MASCEM - an electricity market simulator providing coalition support for virtual power players

This paper presents MASCEM - a multi-agent based electricity market simulator. MASCEM uses game theory, machine learning techniques, scenario analysis and optimization techniques to model market agents and to provide them with decision-support. This paper mainly focus on the MASCEM ability to provide the means to model and simulate Virtual Power Players (VPP). VPPs are represented as a coalition of agents, with specific characteristics and goals. The paper details some of the most important aspects considered in VPP formation and in the aggregation of new producers and includes a case study based on real data.

Dgenco market balancing - cost repartition under light regulation markets

In this paper we present a new methodology, based in game theory, to obtain the market balancing between Distribution Generation Companies (DGENCO), in liberalized electricity markets. The new contribution of this methodology is the verification of the participation rate of each agent based in Nucl��olo Balancing and in Shapley Value. To validate the results we use the Zaragoza Distribution Network with 42 Bus and 5 DGENCO.

Ancillary service market simulation

Power systems operation in a liberalized environment requires that market players have access to adequate decision support tool, allowing them to consider all the business opportunities and take strategic decisions. Ancillary services represent a good negotiation opportunity that must be considered by market players. For this, decision support tools must include ancillary market simulation. This paper deals with ancillary services negotiation in electricity markets. The proposed concepts and methodologies are implemented in MASCEM, a multi-agent based electricity market simulator. A test case concerning the dispatch of ancillary services using two different methods (Linear Programming and Genetic Algorithm approaches) is included in the paper.

Ancillary services market clearing simulation: a comparison between deterministic and heuristic methods

Electricity market players operating in a liberalized environment requires access to an adequate decision support tool, allowing them to consider all the business opportunities and take strategic decisions. Ancillary services represent a good negotiation opportunity that must be considered by market players. For this, decision support tool must include ancillary market simulation. This paper proposes two different methods (Linear Programming and Genetic Algorithm approaches) for ancillary services dispatch. The methodologies are implemented in MASCEM, a multi-agent based electricity market simulator. A test case based on California Independent System Operator (CAISO) data concerning the dispatch of Regulation Down, Regulation Up, Spinning Reserve and Non-Spinning Reserve services is included in this paper.

Ancillary services dispatch using linear programming and genetic algorithm approaches

Electricity market players operating in a liberalized environment requires access to an adequate decision support tool, allowing them to consider all the business opportunities and take strategic decisions. Ancillary services represent a good negotiation opportunity that must be considered by market players. For this, decision support tools must include ancillary market simulation. This paper proposes two different methods (Linear Programming and Genetic Algorithm approaches) for ancillary services dispatch. The methodologies are implemented in MASCEM, a multi-agent based electricity market simulator. A test case concerning the dispatch of Regulation Down, Regulation Up, Spinning Reserve and Non-Spinning Reserve services is included in this paper.

Sistema multi-agente para gest��o de tr��fego em centros urbanos

Mestrado em Engenharia Inform��tica

Adaptive learning in agents behaviour: a framework for electricity markets simulation

Electricity markets are complex environments with very particular characteristics. A critical issue regarding these specific characteristics concerns the constant changes they are subject to. This is a result of the electricity markets��� restructuring, which was performed so that the competitiveness could be increased, but it also had exponential implications in the increase of the complexity and unpredictability in those markets scope. The constant growth in markets unpredictability resulted in an amplified need for market intervenient entities in foreseeing market behaviour. The need for understanding the market mechanisms and how the involved players��� interaction affects the outcomes of the markets, contributed to the growth of usage of simulation tools. Multi-agent based software is particularly well fitted to analyze dynamic and adaptive systems with complex interactions among its constituents, such as electricity markets. This dissertation presents ALBidS ��� Adaptive Learning strategic Bidding System, a multiagent system created to provide decision support to market negotiating players. This system is integrated with the MASCEM electricity market simulator, so that its advantage in supporting a market player can be tested using cases based on real markets��� data. ALBidS considers several different methodologies based on very distinct approaches, to provide alternative suggestions of which are the best actions for the supported player to perform. The approach chosen as the players��� actual action is selected by the employment of reinforcement learning algorithms, which for each different situation, simulation circumstances and context, decides which proposed action is the one with higher possibility of achieving the most success. Some of the considered approaches are supported by a mechanism that creates profiles of competitor players. These profiles are built accordingly to their observed past actions and reactions when faced with specific situations, such as success and failure. The system���s context awareness and simulation circumstances analysis, both in terms of results performance and execution time adaptation, are complementary mechanisms, which endow ALBidS with further adaptation and learning capabilities.

Uma aplica����o de gest��o de redes baseada no paradigma dos agentes m��veis

A massifica����o da utiliza����o das tecnologias de informa����o e da Internet para os mais variados fins, e nas mais diversas ��reas, criou problemas de gest��o das infra-estruturas de inform��tica, ��mpares at�� ao momento. A gest��o de redes inform��ticas converteu-se num factor vital para uma rede a operar de forma eficiente, produtiva e lucrativa. No entanto, a maioria dos sistemas s��o baseados no Simple Network Management Protocol (SNMP), assente no modelo cliente-servidor e com um paradigma centralizado. Desta forma subsiste sempre um servidor central que colecta e analisa dados provenientes dos diferentes elementos dispersos pela rede. Sendo que os dados de gest��o est��o armazenados em bases de dados de gest��o ou Management Information Bases (MIB���s) localizadas nos diversos elementos da rede. O actual modelo de gest��o baseado no SNMP n��o tem conseguido dar a resposta exigida. Pelo que, existe a necessidade de se estudar e utilizar novos paradigmas de maneira a que se possa encontrar uma nova abordagem capaz de aumentar a fiabilidade e a performance da gest��o de redes. Neste trabalho pretende-se discutir os problemas existentes na abordagem tradicional de gest��o de redes, procurando demonstrar a utilidade e as vantagens da utiliza����o de uma abordagem baseada em Agentes m��veis. Paralelamente, prop��e-se uma arquitectura baseada em Agentes m��veis para um sistema de gest��o a utilizar num caso real.

Negotiation platform for personalised advertising

This paper describes a multi-agent brokerage platform for near real time advertising personalisation organised in three layers: user interface, agency and marketplace. The personalisation is based on the classification of viewer profiles and advertisements (ads). The goal is to provide viewers with a personalised advertising alignment during programme intervals. The enterprise interface agents upload new ads and negotiation profiles to producer agents and new user and negotiation profiles to distributor agents. The agency layer is composed of agents that represent ad producer and media distributor enterprises as well as the market regulator. The enterprise agents offer data upload and download operations as Web Services and register the specification of these interfaces at an UDDI registry for future discovery. The market agent supports the registration and deregistration of enterprise delegate agents at the marketplace. This paper addresses the marketplace layer, an agent-based negotiation platform per se, where delegates of the relevant advertising agencies and programme distributors negotiate to create the advertising alignment that best fits a viewer profile and the advertising campaigns available. The whole brokerage platform is being developed in JADE, a multi-agent development platform. The delegate agents download the negotiation profile and upload the negotiation results from / to the corresponding enterprise agent. In the meanwhile, they negotiate using the Iterated Contract Net protocol. All tools and technologies used are open source.

B2B platform for media content personalisation

This paper proposes a novel business model to support media content personalisation: an agent-based business-to-business (B2B) brokerage platform for media content producer and distributor businesses. Distributors aim to provide viewers with a personalised content experience and producers wish to en-sure that their media objects are watched by as many targeted viewers as possible. In this scenario viewers and media objects (main programmes and candidate objects for insertion) have profiles and, in the case of main programme objects, are annotated with placeholders representing personalisation opportunities, i.e., locations for insertion of personalised media objects. The MultiMedia Brokerage (MMB) platform is a multiagent multilayered brokerage composed by agents that act as sellers and buyers of viewer stream timeslots and/or media objects on behalf of the registered businesses. These agents engage in negotiations to select the media objects that best match the current programme and viewer profiles.

Plataforma de negocia����o de componentes multim��dia

O objectivo da tese �� demonstrar a adequa����o do paradigma dos mercados electr��nicos baseados em agentes para transaccionar objectos multim��dia em fun����o do perfil dos espectadores. Esta disserta����o descreve o projecto realizado no ��mbito da plataforma de personaliza����o de conte��dos em constru����o. O dom��nio de aplica����o adoptado foi a personaliza����o dos intervalos publicit��rios difundidos pelos distribuidores de conte��dos multim��dia, i.e., pretende-se gerar em tempo ��til o alinhamento de an��ncios publicit��rios que melhor se adeq��e ao perfil de um espectador ou de um grupo de espectadores. O projecto focou-se no estudo e selec����o das tecnologias de suporte, na concep����o da arquitectura e no desenvolvimento de um prot��tipo que permitisse realizar diversas experi��ncias nomeadamente com diferentes estrat��gias e tipos de mercado. A arquitectura proposta para a plataforma consiste num sistema multiagente organizado em tr��s camadas que disponibiliza interfaces do tipo servi��o Web com o exterior. A camada de topo �� constitu��da por agentes de interface com o exterior. Na camada interm��dia encontram-se os agentes aut��nomos que modelam as entidades produtoras e consumidoras de componentes multim��dia assim como a entidade reguladora do mercado. Estes agentes registam-se num servi��o de registo pr��prio onde especificam os componentes multim��dia que pretendem negociar. Na camada inferior realiza-se o mercado que �� constitu��do por agentes delegados dos agentes da camada superior. O lan��amento do mercado �� efectuado atrav��s de uma interface e consiste na escolha do tipo de mercado e no tipo de itens a negociar. Este projecto centrou-se na realiza����o da camada do mercado e da parte da camada interm��dia de apoio ��s actividades de negocia����o no mercado. A negocia����o �� efectuada em rela����o ao pre��o da transmiss��o do an��ncio no intervalo em preenchimento. Foram implementados diferentes perfis de negocia����o com t��cticas, incrementos e limites de varia����o de pre��o distintos. Em termos de protocolos de negocia����o, adoptou-se uma variante do Iterated Contract Net ��� o Fixed Iterated Contract Net. O prot��tipo resultante foi testado e depurado com sucesso.

Sistema Multi-Agente de presta����o de servi��os de seguimento adaptados aos requisitos do utilizador

Mestrado em Engenharia Electrot��cnica e de Computadores - ��rea de Especializa����o de Telecomunica����es

Evaluating the combination of relaxation and argumentation in ontology matching negotiation

No decorrer dos ��ltimos anos, os agentes (inteligentes) de software foram empregues como um m��todo para colmatar as dificuldades associadas com a gest��o, partilha e reutiliza����o de um crescente volume de informa����o, enquanto as ontologias foram utilizadas para modelar essa mesma informa����o num formato semanticamente expl��cito e rico. �� medida que a popularidade da Web Sem��ntica aumenta e cada vez informa����o �� partilhada sob a forma de ontologias, o problema de integra����o desta informa����o amplifica-se. Em semelhante contexto, n��o �� expect��vel que dois agentes que pretendam cooperar utilizem a mesma ontologia para descrever a sua conceptualiza����o do mundo. Inclusive pode revelar-se necess��rio que agentes interajam sem terem conhecimento pr��vio das ontologias utilizadas pelos restantes, sendo necess��rio que as conciliem em tempo de execu����o num processo comummente designado por Mapeamento de Ontologias [1]. O processo de mapeamento de ontologias �� normalmente oferecido como um servi��o aos agentes de neg��cio, podendo ser requisitado sempre que seja necess��rio produzir um alinhamento. No entanto, tendo em conta que cada agente tem as suas pr��prias necessidades e objetivos, assim como a pr��pria natureza subjetiva das ontologias que utilizam, �� poss��vel que tenham diferentes interesses relativamente ao processo de alinhamento e que, inclusive, recorram aos servi��os de mapeamento que considerem mais convenientes [1]. Diferentes matchers podem produzir resultados distintos e at�� mesmo contradit��rios, criando-se assim conflitos entre os agentes. �� necess��rio que se proceda ent��o a uma tentativa de resolu����o dos conflitos existentes atrav��s de um processo de negocia����o, de tal forma que os agentes possam chegar a um consenso relativamente ��s correspond��ncias que devem ser utilizadas na tradu����o de mensagens a trocar. A resolu����o de conflitos �� considerada uma m��trica de grande import��ncia no que diz respeito ao processo de negocia����o [2]: considera-se que existe uma maior confian��a associada a um alinhamento quanto menor o n��mero de conflitos por resolver no processo de negocia����o que o gerou. Desta forma, um alinhamento com um n��mero elevado de conflitos por resolver apresenta uma confian��a menor que o mesmo alinhamento associado a um n��mero elevado de conflitos resolvidos. O processo de negocia����o para que dois ou mais agentes gerem e concordem com um alinhamento �� denominado de Negocia����o de Mapeamentos de Ontologias. �� data existem duas abordagens propostas na literatura: (i) baseadas em Argumenta����o (e.g. [3] [4]) e (ii) baseadas em Relaxamento [5] [6]. Cada uma das propostas expostas apresenta um n��mero de vantagens e limita����es. Foram propostas v��rias formas de combina����o das duas t��cnicas [2], com o objetivo de beneficiar das vantagens oferecidas e colmatar as suas limita����es. No entanto, �� data, n��o s��o conhecidas experi��ncias documentadas que possam provar tal afirma����o e, como tal, n��o �� poss��vel atestar que tais combina����es tragam, de facto, o benef��cio que pretendem. O trabalho aqui apresentado pretende providenciar tais experi��ncias e verificar se a afirma����o de melhorias em rela����o aos resultados das t��cnicas individuais se mant��m. Com o objetivo de permitir a combina����o e de colmatar as falhas identificadas, foi proposta uma nova abordagem baseada em Relaxamento, que �� posteriormente combinada com as abordagens baseadas em Argumenta����o. Os seus resultados, juntamente com os da combina����o, s��o aqui apresentados e discutidos, sendo poss��vel identificar diferen��as nos resultados gerados por combina����es diferentes e poss��veis contextos de utiliza����o.

Strategic Bidding for Electricity Markets Negotiation Using Support Vector Machines

Energy systems worldwide are complex and challenging environments. Multi-agent based simulation platforms are increasing at a high rate, as they show to be a good option to study many issues related to these systems, as well as the involved players at act in this domain. In this scope the authors��� research group has developed a multi-agent system: MASCEM (Multi- Agent System for Competitive Electricity Markets), which simulates the electricity markets environment. MASCEM is integrated with ALBidS (Adaptive Learning Strategic Bidding System) that works as a decision support system for market players. The ALBidS system allows MASCEM market negotiating players to take the best possible advantages from the market context. This paper presents the application of a Support Vector Machines (SVM) based approach to provide decision support to electricity market players. This strategy is tested and validated by being included in ALBidS and then compared with the application of an Artificial Neural Network, originating promising results. The proposed approach is tested and validated using real electricity markets data from MIBEL - Iberian market operator.