Preferencje help
Widoczny [Schowaj] Abstrakt
Liczba wyników

Znaleziono wyników: 47

Liczba wyników na stronie
first rewind previous Strona / 3 next fast forward last
Wyniki wyszukiwania
Wyszukiwano:
w słowach kluczowych:  EA
help Sortuj według:

help Ogranicz wyniki do:
first rewind previous Strona / 3 next fast forward last
EN
The paper presents a way of using the wavelet transformation for analysis of acoustic emission signals (EA) accompanying the electric treeing process in solid dielectrics. A method of noise removing from the EA signals and the use of wavelet transformation for time - frequency analysis are proposed. Examples of measurement results are presented.
EN
We consider a general Markov chain model of genetic algorithm described in [3], Chapters 5 and 6. For this model, we establish an upper bound for the number of iterations which must be executed in order to find an optimal (or approximately optimal) solution with a prescribed probability. For the classical genetic algorithm with bitwise mutation, our result reduces to the main theorem of [1] in the case of one optimal solution, and gives some improvement over it in the case of many optimal solutions.
EN
The refined model for the biologically inspired agent-based computation system EMAS conformed to BDI standard is presented. The considerations are based on the model of the system dynamics as the stationary Markov chain already presented. In the course of paper space of the system states is modified in order assure state coherency and set of actions is simplified. Such a model allows for better understanding the behavior of the proposed complex systems as well as their limitations.
EN
This paper reports results of an ongoing research on MOSFET parameter extraction using the EKV model. This work continues efforts in finding the best method for efficient and robust parameter extraction based on voltage-current structure characteristics. In the extraction process, voltage-current characteristics are matched by the characteristics generated by the model. Values of parameters for which the best match is observed are the result of the extraction. The extraction process is considered an optimization problem, which is then solved by an evolutionary algorithm followed by the Nelder-Mead simplex. The influence of the measurement error on the extraction results is investigated experimentally.
EN
Evolutionary algorithms (EA) have recently become not only tools for efficient optimization of very difficult problems, but also are applied to simulate behavior of different kinds of systems, among them also games, economic systems and markets. This new domain of EA applications is known as Agent-Based Computational Economics (ACE). This article describes two applications of EA to simple market simulations. The main aim of EA in this approach is to find (sub-) optimal strategies of behavior for the participants of that market game. The first example is a simple market with only several participants and one product, well known as an instance of Cournot oligopoly game. The second example is more complicated and describes a market of permits for CO2 emission, created by the Kyoto Protocol and introduces to the simple Walrasian model the influence of calculated on-line permits prices.
EN
The aim of this paper is to present an implementation of Hierarchic Genetic Strategy (HGS) in solving the Permutation Flowshop Scheduling Problem (PFSP). We defined a hierarchic scheduler based on HGS structure for the exploration of the wide and complicated optimization landscape studied by Reeves. The objective of our work is to examine several variations of HGS operators in order to identify a configuration of operators and parameters that works best for the problem. From the experimental study we observed that HGS implementation outperforms existing schedulers in many of considered instances of a static benchmark for the problem.
EN
The paper is devoted to applications of evolutionary algorithms in identification of structures being under the uncertain conditions. Uncertainties can occur in boundary conditions, in material or geometrical parameters of structures and are modelled by three kinds of granularity: interval mathematics, fuzzy sets and theory of probability. In order to formulate the optimization problem for such a class of problems by means of evolutionary algorithms the chromosomes are considered as interval, fuzzy and random vectors whose genes are represented by: (i) interval numbers, (ii) fuzzy numbers and (iii) random variables, respectively. Description of evolutionary algorithms with granular representation of data is presented in this paper. Various concepts of evolutionary operator such as a crossover and a mutation and methods of selections are described. In order to evaluate the fitness functions the interval, fuzzy and stochastic finite element methods are applied. Several numerical tests and examples of identification of uncertain parameters are presented.
EN
Since few years the Belief Propagation [13, 14, 15] algorithm is reported as a very efficient tool to perform the optimization of systems which can be topologically transformed to the one of acceptable equivalent forms [9, 7]. The Ising system is often mentioned in these papers as a good example to present some basic foundations of BP. It is however rarely used as a tool to solve the Ising system itself. In this article we are going to present the analysis of critical properties, connected to the phase transition of magnetic system described by the Ising hamiltonian and the comparison of results to those obtained using evolutionary algorithm.
EN
This paper describes application of Strength Pareto Evolutionary Algorithm (SPEA) to an optimization problem in weather routing. The paper includes a description of SPEA algorithm and defines the constrained weather routing optimization problem. It also presents a proposal and preliminary test results of SPEA-based weather routing evolutionary algorithm.
EN
The problem of atomic structure optimization related to the minimization of its total energy is a fundamental physical problem as well as hard computational task. For the few last years we have presented some observations concerning the advantages and drawbacks of EA used as a tool to solve such questions. In this paper we would like to present some new approaches devoted to improve the general, not problem oriented part of algorithm. The results obtained for two techniques: population migration and Opposition-Based Learning show that the specific operators, designed for the given problem are still the most important part of algorithm.
EN
The present paper discusses the influence of simulation model accuracy on the convergence of electromagnetic structure simulation-based optimization. Neither response surface approximation method nor the algorithm of moving window filtering, commonly used for simulation error compensation, is not fully capable of guaranteeing proper convergence. The non-expensive device model with coarse meshing and a modified error compensation method can yield satisfactory results in a reasonable time.
EN
Molecular computing created for implementing logic systems, solving NP-difficult problems on nanoscale depends on DNA self-assembly abilities and on modifying DNA with the help of enzymes during genetic operations. In the typical DNA computing a sequence of operations executed on DNA molecules in parallel is called an algorithm, which is also determined by a model of DNA chains. This methodology is similar to the soft hardware specialized architecture driven here by heating, cooling and enzymes, especially polymerases used for copying strings. This work presents a unique approach to implementation of OR, NOR logic gates on molecules. It requires the representation of signals by DNA molecules. The presented method allows for constructing logic gates with many inputs and for executing them at the same quantity of elementary operations, regardless of a number of input signals. The NOR gate was implemented with the help of modified polymerase Taq, which stops its activity, when it meets a molecular obstacle on its way. The appropriate experiment was conducted to confirm the possibilities of the suggested implementation. Laboratory results were discussed.
EN
The method described in this paper helps to syntheses DNA (deoxyribonucleic acid) molecules with length about 1000 bp, using typical techniques enable to create strands of length up to 70 bp. The given DNA strand is divided into smaller fragments, and next these fragments are connected by proposed protocol in genetic laboratory. The evolutionary algorithm is used to find the optimal solution. The freely accessible application called longdna, based on presented ideas was implemented and tested on simulated and real data.
EN
This paper presents a new method of initialization population for sequential niching techniques, in which, evolutionary algorithm (EA) for determining a single local extreme has been employed. Knowledge of the localization of optima, determined in earlier runs of EA has been exploited in this approach. Initialization of a single individual consists in repeating its location until it is placed in the search subspace does not connected with any niche determined earlier. This approach contributes to the enhancement of convergence and to the improvement of achieved results.
EN
The article discusses the problem of identifying dynamic parameters of the Bech-Wagner model in such a way that the output characteristics of the model are close to those of a model ship bearing the shipyard symbol B-481. The identification has been carried out in the off-line mode using a genetic algorithm. The included results of simulation calculations, done using Matlab/Simulink, testify to correct operation, in terms of accuracy and rate, of the genetic algorithm.
EN
This paper is devoted to the application of an Evolutionary Algorithm to the design of Finite Impulse Response filters (FIR). A hybrid algorithm is proposed, which consists of a robust global optimization method (Evolutionary Algorithm - EA) and a good local optimization method (Quasi-Newton - QN). An experimental comparison of the hybrid algorithm against EA and QN alone indicates that EA yields filters with the better amplitude characteristics than QN. Furthermore the hybrid method yields filters with even better amplitude characteristics and in some times, it needs significantly less time than EA alone to reach good solutions.
EN
In this paper we report an on going research on applying evolutionary computation to the identification of technological parameters of MOS transistors (MOSFETs) using the current-voltage measurements. The identification consists in approximating the observed values of the current with the values generated by the transistor model. Values of parameters for which the smallest approximation error is observed are assumed to be the best estimations to the real values. The model is nonlinear and nondifferentiable, and the error function takes multiple local minima with respect to the parameter values. We apply a combination of an evolutionary algorithm together with the Nelder-Mead method to minimize the error function and we experimentally investigate the effectiveness of the proposed approach.
EN
The paper deals with the application of the global optimization methods to the multi-objective optimization of laminates. A multi-layered, fibre-reinforced and hybrid laminates are considered. Different optimization criteria connected with the laminates' cost, modal properties and stiffness are taken into account. As the optimization criteria usually cannot be satisfied simultaneously, the multi-objective optimization methods are employed. The multi-objective evolutionary algorithm with the Pareto approach is used as the global optimization method. The Finite Element Method is used to solve a boundary-value problem for the laminates. The stacking sequence and the number or plies made of different materials are design variables. The discrete as well as continuous fibre orientation angles in particular plies are considered. Numerical example presenting non-dominated solutions for two contradictory optimization criteria is attached.
EN
The paper deals with the identification of material constants in multi-layered composites. Simple and hybrid (with laminas made of different materials) laminates are considered. Material constants are presented in a stochastic form to model their uncertainty. The Evolutionary Algorithm based on such representation of the data is used as the global optimization method. Chromosomes are represented by multidimensional random vectors consisting of random genes in the form of independent random variables with the Gaussian density probability functions. The Stochastic optimization problem is replaced by a deterministic one by evolutionary computing of chromosomes having genes consisting of mean values and standard deviations. Modal analysis methods are employed to collect measurement data necessary for the identification. The Finite Element Method in Stochastic version is used to solve the boundary-value problem for laminates. Numerical examples showing efficiency of the method are presented.
EN
Non-stationary optimization with the immune based algorithms is studied in this paper. The algorithm works with a binary representation of solutions. A set of different types of binary mutation is proposed and experimentally verified. The mutations differ in the way of calculation of the number of bits to be mutated. Obtained results allow to indicate the leading formulas of calculation.
first rewind previous Strona / 3 next fast forward last
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.