Development of a MultiAgent System for Identification and Control

As the complexity and scale of modern buildings increase, there is a growing need for efficient, scalable, and flexible control systems to manage the myriad of sensors, actuators, and meters involved in building automation. Traditional building management systems (BMS) struggle with scalability, especially when dealing with multi-agent control setups that require extensive manual configuration. This limitation poses significant challenges in deploying advanced control strategies across different building environments, leading to inefficiencies and underutilization of building automation technologies.Previous research has explored various aspects of multi-agent systems (MAS) in building automation, including agent-based modeling, decentralized control strategies, and the application of machine learning to optimize building operations. However, much of the existing work has focused on small-scale implementations or specific applications, such as energy management or HVAC control. The problem of automating the setup of multi-agent control systems to adapt to different building environments with minimal manual intervention remains largely unaddressed.The aim of this work is to develop a scalable framework for the automated setup of multi-agent control systems in buildings. This framework will enable the seamless integration of sensors, actuators, and meters into a unified control system, significantly reducing the need for manual configuration. The framework will leverage large language models (LLMs) for semantic tagging and context understanding, thereby facilitating the automated configuration of action and input spaces for individual agents within the control system.The proposed framework will consist of three key components: (1) an LLM-based system for semantic tagging and context interpretation, (2) an automated setup process for defining the action and input spaces of the agents, (3) and an initial offline training of the agents. The LLM will be trained on a diverse dataset of building management setups to ensure it can accurately tag and interpret various building elements. The system will be tested in real-world office buildings to evaluate its scalability, accuracy, and effectiveness.Preliminary simulations indicate that the automated setup framework can significantly reduce the time and effort required to deploy multi-agent control systems in buildings. We expect that this framework will enable more scalable and flexible building management systems, ultimately leading to improved energy efficiency, occupant comfort, and overall operational performance. The significance of this work lies in its potential to transform the deployment of building automation systems, making advanced control strategies more accessible and cost-effective across various building types and scales.

In this dissertation, new algorithms for formation control of multi agent systems (MAS) based on continuum mechanics principles will be suggested. For this purpose, agents of the MAS are considered as particles in a continuum, evolving in R^n, whose desired configuration is required to satisfy an admissible deformation function. Considered is a specific class of mappings that are called homogenous where the Jacobian of the mapping is only a function of time and is not spatially varying. The primary objectives of this dissertation are to develop the necessary theory and its validation on a mobile-agent based swarm test bed that includes two primary tasks: 1) homogenous transformation of MAS and 2) deployment of a random distribution of agents on a desired configuration. Developed will be a framework based on homogenous transformations for the evolution of an MAS in an n-dimensional space (n=1,2, and 3), under1) no inter-agent communication (predefined motion plan), 2) local inter-agent communication, and 3) intelligent perception by agents. In this dissertation, different communication protocols for MAS evolution that are based on certain special features of a homogenous transformation will be developed. It is also aimed to deal with the robustness of tracking of a desired motion by an MAS evolving in R^n. Furthermore, the effect of communication delays in an MAS evolving under consensus algorithms or homogenous maps is investigated. In this regard, the maximum allowable communication delay for MAS evolution is formulated on the basis of eigen-analysis.

This paper proposes a new multiagent control system (MACS) for service restoration in distribution systems with integrated distributed generation (DG) units. First, the MACS detects and locates faults, then decides the optimal reconfiguration of the network for restoring de-energized loads, and finally regulates the nodes voltages. Unintentional islanding operation of DG units is avoided and different postfault response modes of DG unit are addressed in the MACS design. The MACS has a hybrid centralizeddecentralized structure where agents are arranged in two layers with different responsibilities and communication capabilities. Agents at load buses in the first layer can only communicate directly with its next neighbor load agents on the same feeder and to its feeder agent, whereas the higher level agents in the second layer can communicate directly with each other. This MACS structure reduces the possibilities of control system failures for a moderate communication network infrastructure. Full dynamic simulation model for evaluating the MACS is implemented.

Chapter 6 - Protecting Critical Infrastructure: Process Control and SCADA

This paper proposes a multiagent control system (MACS) for a fossil-fuel power unit (FFPU). The MACS consists of fossil-fuel power unit agent (FFPUA), reference agent (RA), feedforward control agents (FFCAs), feedback control agents (FBCAs) and coordinator agent (COA). The reference agent decomposes the task of the MACS by multiobjective optimization. The FFCAs make decision using the neuro-fuzzy systems and the FBCAs make decision using the genetic algorithm-based fuzzy systems. The coordinator agent coordinates the agents in the MACS according to different operating conditions.

The electric power industry is currently in transition to a new technological platform, which is based on the smart grid concept. Important components of this grid include distributed generation plants as well as those based on renewable energy sources. The centralized system of control of numerous distributed generation plants becomes inefficient due to the need for transferring large amounts of data to the control center. The control efficiency can be enhanced by use of peripheral systems based on the agent approach. The paper suggests the design, description and Joiner-network of an event model of the multi-agent control system of distributed generation plants based on turbine-generators. The main functions of the agents of the interactive network are presented. A system of differential equations defining processes within the power supply system including distributed generation plants is considered. The model of the proposed multi-agent control system is developed in the AnyLogic simulation environment. Modeling results showed positive effects gained with the multi-agent control system. The adaptability of the system to changing conditions and applicability of the multi-agent technologies when establishing active-adaptive networks of intelligent power supply systems are demonstrated.

In this paper, a new multi-agent control system (MACS) is proposed for smart distribution networks. It can: 1) eliminate the feeders' congestion, 2) globally correct voltage violations, and 3) coordinate the operation of reactive power control devices. In performing its functions, the MACS must prevent overstress on the substation voltage regulator tap changer under all operating conditions, and avoid selective active power curtailment from DG units. This can help to improve reliability, economic revenue, and power quality. A fuzzy rule-based system is employed for decision-making support in MACS. The proposed MACS can achieve the self-healing requirement of smart distribution network. It eliminates feeders' congestion by properly reconfiguring the distribution network. Meanwhile, it amends most of voltage violation problems by coordinated operation of voltage control devices in the distribution system. The MACS overcomes the severe voltage violation problems via distribution network reconfiguration. It is applied to the IEEE 33-bus test feeder. A full dynamic model is constructed using a parallel MATLAB-JADE simulation arrangement. Comparative performance evaluations of the MACS and other schemes are presented under various conditions.

This paper presents a fuzzy multi-agent system for control of the traffic signals of an urban network, where the aim is to reduce the total average delay time of the vehicles. The large-scale traffic system is first divided into sub-areas and an agent, using a fuzzy controller, is considered for each sub-area. In order to develop the fuzzy rule bases for the agents, an extensive set of data is collected and the corresponding origin-destination (OD) matrices are calculated. The OD matrices are then clustered using a new clustering algorithm proposed in this paper. Finally, the resulting cluster centers, which are matrices of the same dimension as the OD matrices, are mapped to a 2-dimensional space and the corresponding triangular fuzzy sets are extracted. An optimal cycle plan is found for each fuzzy set and this way the fuzzy rule bases are constructed (the triangular fuzzy sets form the IF-parts and the optimal cycle plans form the THEN-parts). In a case study the proposed multi-agent control system is applied to a microscopic urban traffic network and the results are compared with a controller that applies optimal signal plans calculated by PASSER V. The results show that the proposed fuzzy multi-agent system outperforms the non-fuzzy control system, where the average delay time of the traveling vehicles is decreased by 19% using the multi-agent control system.

Increasing operating costs and decreasing net revenues have greatly underlined the importance of optimizing production operations. The development of automation systems in the Petroleum Industry has become important in uniting isolated production facilities to form integrated systems. This paper briefly traces the history and development of remote and computer control systems for gas and oil production. It also describes the development of the remote control and monitoring system installation at Union Oil onshore and offshore production facilities in East Kalimantan. Emphasis is placed more on the description of the remote process system monitoring and control in the Attaka field. The wireless communication and process control system are being used to transmit alarm and flow metering, emergency shutdown, and paging functions between five satellite wellhead platforms and production platform central monitoring system at the Attaka field. Each producing facility is designed to be compatible with remote supervisory control. To obtain this compatibility and a high degree of safety, conditions in each portion of the process system are monitored. Paralleling the development of automatic control techniques were computer applications. A simple computer system are being used to control and monitor three gas dehydration towers which comprise a portion of the Union LEX Plant operation at Santan onshore facility. Expectations and development of future systems and evaluation of past experience are discussed in the conclusion.

Process Control Algorithms for Microprocessors

Continuous rolling of plate and strip is a typical complex, multi-variable and multi-objective system, which has shown the characteristics such as complication, large-scale and close coupling increasingly. Therefore, it has become more and more difficult to design complex industrial control systems. This paper describes a multi-agent control system for a strip rolling process. In order to realize the distributed and intelligent control of the system, after analyzing the characteristics and structure of this control system, the multi-agent theory was applied and then a control strategy was designed based on multi-agent interaction. Finally, the partition method, the coordination mechanism, and the model structure of the multi-Agent control system for the strip rolling process were analyzed. This paper concludes with a new control structure and system for strip rolling process, which helps to build up the generally integrated model, to eliminate the coupling among control systems, and to realize more intelligent control.

Survey of multi-agent systems for microgrid control

Nowadays developing methods for distributed computing is received much attention. One of the methods of distributed computing is using of multi-agent systems. The organization of distributed computing based on the conventional network computers can experience security threats performed by computational processes. Authors have developed the unified agent algorithm of control system of computing network nodes operation. Network PCs is used as computing nodes. The proposed multi-agent control system for the implementation of distributed computing allows in a short time to organize using of the processing power of computers any existing network to solve large-task by creating a distributed computing. Agents based on a computer network can: configure a distributed computing system; to distribute the computational load among computers operated agents; perform optimization distributed computing system according to the computing power of computers on the network. The number of computers connected to the network can be increased by connecting computers to the new computer system, which leads to an increase in overall processing power. Adding multi-agent system in the central agent increases the security of distributed computing. This organization of the distributed computing system reduces the problem solving time and increase fault tolerance (vitality) of computing processes in a changing computing environment (dynamic change of the number of computers on the network). Developed a multi-agent system detects cases of falsification of the results of a distributed system, which may lead to wrong decisions. In addition, the system checks and corrects wrong results.

This study proposes a new multi‐agent control system (MACS) for energy management in a microgrid (MG). The latter includes photovoltaic arrays and wind turbine generators as renewable energy resources. It also has microturbines, fuel cell units, and an energy storage system (ESS). The MACS determines the optimal active and reactive powers of the dispatchable energy resources. It also identifies the ESS charging/discharging times. The MACS supervises energy management among local distributed generators (DGs), main grid, and loads to maximise profit while satisfying all operational constraints. The MG collaborates in the electricity market. It purchases/sells active and reactive powers from/to the main grid. Furthermore, MACS dynamically corrects the possible voltage violations and eliminates line congestion using the coordination of reactive power control. Agents at dispatchable DGs and ESS buses tackle the optimal economical operation. Other agents regulate the nodes voltages and remove line congestions. A parallel computing‐based full dynamic simulation model of the MACS and the MG power system is implemented. Moreover, the performance comparisons between the proposed MACS and recent approaches are presented to evaluate the performance of MACS.

Process and Production Control Package