• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1998.10a
• /
• pp.361-367
• /
• 1998

The main two applications of the Genetic Algorithms(GA) are the optimization and the machine learning. Machine Learning has two objectives that make the complex system learn its environment and produce the proper output of a system. The machine learning using the Genetic Algorithms is called GA machine learning or genetic-based machine learning (GBML). The machine learning is different from the optimization problems in finding the rule set. In optimization problems, the population of GA should converge into the best individual because optimization problems, the population of GA should converge into the best individual because their objective is the production of the individual near the optimal solution. On the contrary, the machine learning systems need to find the set of cooperative rules. There are two methods in GBML, Michigan method and Pittsburgh method. The former is that each rule is expressed with a string, the latter is that the set of rules is coded into a string. Th classifier system of Holland is the representative model of the Michigan method. The classifier systems arrange the strength of classifiers of classifier list using the message list. In this method, the real time process and on-line learning is possible because a set of rule is adjusted on-line. A classifier system has three major components: Performance system, apportionment of credit system, rule discovery system. In this paper, we solve the food search problem with the learning and evolution of an artificial ant using the learning classifier system.

• Proceedings of the KIEE Conference
• /
• 1996.07b
• /
• pp.826-828
• /
• 1996

Genetic Algorithm(GA) is a searching or optimizing algorithm based on natural evolution principle. GA has demonstrated considerable success in providing good solutions to many nonlinear, multi-dimensional optimization problems. The preventative maintenance scheduling is a kind of dynamic optimization problem with constraints. This paper applies GA to the preventative maintenance scheduling problem. In the case study, we can get the preventative maintenance scheduling of 3-generators during 8 weeks using GA. It is shown that GA can be available to the preventative maintenance scheduling problem.

• Journal of the Korean Society for Precision Engineering
• /
• v.16 no.1 s.94
• /
• pp.116-124
• /
• 1999

In most conventional design optimization of dynamic system, design sensitivities are utilized. However, design sensitivities based optimization method has numbers of drawback. First, computing design sensitivities for dynamic system is mathematically difficult, and almost impossible for many complex problems as well. Second, local optimum is obtained. On the other hand, Genetic Algorithm is the search technique based on the performance of system, not on the design sensitivities. It is the search algorithm based on the mechanics of natural selection and natural genetics. GA search, differing from conventional search techniques, starts with an initial set of random solutions called a population. Each individual in the population is called a chromosome, representing a solution to the problem at hand. The chromosomes evolve through successive iterations, called generations. As the generation is repeated, the fitness values of chromosomes were maximized, and design parameters converge to the optimal. In this study, Genetic Algorithm is applied to the actual dynamic optimization problems, to determine the optimal design parameters of the dynamic system.

• Communications for Statistical Applications and Methods
• /
• v.10 no.1
• /
• pp.153-166
• /
• 2003

This study presents the time-divided forecasting model to integrate evolutionary optimization algorithm and change point detection based on artificial neural networks (ANN) for the prediction of (Korea) stock price index. The genetic algorithm(GA) is introduced as an evolutionary optimization method in this study. The basic concept of the proposed model is to obtain intervals divided by change points, to identify them as optimal or near-optimal change point groups, and to use them in the forecasting of the stock price index. The proposed model consists of three phases. The first phase detects successive change points. The second phase detects the change-point groups with the GA. Finally, the third phase forecasts the output with ANN using the GA. This study examines the predictability of the proposed model for the prediction of stock price index.

• Journal of Navigation and Port Research
• /
• v.30 no.2
• /
• pp.137-143
• /
• 2006

The optimum grillage design belongs to nonlinear constrained optimization problem. The determination of beam scantlings for the grillage structure is a very crucial matter out of whole structural design process. The performance of optimization methods, based on penalty functions, is highly problem-dependent and many methods require additional tuning of some variables. This additional tuning is the influences of penalty coefficient, which depend strongly on the degree of constraint violation. Moreover, Binary-coded Genetic Algorithm (BGA) meets certain difficulties when dealing with continuous and/or discrete search spaces with large dimensions. With the above reasons, Real-coded Micro-Genetic Algorithm ($R{\mu}GA$) is proposed to find the optimum beam scantlings of the grillage structure without handling any of penalty functions. $R{\mu}GA$ can help in avoiding the premature convergence and search for global solution-spaces, because of its wide spread applicability, global perspective and inherent parallelism. Direct stiffness method is used as a numerical tool for the grillage analysis. In optimization study to find minimum weight, sensitivity study is carried out with varying beam configurations. From the simulation results, it has been concluded that the proposed $R{\mu}GA$ is an effective optimization tool for solving continuous and/or discrete nonlinear real-world optimization problems.

• Proceedings of the KIEE Conference
• /
• 1996.07b
• /
• pp.820-822
• /
• 1996

This paper proposes a method for seeking the scheme of substation restoration by using genetic algorithm. Genetic algorithm (GA), first introduced by John Holland, is becoming an important tool in machine learning and function optimization. GA is a searching or optimization algorithm based on Darwinian biological evolution principle. As a test system, we assume a simple substation system and for the transformer fault, the result is obtained.

• Journal of Institute of Control, Robotics and Systems
• /
• v.12 no.4
• /
• pp.328-338
• /
• 2006

A new evolutionary computation technique, called particle swarm optimization(PSO), has been proposed and introduced recently. PSO has been inspired by the social behavior of flocking organisms, such as swarms of birds and fish schools and PSO is an algorithm that follows a collaborative population-based search model. Each particle of swarm flies around in a multidimensional search space looking for the optimal solution. Then, Particles adjust their position according to their own and their neighboring-particles experience. In this paper, characteristics of PSO such as mentioned are reviewed and compared with GA which is based on the evolutionary mechanism in natural selection. Also dimensionalities of PSO and GA are compared throughout numeric experimental studies. The comparative studies demonstrate that PSO is characterized as simple in concept, easy to implement, and computationally efficient and can generate a high-quality solution and stable convergence characteristic than GA.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.21 no.1
• /
• pp.111-117
• /
• 2007

This paper presents a new approach to evaluate reliability indices of electric distribution systems using genetic Algorithm (GA). The use of reliability evaluation is an important aspect of distribution system planning and operation to adjust the reliability level of each area. In this paper, the reliability model is based on the optimal load transforming problem to minimize load generated load point outage in each sub-section. This approach is one of the most difficult procedures and become combination problems. A new approach using GA was developed for this problem. GA is a general purpose optimization technique based on principles inspired from the biological evolution using metaphors of mechanisms such as natural selection, genetic recombination and survival of the fittest. Test results for the model system with 24 nodes 29 branches are reported in the paper.

• Proceedings of the KIEE Conference
• /
• 1999.07g
• /
• pp.2827-2829
• /
• 1999

Genetic algorithms were described as a method of solving large-scaled optimization problems with complex constraints. It has overcome their slowness, a major drawback of genetic algorithms using hardware implementation of genetic algorithm processor (GAP). In this study, we proposed GAP effectively connecting the goodness of survival-based GA, steady-state GA, tournament selection. Using Pipeline Parallel processing, handshaking protocol effectively, the proposed GAP exhibits 50% speed-up over survival-based GA which runs one million crossovers per second(1MHz). It will be used for high speed processing such of central processor of EHW, robot control and many optimization problem.

• Journal of Electrical Engineering and Technology
• /
• v.2 no.4
• /
• pp.413-419
• /
• 2007

The optimal power flow (OPF) problem was introduced by Carpentier in 1962 as a network constrained economic dispatch problem. Since then, it has been intensively studied and widely used in power system operation and planning. In the past few decades, many stochastic optimization methods such as Genetic Algorithm (GA), Evolutionary Programming (EP), and Particle Swarm Optimization (PSO) have been applied to solve the OPF problem. In particular, PSO is a newly proposed population based stochastic optimization algorithm. The main idea behind it is based on the food-searching behavior of birds and fish. Compared with other stochastic optimization methods, PSO has comparable or even superior search performance for some hard optimization problems in real power systems. Nowadays, some modifications such as breeding and selection operators are considered to make the PSO superior and robust. In this paper, we propose the Modified PSO (MPSO), in which the mutation operator of GA is incorporated into the conventional PSO to improve the search performance. To verify the optimal solution searching ability, the proposed approach has been evaluated on an IEEE 3D-bus test system. The results showed that performance of the proposed approach is better than that of the standard PSO.