Felipe Campelo

A clonal selection algorithm for optimization in electromagnetics

This paper proposes the real-coded clonal selection algorithm (RCSA) for use in electromagnetic design optimization. Some features of the algorithm, such as the number of clones, mutation range, and the fraction of the population selected each generation are discussed. The TEAM Workshop problem 22 is investigated, in order to illustrate the performance of the algorithm in a real electromagnetic problem. The results obtained, a set of optimal solutions representing a broader range of options for the designer, are compared with those achieved by a genetic algorithm, showing the efficiency of the RCSA in practical optimization problems.

Electric Distribution Network Expansion Under Load-Evolution Uncertainty Using an Immune System Inspired Algorithm

This paper addresses the problem of electric distribution network expansion under condition of uncertainty in the evolution of node loads in a time horizon. An immune-based evolutionary optimization algorithm is developed here, in order to find not only the optimal network, but also a set of suboptimal ones, for a given most probable scenario. A Monte-Carlo simulation of the future load conditions is performed, evaluating each such solution within a set of other possible scenarios. A dominance analysis is then performed in order to compare the candidate solutions, considering the objectives of: smaller infeasibility rate, smaller nominal cost, smaller mean cost and smaller fault cost. The design outcome is a network that has a satisfactory behavior under the considered scenarios. Simulation results show that the proposed approach leads to resulting networks that can be rather different from the networks that would be found via a conventional design procedure: reaching more robust performances under load evolution uncertainties

Optimization of Inductors Using Evolutionary Algorithms and Its Experimental Validation

This paper presents parameter and topology optimization of inductor shapes using evolutionary algorithms. The goal of the optimization is to reduce the size of inductors satisfying the specifications on inductance values under weak and strong bias-current conditions. The inductance values are computed from the finite-element (FE) method taking magnetic saturation into account. The result of the parameter optimization, which leads to significant reduction in the volume, is realized for test, and the dependence of inductance on bias currents is experimentally measured, which is shown to agree well with the computed values. Moreover, novel methods are introduced for topology optimization to obtain inductor shapes with homogeneous ferrite cores suitable for mass production.

A modified immune network algorithm for multimodal electromagnetic problems

Some optimization algorithms based on theories from immunology have the feature of finding an arbitrary number of optima, including the global solution. However, this advantage comes at the cost of a large number of objective function evaluations, in most cases, prohibitive in electromagnetic design. This paper proposes a modified version of the artificial immune network algorithm (opt-AINet) for electromagnetic design optimization. The objective of this modified AINet (m-AINet) is to reduce the computational effort required by the algorithm, while keeping or improving the convergence characteristics. Another improvement proposed is to make it more suitable for constrained problems through the utilization of a specific constraint-handling technique. The results obtained over an analytical problem and the design of an electromagnetic device show the applicability of the proposed algorithm

Optimization of Cost Functions Using Evolutionary Algorithms with Local Learning and Local Search

Evolutionary algorithms can benefit from their association with local search operators, giving rise to hybrid or memetic algorithms. The cost of the local search may be prohibitive, particularly when dealing with computationally expensive functions. We propose the use of local approximations in the local search phase of memetic algorithms for optimization of cost functions. These local approximations are generated using only information already collected by the algorithm during the evolutionary process, requiring no additional evaluations. The local search improves some individuals of the population, hence speeding up the overall optimization process. We investigate the design of a loudspeaker magnet with seven variables. The results show the improvement achieved by the proposed combination of local learning and search within evolutionary algorithms

Pareto cone ε-dominance: improving convergence and diversity in multiobjective evolutionary algorithms

Relaxed forms of Pareto dominance have been shown to be the most effective way in which evolutionary algorithms can progress towards the Pareto-optimal front with a widely spread distribution of solutions. A popular concept is the e-dominance technique, which has been employed as an archive update strategy in some multiobjective evolutionary algorithms. In spite of the great usefulness of the e-dominance concept, there are still difficulties in computing an appropriate value of e that provides the desirable number of nondominated points. Additionally, several viable solutions may be lost depending on the hypergrid adopted, impacting the convergence and the diversity of the estimate set. We propose the concept of cone e-dominance, which is a variant of the e-dominance, to overcome these limitations. Cone e-dominance maintains the good convergence properties of e-dominance, provides a better control over the resolution of the estimated Pareto front, and also performs a better spread of solutions along the front. Experimental validation of the proposed cone e-dominance shows a significant improvement in the diversity of solutions over both the regular Pareto-dominance and the e-dominance.

Overview of artificial immune systems for multi-objective optimization

Evolutionary algorithms have become a very popular approach for multiobjective optimization in many fields of engineering. Due to the outstanding performance of such techniques, new approaches are constantly been developed and tested to improve convergence, tackle new problems, and reduce computational cost. Recently, a new class of algorithms, based on ideas from the immune system, have begun to emerge as problem solvers in the evolutionary multiobjective optimization field. Although all these immune algorithms present unique, individual characteristics, there are some trends and common characteristics that, if explored, can lead to a better understanding of the mechanisms governing the behavior of these techniques. In this paper we propose a common framework for the description and analysis of multiobjective immune algorithms.

A multiobjective proposal for the TEAM benchmark problem 22

The TEAM benchmark problem 22 is an important optimization problem in electromagnetic design, which can be formulated as a constrained mono-objective problem or a multiobjective one with two objectives. In this paper, we propose a multiobjective version with three objectives, whose third objective is related to the quench constraint and the better use of the superconducting material. The formulation proposed yields results that provide new alternatives to the designer. We solved the formulation proposed using the multiobjective clonal selection algorithm. After that, we selected a particular solution using a simple decision making procedure

A comparison of dominance criteria in many-objective optimization problems

In this paper, we analyze four dominance criteria in terms of their ability to adequately order sets of points in multi-and many-objective optimization problems. The use of relaxed and alternative dominance relationships has been an important tool for improving the performance of multiobjective evolutionary optimization algorithms, and their ordering ability is among the most important characteristics responsible for such improvement. Three relaxed formulations of dominance are investigated, along with the traditional Pareto ordering, in order to provide a comparison baseline. The results obtained show that all three relaxed dominance approaches presented greater robustness to the increase in the number of objectives, and are therefore more appropriate for use in many-objective optimization algorithms.

Local Learning and Search in Memetic Algorithms

The use of local search in evolutionary techniques is believed to enhance the performance of the algorithms, giving rise to memetic or hybrid algorithms. However, in many continuous optimization problems the additional cost required by local search may be prohibitive. Thus we propose the local learning of the objective and constraint functions prior to the local search phase of memetic algorithms, based on the samples gathered by the population through the evolutionary process. The local search operator is then applied over this approximated model. We perform some experiments by combining our approach with a real-coded genetic algorithm. The results demonstrate the benefit of the proposed methodology for costly black-box functions.