Decision-making problems often require characterization of alternatives through multiple criteria. In contexts where some of these criteria interact, the decision maker (DM) must consider the interaction effects durin...
详细信息
Decision-making problems often require characterization of alternatives through multiple criteria. In contexts where some of these criteria interact, the decision maker (DM) must consider the interaction effects during the aggregation of criteria scores. The well-known ELECTRE (ELimination Et Choix Traduisant la REalite) methods were recently improved to deal with interacting criteria fulfilling several relevant properties, addressing the main types of interaction, and retaining most of the fundamental characteristics of the classical methods. An important criticism to such a family of methods is that defining its parameter values is often difficult and can involve significant challenges and high cognitive effort for the DM;this is exacerbated in the improved version whose parameters are even less intuitive. Here, we describe an evolutionary-based method in which parameter values are inferred by exploiting easy-to-make decisions made or accepted by the DM, thereby reducing his/her cognitive effort. A genetic algorithm is proposed to solve a regression-inspired nonlinear optimization problem. To the best of our knowledge, this is the first paper addressing the indirect elicitation of the ELECTRE model's parameters with interacting criteria. The proposal is assessed through both in-sample and out-of-sample experiments. Statistical tests indicate robustness of the proposal in terms of the number of criteria and their possible interactions. Results show almost perfect effectiveness to reproduce the DM's preferences in all situations.
Many multiobjective optimization problems in the engineering field are required to be solved within more or less severe time restrictions. Because the optimization criteria, the parameters, and/or constraints might ch...
详细信息
Many multiobjective optimization problems in the engineering field are required to be solved within more or less severe time restrictions. Because the optimization criteria, the parameters, and/or constraints might change with time, the optimization solutions must be recalculated when a change takes place. The time required by the optimization procedure to arrive at the new solutions should be bounded accordingly with the rate of change observed in these dynamic problems. This way, the faster the optimization algorithm is to obtain solutions, the wider is the set of dynamic problems to which that algorithm can be applied. Here, we analyze the performance of the nondominated sorting algorithm (NSGA-II), strength Pareto evolutionary algorithm (SPEA2), and single front genetic algorithms (SFGA, and SFGA2) on two different multiobjective optimization problems, with two and three objectives, respectively. For these two studied problems, the single front genetic algorithms have obtained adequate quality in the solutions in very little time. Moreover, for the second and more complex problem approached, SFGA2 and NSGA-II obtain the best hypervolume in the found set of nondominated solutions, but SFGA2 employs much less time than NSGA-II. These results may suggest that single front genetic algorithms, especially SFGA2, could be appropiate to deal with optimization problems with high rates of change, and thus stronger time constraints.
Hybrid photovoltaic (PV)-wind turbine (WT) systems with battery storage have been introduced as a green and reliable power system for remote areas. There is a steady increase in usage of hybrid energy system (HES) and...
详细信息
Hybrid photovoltaic (PV)-wind turbine (WT) systems with battery storage have been introduced as a green and reliable power system for remote areas. There is a steady increase in usage of hybrid energy system (HES) and consequently optimum sizing is the main issue for having a cost-effective system. This paper evaluates the performance of different evolutionary algorithms for optimum sizing of a PV/WT/battery hybrid system to continuously satisfy the load demand with the minimal total annual cost (TAC). For this aim, all the components are modeled and an objective function is defined based on the TAC. In the optimization problem, the maximum allowable loss of power supply probability (LPSPmax) is also considered to have a reliable system, and three well-known heuristic algorithms, namely, particle swarm optimization (PSO), tabu search (TS) and simulated annealing (SA), and four recently invented metaheuristic algorithms, namely, improved particle swarm optimization (IPSO), improved harmony search (IHS), improved harmony search-based simulated annealing (IHSBSA), and artificial bee swarm optimization (ABSO), are applied to the system and the results are compared in terms of the TAC. The proposed methods are applied to a real case study and the results are discussed. It can be seen that not only average results produced by ABSO are more promising than those of the other algorithms but also ABSO has the most robustness. Also considering LPSPmax set to 5%, the PV/battery is the most cost-effective hybrid system, and in other LPSPmax values, the PV/WT/battery is the most cost-effective systems. (C) 2015 Elsevier Ltd. All rights reserved.
A novel approach to broadband log-periodic antenna design is presented, where some of the most powerful evolutionary algorithms are applied and compared for the optimal design of wire log-periodic dipole arrays (LPDA)...
详细信息
A novel approach to broadband log-periodic antenna design is presented, where some of the most powerful evolutionary algorithms are applied and compared for the optimal design of wire log-periodic dipole arrays (LPDA) using Numerical Electromagnetics Code. The target is to achieve an optimal antenna design with respect to maximum gain, gain flatness, front-to-rear ratio (F/R) and standing wave ratio. The parameters of the LPDA optimized are the dipole lengths, the spacing between the dipoles, and the dipole wire diameters. The evolutionary algorithms compared are the Differential Evolution (DE), Particle Swarm (PSO), Taguchi, Invasive Weed (IWO), and Adaptive Invasive Weed Optimization (ADIWO). Superior performance is achieved by the IWO (best results) and PSO (fast convergence) algorithms.
Distributed evolutionary algorithms are traditionally executed on homogeneous dedicated clusters, despite most scientists have access mainly to networks of heterogeneous nodes (e.g., desktop PCs in a lab). Fitting thi...
详细信息
Distributed evolutionary algorithms are traditionally executed on homogeneous dedicated clusters, despite most scientists have access mainly to networks of heterogeneous nodes (e.g., desktop PCs in a lab). Fitting this kind of algorithms to these environments, so that they can take advantage of their heterogeneity to save running time, is still an open problem. The different computational power of the nodes affects the performance of the algorithm, and tuning or fitting it to each node properly could reduce execution time. Since the distributed evolutionary algorithms include a whole range of parameters that influence the performance, this paper proposes a study on the population size. This parameter is one of the most important, since it has a direct relationship with the number of iterations needed to find the solution, as it affects the exploration factor of the algorithm. The aim of this paper consists in validating the following hypothesis: fitting the sub-population size to the computational power of the heterogeneous cluster node can lead to an improvement in running time with respect to the use of the same population size in every node. Two parameter size schemes have been tested, an offline and an online parameter setting, and three problems with different characteristics and computational demands have been used. Results show that setting the population size according to the computational power of each node in the heterogeneous cluster improves the time required to obtain the optimal solution. Meanwhile, the same set of different size values could not improve the running time to reach the optimum in a homogeneous cluster with respect to the same size in all nodes, indicating that the improvement is due to the interaction of the different hardware resources with the algorithm. In addition, a study on the influence of the different population sizes on each stage of the algorithm is presented. This opens a new research line on the fitting (offline or online) o
The utilization of populations is one of the most important features of evolutionary algorithms (EAs). There have been many studies analyzing the impact of different population sizes on the performance of EAs. However...
详细信息
The utilization of populations is one of the most important features of evolutionary algorithms (EAs). There have been many studies analyzing the impact of different population sizes on the performance of EAs. However, most of such studies are based on computational experiments, except for a few cases. The common wisdom so far appears to be that a large population would increase the population diversity and thus help an EA. Indeed, increasing the population size has been a commonly used strategy in tuning an EA when it did not perform as well as expected for a given problem. He and Yao (2002) [8] showed theoretically that for some problem instance classes, a population can help to reduce the runtime of an EA from exponential to polynomial time. This paper analyzes the role of population further in EAs and shows rigorously that large populations may not always be useful. Conditions, under which large populations can be harmful, are discussed in this paper. Although the theoretical analysis was carried out on one multimodal problem using a specific type of EAs, it has much wider implications. The analysis has revealed certain problem characteristics, which can be either the problem considered here or other problems, that lead to the disadvantages of large population sizes. The analytical approach developed in this paper can also be applied to analyzing EAs on other problems. (C) 2011 Elsevier B.V. All rights reserved.
This paper deals with the design of electronically steerable linear arrays for intelligent antenna systems. The design problem is modeled as a multi-objective optimization problem with non-linear constraints. The well...
详细信息
This paper deals with the design of electronically steerable linear arrays for intelligent antenna systems. The design problem is modeled as a multi-objective optimization problem with non-linear constraints. The well-known NSGA-II and SPEA 2 algorithms are employed as the methodologies to solve the resulting optimization problem. The main goal and contribution of this paper is computation of the trade-off curves between side lobe level and main beam width for steerable linear arrays. The addressed problem considers a driving-point impedance restriction placed on each element in the array. This consideration makes the problem more restrictive and therefore more difficult to solve. Experimental results show the effectiveness of the algorithms for the design of steerable linear arrays. (c) 2006 Elsevier B. V. All rights reserved.
evolutionary algorithms have been used to solve a number of variable-length problems, many of which share a common representation. A set of design variables is repeatedly defined, giving the genome a segmented structu...
详细信息
evolutionary algorithms have been used to solve a number of variable-length problems, many of which share a common representation. A set of design variables is repeatedly defined, giving the genome a segmented structure. Each segment encodes a portion, frequently a single component, of the solution. For example, in a wind farm design problem each segment may encode the position and height of a single turbine. This is described as a metameric representation, with each segment referred to as a metavariable. The number of metavariables can vary among solutions, requiring modifications to the traditional fixed-length evolutionary operators. This paper surveys the literature that uses metameric representations with a focus on the problems being solved, the specifics of the representation, and the modifications to evolutionary operators. While there is little cross-referencing among the cited articles, it is demonstrated that there is already a strong overlap in their methodologies. By considering problems using a metameric representation as a single class, greater recognition of commonalities and differences among these works can be achieved. This could allow for the development of more efficient metameric evolutionary algorithms.
Recently, many evolutionary algorithms have been proposed. Compared to other algorithms, the core of the many-objective evolutionary algorithm using a one-by-one selection strategy is to select offspring one by one in...
详细信息
Recently, many evolutionary algorithms have been proposed. Compared to other algorithms, the core of the many-objective evolutionary algorithm using a one-by-one selection strategy is to select offspring one by one in environmental selection. However, it does not perform well in resolving large-scale many-objective optimization problems. In addition, a large amount of meaningful information in the population of the previous iteration is not retained. The information feedback model is an effective strategy to reuse the information from previous populations and integrate it into the update process of the offspring. Based on the original algorithm, this paper proposes a series of many-objective evolutionary algorithms, including six new algorithms. Experiments were carried out in three different aspects. Using the same nine benchmark problems, we compared the original algorithm with six new algorithms. algorithms with excellent performance were selected and compared with the latest studies using the information feedback model from two aspects. Then, the best one was selected for comparison with six state-of-the-art many-objective evolutionary algorithms. Additionally, non-parametric statistical tests were conducted to evaluate the different algorithms. The comparison, with up to 15 objectives and 1500 decision variables, showed that the proposed algorithm achieved the best performance, indicating its strong competitiveness.
Population initialization is a crucial task in evolutionary algorithms because it can affect the convergence speed and also the quality of the final solution. If no information about the solution is available, then ra...
详细信息
Population initialization is a crucial task in evolutionary algorithms because it can affect the convergence speed and also the quality of the final solution. If no information about the solution is available, then random initialization is the most commonly used method to generate candidate solutions (initial population). This paper proposes a novel initialization approach which employs opposition-based learning to generate initial population. The conducted experiments over a comprehensive set of benchmark functions demonstrate that replacing the random initialization with the opposition-based population initialization can accelerate convergence speed. (C) 2007 Elsevier Ltd. All rights reserved.
暂无评论