Abstract

Abstract
Decision problems cannot be fully represented without underlying assumptions about the Decision Maker motivations and behavior. This paper describes one technique to build a rule model representing the interaction of preferences of a Decision Maker, by training a Fuzzy Inference System based on examples. © 2005 ISAP.

Abstract
This paper presents a comparison in performance of 3 variants of Genetic Algorithms (GA) vs. 2 variants of Evolutionary Particle Swarm Optimization (EPSO), made in the extremely complex context of a multi-energy market simulation where the behavior of energy retailers is observed. The simulations are on JADE, a FIPA compliant platform based on intelligent autonomous agents running in a cluster of PCs. Each agent formulates its strategy by an inner complex simulation process using a meta-heuristic that tries to define optimum decisions. The results suggest that an EPSO approach is more efficient than GA. © 2005 ISAP.

Abstract

Abstract
This text introduces a family of Evolutionary Algorithms named EPSO - Evolutionary Particle Swarm Optimization. EPSO algorithms are evolutionary methods that borrow the movement rule from Particle Swarm Optimization methods (PSO) and use it as a recombination operator that evolves under the pressure of selection. This hybrid approach builds up an algorithm that, in several cases, in application to complex problems in Power Systems, has already proven to be more efficient, accurate and robust than classical evolutionary methods or classical PSO. The text presents the description of the method, didactic examples and examples of applications in real world problems. © 2005 ISAP.

Abstract
Grid environments are ideal for executing applications that require a huge amount of computational work, both due to the big number of tasks to execute and to the large amount of data to be analysed. Unfortunately, current tools may require that users deal themselves with corrupted outputs or early termination of tasks. This becomes incovenient as the number of parallel runs grows to easily exceed the thousands. ReCS is a user-level software designed to provide automatic detection and restart of corrupted or early terminated tasks. ReGS uses a web interface to allow the setup and control of grid execution, and provides automatic input data setup. ReGS allows the automatic detection of job dependencies, through the GRID-ADL task management language. Our results show that besides automatically and effectively managing a huge number of tasks in grid environments, ReGS is also a good monitoring tool to spot grid nodes pitfalls. © 2005 IEEE.