existing in VPP,a two-stage bi-level decentralized planning (BLDP) model incorporating the conditional value-at-risk (CVaR) based on the characteristics of hierarchical control in the multi-agent system is built. Multi objective programming (MOP) model is selected and compared with the BLDP model constructed in this paper to verify the effectiveness of the BLDP model. It indicates that when VPP participates in the market independently and trades with other VPPs under the BLDP scenario,the system will achieve higher profits and be affected less by the risk levels. Meanwhile,the system can get higher yield rate under BLDP scenario than that under MOP scenario under different trading mechanisms,when VPP participates the power market independently without mutual dealing,results obtained under the BLDP scenario are 103% higher than that under the MOP scenario,while when it takes part in both the market and mutual dealing,the results under the BLDP scenario are 147% higher than that under the MOP scenario. The results have demonstrated the availability of the proposed model under different risk levels.

In this paper, the problem of distributed adaptive consensus tracking control for second-order nonlinear heterogeneous multi-agent systems (MASs) with input quantization is considered. A distributed output feedback control scheme based on a K-filter is developed to suppress the influences of unknown disturbances and input quantization. In contrast to existing approaches, an additional design parameter is introduced into the controller design to ensure that the subsystem tracking error converges to an arbitrarily small residual set. Through Lyapunov stability analysis, it can be proved that the proposed control scheme can achieve distributed consensus tracking control of second-order nonlinear heterogeneous MASs. In addition, all signals in the closed-loop system are shown to be globally uniformly bounded. Finally, a practical example demonstrates the effectiveness of the proposed control method. [ABSTRACT FROM AUTHOR]

Flocking guidance addresses a challenging problem considering the navigation and control of a group of passive agents. To solve this problem, shepherding offers a bio-inspired technique for navigating such a group of agents using external steering agents with appropriately designed movement law. Although most shepherding research is mainly based on the availability of centralized instructions, these assumptions are not realistic enough to solve some emerging application problems. Therefore, this paper presents a decentralized shepherding method where each steering agent makes movements based on its own observation without any inter-agent communication. Our numerical simulations confirm the effectiveness of the proposed method by showing its high success rate and low costs in various placement patterns. These advantages particularly improve with the increase in the number of steering agents. We also confirm the robustness and resilience properties of the proposed method via numerical simulations. [ABSTRACT FROM AUTHOR]

Resilience in autonomous agent systems is about having the capacity to anticipate, respond to, adapt to, and recover from adverse and dynamic conditions in complex environments. It is associated with the intelligence possessed by the agents to preserve the functionality or to minimize the impact on functionality through a transformation, reconfiguration, or expansion performed across the system. Enhancing the resilience of systems could pave way toward higher autonomy allowing them to tackle intricate dynamic problems. The state-of-the-art systems have mostly focussed on improving the redundancy of the system, adopting decentralized control architectures, and utilizing distributed sensing capabilities. While machine learning approaches for efficient distribution and allocation of skills and tasks have enhanced the potential of these systems, they are still limited when presented with dynamic environments. To move beyond the current limitations, this paper advocates incorporating counterfactual learning models for agents to enable them with the ability to predict possible future conditions and adjust their behavior. Counterfactual learning is a topic that has recently been gaining attention as a model-agnostic and post-hoc technique to improve explainability in machine learning models. Using counterfactual causality can also help gain insights into unforeseen circumstances and make inferences about the probability of desired outcomes. We propose that this can be used in agent systems as a means to guide and prepare them to cope with unanticipated environmental conditions. This supplementary support for adaptation can enable the design of more intelligent and complex autonomous agent systems to address the multifaceted characteristics of real-world problem domains.

In this article, a robust index named first-order network coherence (FONC) for the multi-agent systems (MASs) with layered lattice-like structure is studied via the angle of the graph spectra theory. The union operation of graphs is utilized to construct two pairs of non-isomorphic layered lattice-like structures, and the expression of the index is acquired by the approach of Laplacian spectra, then the corresponding asymptotic results are obtained. It is found that when the cardinality of the node sets of coronary substructures with better connectedness tends to infinity, the FONC of the whole network will have the same asymptotic behavior with the central lattice-like structure in the considered classic graph frameworks. The indices of the networks were simulated to illustrate the the asymptotic results, as described in the last section.

Effective fault-diagnosis strategies have been the focus of research on multi-agent systems (MASs). In this paper, the belief rule base (BRB)-based distributed fault-diagnosis problem for MASs is investigated, and a topology-switching strategy is developed to increase the reliability of fault-diagnosis model. Firstly, a BRB-based distributed fault-diagnosis model is constructed for the MAS with multiple faults, then expert knowledge is used to judge whether the agent is faulty. Then, considering that the system may be influenced by the fault or some other factors and thus leading to a decrease in the accuracy of the fault-diagnosis results, a topology-switching strategy based on the average distance of the output diagnosis accuracy is proposed to update the topology of the agent so that the fault-diagnosis results can be more reliable. Note that the topology-switching threshold is designed based on the average distance between the accuracy of the fault diagnosis of each agent. The method proposed in this paper can solve the problem when the fault-diagnosis accuracy of the model is affected by some common factors and thus decreases, and can improve the reliability of the fault-diagnosis model very well. Finally, the effectiveness of the BRB-based distributed fault-diagnosis model and the proposed topology-switching strategy to improve the fault-diagnosis accuracy is verified by simulation examples. [ABSTRACT FROM AUTHOR]

One of the main objectives of the Energy Management System (EMS) is to achieve a very high level of flexibility, stability, and the system must be able to adapt to most changes in the distribution network. This paper proposes a multi-agent system-based microgrid energy management to balance the energy supply and demand by feasibly integrating the energy storage system and demand response. An overall power management strategy is necessary to manage power flows among all interconnected elements of the microgrid. The principal contribution of this paper is an energy management system based on intelligent agents; each agent uses the microgrid data to manage the power flow in the microgrid. The interaction between agents proposed is simulated using JADE (Java Agent Development Environment) and the microgrid model is simulated using MATLAB/Simulink, both platforms are connected through the MACSimJX middleware for a dynamic simulation. [ABSTRACT FROM AUTHOR]

The demand for customized products is increasing, which brings great challenges to the traditional manufacturing system. The traditional manufacturing system lacks the analysis and feedback of the data collected in manufacturing, which hinders the rational utilization of the resources. In order to reasonably allocate resources according to the production requirements and effectively avoid the influence of external interference on the manufacturing system, a framework for the intelligent digital twin shopfloor is proposed in this paper. The framework consists of three primary components, namely physical shopfloor, virtual shopfloor, and digital twin service system. A model of digital twin shopfloor, a self-organizing model which achieves automation function in physical shopfloor, and an adaptive model which provides additional external disturbance resistance in virtual shopfloor are developed to improve the above-mentioned components. The mechanism of a contract network protocol and some cooperation strategies used in multi-agents are applied to achieve the capabilities above, which improve the reconfiguration and response ability of the shopfloor. A design case is studied to illustrate that the intelligent digital twin shopfloor can configure resources and deal with disturbances effectively. [ABSTRACT FROM AUTHOR]

The development of electric vehicle (EV) industry is facing a series of issues, among which the efficient charging of multiple EVs needs solving desperately. This paper investigates the coordinated charging of multiple EVs with the aim of reducing the charging cost, ensuring a high battery state of charge (SoC), and avoiding the transformer overload. To this end, we first formulate the EV coordinated charging problem with the above multiple objectives as a Markov Decision Process (MDP) and then propose a multi-agent deep reinforcement learning (DRL)-based algorithm. In the proposed algorithm, a novel interaction model, i.e., communication neural network (CommNet) model, is adopted to realize the distributed computation of global information (namely the electricity price, the transformer load, and the total charging cost of multiple EVs). Moreover, different from the most existing works which make specific constraints on the size, the location, or the topology of the distribution network, what we need in the proposed method is only the transformer load. Besides, due to the use of long and short-term memory (LSTM) for price prediction, the proposed algorithm can flexibly deal with various uncertain price mechanisms. Finally, simulations are presented to verify the effectiveness and practicability of the proposed algorithm in a residential charging area. [ABSTRACT FROM AUTHOR]

This paper presents an online adaptive learning solution to the optimal synchronization control problem of heterogeneous multi‐agent systems via a novel distributed policy iteration approach. For the leader‐follower multi‐agents, the dynamics of all the followers are heterogeneous with leader disturbance. To make the output of each follower synchronize with the leader's output, we propose a synchronization control protocol where the stability conditions for selecting the feedback gains are given. Next, with a minimization of the output tracking errors, we optimize the feedback gains for the synchronization control protocol, and the unique solutions for those feedback gains are learned via a novel distributed policy iteration approach. We prove that the proposed online adaptive learning solution will converge to the optimal control solution under some mild conditions. Finally, an illustrative numerical example is provided to show the effectiveness of the approach. [ABSTRACT FROM AUTHOR]