• Journal of information and communication convergence engineering
• /
• v.13 no.2
• /
• pp.123-131
• /
• 2015

Training neural networks is a complex task with great importance in the field of supervised learning. In the training process, a set of input-output patterns is repeated to an artificial neural network (ANN). From those patterns weights of all the interconnections between neurons are adjusted until the specified input yields the desired output. In this paper, a new hybrid algorithm is proposed for global optimization of connection weights in an ANN. Dynamic swarms are shown to converge rapidly during the initial stages of a global search, but around the global optimum, the search process becomes very slow. In contrast, the gradient descent method can achieve faster convergence speed around the global optimum, and at the same time, the convergence accuracy can be relatively high. Therefore, the proposed hybrid algorithm combines the dynamic particle swarm optimization (DPSO) algorithm with the backpropagation (BP) algorithm, also referred to as the DPSO-BP algorithm, to train the weights of an ANN. In this paper, we intend to show the superiority (time performance and quality of solution) of the proposed hybrid algorithm (DPSO-BP) over other more standard algorithms in neural network training. The algorithms are compared using two different datasets, and the results are simulated.

• Smart Structures and Systems
• /
• v.18 no.3
• /
• pp.425-448
• /
• 2016

In this paper, a new Pigeon Colony Algorithm (PCA) based on the features of a pigeon colony flying is proposed for solving global numerical optimization problems. The algorithm mainly consists of the take-off process, flying process and homing process, in which the take-off process is employed to homogenize the initial values and look for the direction of the optimal solution; the flying process is designed to search for the local and global optimum and improve the global worst solution; and the homing process aims to avoid having the algorithm fall into a local optimum. The impact of parameters on the PCA solution quality is investigated in detail. There are low-dimensional functions, high-dimensional functions and systems of nonlinear equations that are used to test the global optimization ability of the PCA. Finally, comparative experiments between the PCA, standard genetic algorithm and particle swarm optimization were performed. The results showed that PCA has the best global convergence, smallest cycle indexes, and strongest stability when solving high-dimensional, multi-peak and complicated problems.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.15 no.2
• /
• pp.305-313
• /
• 2002

The genetic algorithm(GA), an optimization technique based on the theory of natural selection, has proven to be relatively robust means to search for global optimum. It is converged near to the global optimum point without auxiliary information such as differentiation of function. When studying some optimization problems with continuous variables, it was found that premature saturation was reached that is no further improvement in the object function could be found over a set of iterations. Also, the general GA oscillates in the region of the new global optimum point so that the speed of convergence is decreased. This paper is to propose the concept of restarting and elitist preserving strategy as a measure to overcome this difficulty. Some benchmark examples are studied involving 3-bar truss and cantilever beam with plane stress elements. The modifications to GA improve the speed of convergence.

• Journal of Ocean Engineering and Technology
• /
• v.17 no.6
• /
• pp.53-57
• /
• 2003

Optimum design of most structural system requires that design variables are regarded as discrete quantities. This paper presents the use of Genetic Algorithm for determining the optimum design for truss with discrete variables. Genetic Algorithm are know as heuristic search algorithms, and are effective global search methods for discrete optimization. In this paper, Elitism and the method of conferring penalty parameters in the design variables, in order to achieve improved fitness in the reproduction process, is used in the Genetic Algorithm. A 10-Bar plane truss and a 25-Bar space truss are used for discrete optimization. These structures are designed for stress and displacement constraints, but buckling is not considered. In particular, we obtain continuous solution using Genetic Algorithms for a 10-bar truss, compared with other results. The effectiveness of Genetic Algorithms for global optimization is demonstrated through two truss examples.

• Journal of Korean Institute of Industrial Engineers
• /
• v.16 no.2
• /
• pp.51-62
• /
• 1990

One of the important scheduling and control problems that must be solved for the efficient operation of FMS could be the "part ordering problem" which is finding optimal sequence of parts to be released into a manufacturing system. In this paper an approach which solves the problem using simulation-optimization technique will be presented. Currently available heuristic approaches for dispatching rules can only get the near optimum at the local level because of the complexities of the system and the dependencies of its components whereas the proposed approach will try to get the global optimum for a given criterion.

• Proceedings of the KSME Conference
• /
• 2001.11a
• /
• pp.533-538
• /
• 2001

This paper describes a combinational method to compute the global and local solutions of optimization problems. The present hybrid algorithm uses both a genetic algorithm and a local concentrate search algorithm (e. g simplex method). The hybrid algorithm is not only faster than the standard genetic algorithm but also supplies a more accurate solution. In addition, this algorithm can find the global and local optimum solutions. The present algorithm can be supplied to minimize the resonance response (Q factor) and to yield the critical speeds as far from the operating speed as possible. These factors play very important roles in designing a rotor-bearing system under the dynamic behavior constraint. In the present work, the shaft diameter, the bearing length, and clearance are used as the design variables.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.19 no.7
• /
• pp.1665-1674
• /
• 1995

An optimum design has been performed to maximize specific energy (SED) of composite flywheel rotor for energy storage system. The flywheel rotor is assumed to be an axisymmetric thick laminated shell with a plane strain state for structural analysis. For the structural analysis the centrifugal force is considered and the stiffness matrix equation was derived for each ring considering the interferences between the rings. The global stiffness matrix was derived by integrating the local stiffness matrix satisfying the conditions of force and displacement compatibilities. Displacements are then calculated from the global stiffness matrix and the stresses in each ring are also calculated. 3-D intra-laminar quadratic Tsai-Wu criterion is then used for the strength analysis. An optimum procedure is also developed to find the optimal interferences and lay up angle to maximize SED using the sensitivity analysis.

• Computational Structural Engineering
• /
• v.5 no.2
• /
• pp.93-104
• /
• 1992

The final objective of optimization methods is to find global optimum accurately and efficiently. The optmization processes by simulated annealing and genetic algorithm which have stochastic search process are examined and are applied to several mathematical models and truss, beam structures. Then the robustnesses of these two methods are studied and compared with the results of deterministic optimization methods from the viewpoints of reliability and running time in obtaining the global optimum.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.52 no.6
• /
• pp.325-332
• /
• 2003

The loss minimization is one of the most important problems to save the operational cost in distribution systems. This paper presents an efficient method for optimal feeder reconfiguration of distribution systems. Chaos search algorithm (CSA) is used to reconfigure distribution systems so that active power losses are globally minimized with turning on/off sectionalizing switches. In optimization problem, the CSA searches the global optimal solution on the basis of regularity in chaotic motions and easily escapes from local or near optimal solution. The CSA is tested on 15 buses and 32 buses distribution systems, and the results indicate that it is able to determine appropriate switching options for global optimum reconfiguration.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1996.10a
• /
• pp.290-293
• /
• 1996

This paper proposes a scheme of initial cluster center selection in FCM algorithm using the genetic algorithms. The FCM algorithm often fails in the search for global optimum because it is local search techniques that search for the optimum by using hill-climbing procedures. To solve this problem, we search for a hypersphere encircling each clusters whose parameters are estimated by the genetic algorithms. Then instead of a randomized initialization for fuzzy partition matrix in FCM algorithm, we initialize each cluster center by the center of a searched hypersphere. Our experimental results show that the proposed initializing scheme has higher probabilities of finding the global or near global optimal solutions than the traditional FCM algorithm.