Broker Agent Research Articles

Demand Management of Distributed Energy Loads Based on Genetic Algorithm Optimization

Management of a very large number of distributed energy resources, energy loads, and generators, is a hot research topic. Such energy demand management techniques enable appliances to control and defer their electricity consumption when price soars and can be used to cope with the unpredictability of the energy market or provide response when supply is strained by demand. We consider a multi-agent system comprising multiple energy loads, each with a dedicated controller. This paper introduces our latest research in self-organization of coordinated behavior of multiple agents. Energy resource agents (RAs) coordinate with each other to achieve a balance between the overall consumption by the multi-agent collective and the stress on the community. In order to reduce the overall communication load while permitting efficient coordinated responses, information exchange is through indirect communications between RAs and a broker agent (BA). This gives a decentralized coordination approach that does not rely on intensive computation by a central processor. The algorithm presented here can coordinate different types of loads by controlling their set-points. The coordination strategy is optimized by a genetic algorithm (GA) and a fast coordination convergence has been achieved.

가상 전력 도매 시장의 최적 경매 가격 예측

Power TAC (Power Trading Agent Competition) is an agent-based simulation for competitions between electricity brokering agents on the smart grid. To win the competition, agents obtain electricity from the electricity wholesale market among the power plants. In this operation, a key to success is balancing the demand of the customer and the supply from the plants because any imbalance results in a significant penalty to the brokering agent. Given the bidding on the wholesale market requires the price and the quantity on the electricity, this paper proposes four different price estimation strategies: exponentially moving average, linear regression, fuzzy logic, and support vector regression. Our evaluations with the competition simulation show which strategy is better than which, and which strategy wins in the free-for-all situations. This result is a crucial component in designing an electricity brokering agent in both Power TAC and the real world.

Autonomous Agents in Future Energy Markets: The 2012 Power Trading Agent Competition

Sustainable energy systems of the future will need more than efficient, clean, and low-cost energy sources. They will also need efficient price signals that motivate sustainable energy consumption behaviors and a tight real-time alignment of energy demand with supply from renewable and traditional sources. The Power Trading Agent Competition (Power TAC) is a rich, competitive, open-source simulation platform for future retail power markets built on real-world data and state-of-the-art customer models. Its purpose is to help researchers understand the dynamics of customer and retailer decision-making as well as the robustness of proposed market designs. Power TAC invites researchers to develop autonomous electricity broker agents and to pit them against best-in-class strategies in global competitions, the first of which will be held at AAAI 2013. Power TAC competitions provide compelling, actionable information for policy makers and industry leaders. We describe the competition scenario, demonstrate the realism of the Power TAC platform, and analyze key characteristics of successful brokers in one of our 2012 pilot competitions between seven research groups from five different countries.

Power TAC: A competitive economic simulation of the smart grid

Sustainable energy systems of the future will need more than efficient, clean, low-cost, renewable energy sources; they will also need efficient price signals that motivate sustainable energy consumption as well as a better real-time alignment of energy demand and supply. The Power Trading Agent Competition (Power TAC) is a rich competitive simulation of future retail power markets. This simulation will help us to understand the dynamics of customer and retailer decision-making and the robustness of market designs, by stimulating researchers to develop broker agents and benchmark them against each other. This will provide compelling, actionable information for policymakers and industry leaders. We describe the competition scenario in detail, and we demonstrate behaviors that arise from the interaction of customer and broker models.

A reinforcement learning approach to autonomous decision-making in smart electricity markets

The vision of a Smart Electric Grid relies critically on substantial advances in intelligent decentralized control mechanisms. We propose a novel class of autonomous broker agents for retail electricity trading that can operate in a wide range of Smart Electricity Markets, and that are capable of deriving long-term, profit-maximizing policies. Our brokers use Reinforcement Learning with function approximation, they can accommodate arbitrary economic signals from their environments, and they learn efficiently over the large state spaces resulting from these signals. We show how feature selection and regularization can be leveraged to automatically optimize brokers for particular market conditions, and demonstrate the performance of our design in extensive experiments using real-world energy market data.

Complex and Concurrent Negotiations for Multiple Interrelated e-Markets

To date, most of the existing bargaining models are designed for supporting negotiation in only one market involving only two types of participants (buyers and sellers). This work devises a complex negotiation mechanism that supports negotiation activities among three types of participants in multiple interrelated markets. The complex negotiation mechanism consists of: 1) a bargaining-position-estimation (BPE) strategy for the multilateral negotiations between consumer and broker agents in a service market and 2) a regression-based coordination (RBC) strategy for concurrent negotiations between broker and provider agents in multiple resource markets. The negotiation outcomes between broker and provider agents in a resource market can potentially influence the negotiation outcomes between broker and consumer agents in a service market. Empirical results show that agents adopting the BPE strategy can better respond to different market conditions than agents adopting the time-dependent strategy because they do not make excessive (respectively, inadequate) amounts of concessions in favorable (respectively, unfavorable) markets. In the concurrent negotiations in multiple resource markets, empirical results show that broker agents adopting the RBC strategy achieved significantly higher utilities, higher success rates, and faster negotiation speed than broker agents adopting the utility-oriented and patient coordination strategies.

A multiagent brokering protocol for supporting Grid resource discovery

Resource discovery provides a mean to identify the set of resources that are capable of satisfying the job requirements. Scalability and adaptability are two key challenges of resource discovery in a Grid environment. To deal with both issues, the agent concept has been adopted to cope with unexpected events such as the failure of provider agents. Multiple broker agents are used to deal with huge amounts of data from multiple sources. In our multiagent protocol, three types of agents (user agent, provider agent, and broker agent) are used. Each broker agent connects user agents to provider agents using the connection algorithm which mainly consists of 4 stages: selection, evaluation, filtering, and recommendation. In the recommendation stage, two kinds of recommendation approaches (circular approach and multicast approach) are used for making recommendations to the user agents that failed to be matched to provider agents. Empirical results show that our approach using multiagent brokering protocol with the adaptable feature of coordinating load balance (the results of balanced case) has significantly good performance in terms of scalability and adaptability.

Agent-Based Cloud Computing

Agent-based cloud computing is concerned with the design and development of software agents for bolstering cloud service discovery, service negotiation, and service composition. The significance of this work is introducing an agent-based paradigm for constructing software tools and testbeds for cloud resource management. The novel contributions of this work include: 1) developing Cloudle: an agent-based search engine for cloud service discovery, 2) showing that agent-based negotiation mechanisms can be effectively adopted for bolstering cloud service negotiation and cloud commerce, and 3) showing that agent-based cooperative problem-solving techniques can be effectively adopted for automating cloud service composition. Cloudle consists of 1) a service discovery agent that consults a cloud ontology for determining the similarities between providers' service specifications and consumers' service requirements, and 2) multiple cloud crawlers for building its database of services. Cloudle supports three types of reasoning: similarity reasoning, compatibility reasoning, and numerical reasoning. To support cloud commerce, this work devised a complex cloud negotiation mechanism that supports parallel negotiation activities in interrelated markets: a cloud service market between consumer agents and broker agents, and multiple cloud resource markets between broker agents and provider agents. Empirical results show that using the complex cloud negotiation mechanism, agents achieved high utilities and high success rates in negotiating for cloud resources. To automate cloud service composition, agents in this work adopt a focused selection contract net protocol (FSCNP) for dynamically selecting cloud services and use service capability tables (SCTs) to record the list of cloud agents and their services. Empirical results show that using FSCNP and SCTs, agents can successfully compose cloud services by autonomously selecting services.

Optimal pricing solution for VoIP service in 802.16/802.11 combined networks

Future wireless networks are envisioned to be all IP and capable of running real time applications like VoIP. IEEE 802.16 and IEEE 802.11 based networks popularly known as Wimax and Wifi are already being widely used for internet and other services. The combined Wimax/Wifi scenario can be ideal public-private partnership solution for distributing VoIP-based telephony solutions in villages using Wimax-based private backbone operator and Wifi community network for end distribution. Internet bandwidth is still a scarce resource and should be used optimally from both the user as well as backbone service provider’s point of view. Game theory can be effectively used for optimising service for both the parties in combined Wimax/Wifi scenarios. The results indicate that the utilities for both user and backbone service provider can be improved with cooperation between both parties. Third party broker agent realises cooperation between the two entities for an optimal solution. Using this approach, it is possible to obtain an optimal VoIP solution, which is beneficial for both user and backbone service provider.

Learned Behaviors of Multiple Autonomous Agents in Smart Grid Markets

One proposed approach to managing a large complex Smart Grid is through Broker Agents who buy electrical power from distributed producers, and also sell power to consumers, via a Tariff Market--a new market mechanism where Broker Agents publish concurrent bid and ask prices. A key challenge is the specification of the market strategy that the Broker Agents should use in order to earn profits while maintaining the market's balance of supply and demand. Interestingly, previous work has shown that a Broker Agent can learn its strategy, using Markov Decision Processes (MDPs) and Q-learning, and outperform other Broker Agents that use predetermined or randomized strategies. In this work, we investigate the more representative scenario in which multiple Broker Agents, instead of a single one, are independently learning their strategies. Using a simulation environment based on real data, we find that Broker Agents who employ periodic increases in exploration achieve higher rewards. We also find that varying levels of market dominance in customer allocation models result in remarkably distinct outcomes in market prices and aggregate Broker Agent rewards. The latter set of results can be explained by established economic principles regarding the emergence of monopolies in market-based competition, further validating our approach.

Schools, Governance and Community: A Next Practice Intervention

The article analyses an innovative scheme for the re-organization of school governance that sought to enhance the voice of community members and contribute towards the improvement of educational achievement. The scheme was located in the north of England, a region characterized by high levels of social disadvantage and low participation rates in higher education. The research adopted a case study methodology informed by observation, minutes of governors’ meetings, training workshops, conferences, face-to-face interviews, online surveys and local authority (LA) documentation. The fieldwork was conducted in 2007—2008. The findings suggest that although aspects of community engagement are present they are inhibited by a number of factors operating at the strategic level in LA policy. This results in an aspirational model of school governance that reflects traditional practices in which the LA intercedes as the indispensable broker and external agent for innovation and change. In order to move away from the restrictions of this approach the LA needs to expand its strategies to involve members of the community in governance and develop mechanisms to devolve greater decision-making power to the community.

Commentary: Cultural Adaptation, Collaboration, and Exchange

This commentary reviews three articles linked together by two themes (a) the use of cultural adaptation of evidence-based practices to reduce disparities in health and services delivery and (b) the importance of collaboration involving intervention developers, practitioners, and consumers when delivering services. Both themes illustrate a process of cultural exchange, enabling researchers to develop interventions that are more meaningful and acceptable to consumers; providers to develop a stronger therapeutic alliance with consumers; and consumers to develop greater understanding and acceptance of treatment process. Such exchanges lead to improved consumer outcomes and greater satisfaction with services. By serving as a culture broker or change agents, social workers can play a leadership role in the translation of research to practice.

Formal Aspects of Grid Brokering

Coordination in distributed environments, like Grids, involves selecting the most appropriate services, resources or compositions to carry out the planned activities. Such functionalities appear at various levels of the infrastructure and in various means forming a blurry domain, where it is hard to see how the participating components are related and what their relevant properties are. In this paper we focus on a subset of these problems: resource brokering in Grid middleware. This paper aims at establishing a semantical model for brokering and related activities by defining brokering agents at three levels of the Grid middleware for resource, host and broker selection. The main contribution of this paper is the definition and decomposition of different brokering components in Grids by providing a formal model using Abstract State Machines.

A multi-agent framework for stock trading

TrAgent is a secure multi-agent environment which models a stock exchange such as the New York Stock Exchange and the procedure of trading securities on the trading floor. The model comprises the complete process from the initiation of a trading order until its execution and also deals with the security issues involved. The multi-agent paradigm is the framework for implementation of the system. The properties of intelligent software agents meet the characteristics of the actors on the trading floor and provide capabilities for efficient distributed computing. These characteristics include dynamic and flexible adoption to a changing scene, asynchronous and autonomous behaviour, and efficient use of resources in a distributed environment. The main focus in this paper is on one of the important agents in TrAgent, the stock broker agent. Intelligence is provided to the stock broker agent in order to make decisions and give reasons on the profitability of a firm. This intelligence is formed using fuzzy expert system. The paper further discusses the implementation issues concerning different components of the model and their security framework.

Integration of rules and policies for Semantic Web Services

Multimedia service providers on the web need their services to be well protected and easily accessible worldwide. This has initiated several lines of research to provide semantically rich policy-aware web services. However, the existence of various technologies to define and protect services has made the communication of services questionable. Currently, it is ambiguous how a broker agent looking for a special object would communicate to an online service provider, with different policy and service definition languages, to obtain the appropriate object. Here, we propose a solution to such problems by introducing an interchange framework to transform business rules and policies between different enterprises using REWERSE Rule Markup Language.

Resource discovery and allocation in wireless grids by using a multiagent system

This paper presents an Agent-based Discovery and Allocation of Resources (ADAR) model for wireless grids of virtual organizations. ADAR allocates cost effective resources for the jobs in order to maximize the resource utilization. It employs resource discovery mechanism in a hierarchical fashion: firstly within a local cluster, secondly within a virtual organization if a resource is not available in the local cluster, and finally among the virtual organizations if a resource is not available within the local virtual organization. ADAR uses five types of agents namely Job Processing Agents (JPAs), Job Mobile Agents (JMAs), Resource Monitoring Agents (RMAs), Actual Organization Resource Broker Agents (AORBAs), and Virtual Organization Resource Manager Agents (VORMAs). JPA is static and executes a job within the same machine if the resources are available; otherwise it creates a JMA that carries the job requirements with it and communicates with its associated AORBA to discover the required resources. The AORBA is a static agent and discovers the optimum cost resources within its cluster. AORBA interacts with the VORMAs in case the resources are not available within a cluster. VORMA is a static agent and can interact with other VORMAs. The RMA is a static agent and monitors the resource status of a machine. The discovery results are informed to JMA, which makes the source device to migrate its job to the discovered resource. ADAR is evaluated in a simulated environment by using different wireless grid scenarios for virtual organizations. The performance parameters evaluated are: job cost function, resource utilization, bandwidth utilization, agent overheads, and resource discovery time against different mobility factors and varying system loads.

An Agent-based Resource Allocation Model for computational grids

The paper presents an Agent-based Resource Allocation Model (ARAM) for grid computing. Three types of agents, namely Job Agents (JAs), Resource Brokering Agents (RBAs), and Resource Monitoring Agents (RMAs), are used. The JAs are mobile and they repr

Toward a Meta-Level Framework for Agent-Supported Interoperation of Defense Simulations

Interoperation and dynamic composability of disparate simulations is a long-standing challenge within the defense modeling and simulation community. While both issues are extensively studied, a unified and coherent strategy that facilitates achieving seamless and transparent congruity among conceptual and realization spaces of simulations is still an elusive goal. In this paper, an agent-supported meta-level interoperation architecture is proposed to address these challenges. The strategy involves the introduction of an agent framework independent of the simulation infrastructure for explicit separation of interoperation protocols from the simulation environment. To this end, agent-based mediation, brokering, matchmaking, and facilitation services are suggested as critical components for interoperation and composability. The proposed approach is based on the following premises: 1) deployment of brokering protocols improves transparency and balances the workload in interoperation, 2) partial matchmaking mechanisms provide more efficient and effective model and data qualification strategies than the conventional keyword-based matching mechanism, and 3) automated mediation facilities that are deployed independently from the simulation infrastructure improve transparency and runtime extensibility. Stand-alone prototypes of the matchmaker and broker agents are developed to facilitate testing these propositions. We conclude by discussing the extension of the applicability of the proposed strategy to DoDAF architectural views and associated products.

Intelligent agents meet the semantic Web in smart spaces

A new smart meeting room system called EasyMeeting explores the use of multi-agent systems, Semantic Web ontologies, reasoning, and declarative policies for security and privacy. Building on an earlier pervasive computing system, EasyMeeting provides relevant services and information to meeting participants based on their situational needs. The system also exploits the context-aware support provided by the Context Broker Architecture (Cobra). Cobra's intelligent broker agent maintains a shared context model for all computing entities in the space and enforces user-defined privacy policies.

A conceptual framework for agent‐based agile manufacturing cells

Abstract. Agile manufacturing techniques are perceived as the manufacturing systems of the future. Agile manufacturing cells are dynamic and reconfigurable and the modelling of the manufacturing cells and its interaction mechanism is critical to its successful use. This paper deals with the architecture and cooperation mechanism of web‐based agile manufacturing cells. Based on an analysis of structure and organization requirements of agile manufacturing cells and a comparison of three basic architectures of manufacturing systems, the quasi‐heterarchical architecture is used for the agile manufacturing cell. Functional layers are defined in this architecture to make the cell's control system reconfigurable and reusable. Agent technology is adopted for implementation of each layer's functions to establish an agent‐based model of agile manufacturing cells. Four types of agents including cooperation agent, job management agent, resource broker agent, and resource control agent are defined, and their functions discussed. Finally, a real time interaction mechanism of the agents is presented by considering the activities during the agents’ cooperation in an agile manufacturing cell.