• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.61 no.1
• /
• pp.135-142
• /
• 2012

In this paper, we proposed a new architecture called radial basis function-based polynomial neural networks classifier that consists of heterogeneous neural networks such as radial basis function neural networks and polynomial neural networks. The underlying architecture of the proposed model equals to polynomial neural networks(PNNs) while polynomial neurons in PNNs are composed of Fuzzy-c means-based radial basis function neural networks(FCM-based RBFNNs) instead of the conventional polynomial function. We consider PNNs to find the optimal local models and use RBFNNs to cover the high dimensionality problems. Also, in the hidden layer of RBFNNs, FCM algorithm is used to produce some clusters based on the similarity of given dataset. The proposed model depends on some parameters such as the number of input variables in PNNs, the number of clusters and fuzzification coefficient in FCM and polynomial type in RBFNNs. A multiobjective particle swarm optimization using crowding distance (MoPSO-CD) is exploited in order to carry out both structural and parametric optimization of the proposed networks. MoPSO is introduced for not only the performance of model but also complexity and interpretability. The usefulness of the proposed model as a classifier is evaluated with the aid of some benchmark datasets such as iris and liver.

• Proceedings of the KIEE Conference
• /
• 2008.04c
• /
• pp.44-46
• /
• 2008

This paper Proposes a new couple particle swarm optimization (CPSO) for multimodal functions. In this method, main particles are generated uniformly using Faure-sequences, and move accordingly to cognition only model. If any main particle detects the movement direction which has local optimum, this particle would create a new particle beside itself and make a couple. After that, all couples move accordingly to conventional particle swarm optimization (PSO) model. If these couples tend toward the same local optimum, only the best couple would be kept and the others would be eliminated. We had applied this method to some analytic multimodal functions and successfully locate all local optima.

• Proceedings of the KIEE Conference
• /
• 2011.07a
• /
• pp.187-188
• /
• 2011

This paper proposes the hybrid Complex-PSO algorithm based on the complex search method and particle swarm optimization (PSO) for unconstrained optimization. This hybridization intends to produce faster and more accurate convergence to the optimum value. These hybrid will concentrate on determining the dynamic load model parameters, the ZIP model and induction motor model parameters. Measurement-based parameter estimation, which employs measurement data to derive load model parameters, is used. The theoretical foundation of the measurement-based approach is system identification. The main objective of this paper is to demonstrate how the standard particle swarm optimization and complex method can be improved through hybridization of the two methods and the results will be compared with that of their original forms.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.26 no.12
• /
• pp.50-55
• /
• 2012

This paper presents an application of optimization method for efficient operation in micro grid. For operational efficiency, the objective function in a diesel generator consists of the fuel cost function similar to the cost functions used for the conventional fossil-fuel generating plants. The wind turbine generator is modeled by the characteristics of variable output. The cost function of fuel cell plant considers the efficiency of fuel cell. Particle swarm optimization(PSO) and sequential quadratic programming(SQP) are used for solving the problem of microgrid system operation. Also, from the results this paper presents the way to attend power markets which can buy and sell power from upper lever grids by connecting a various generation resources to micro grid.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.8
• /
• pp.1796-1803
• /
• 2013

In this paper, a novel particle swarm optimization method based on IE3D is used to design a mobile communication microstrip patch antenna. The aim of the paper is to design and fabricate an inset fed rectangular microstrip antenna and study the effect of antenna dimensions length (L), width (W) and substrate parameters relative dielectric constant (${\varepsilon}r$), substrate thickness on radiation parameters of band width. PSO algorism was applied to IE3D, low resistance against, band width and advantage, were improved.

• Proceedings of the KIEE Conference
• /
• 2008.04a
• /
• pp.77-78
• /
• 2008

This paper proposes a new algorithm for path planning and obstacles avoidance using the ant colony optimization algorithm and the particle swarm optimization. The proposed algorithm is a new hybrid algorithm that composes of the ant colony algorithm method and the particle swarm optimization method. At first, we produce paths of a mobile robot in the static environment. And then, we find midpoints of each path using the Maklink graph. Finally, the hybrid algorithm is adopted to get a shortest path. We prove the performance of the proposed algorithm is better than that of the path planning algorithm using the ant colony optimization only through simulation.

• Structural Engineering and Mechanics
• /
• v.31 no.4
• /
• pp.407-421
• /
• 2009

The present paper addresses the optimal design of composite laminates with the aim of minimizing free-edge delamination stresses. A technique involving the application of particle swarm optimization (PSO) integrated with FEM was developed for the optimization. Optimization was also conducted with the zero-order method (ZOM) included in ANSYS. The semi-analytical method, which provides an approximation of the interlaminar normal stress of laminates under in-plane load, was used to partially validate the optimization results. It was found that optimal results based on ZOM are sensitive to the starting design points, and an unsuitable initial design set will lead to a result far from global solution. By contrast, the proposed method can find the global optimal solution regardless of initial designs, and the solutions were better than those obtained by ZOM in all the cases investigated.

• ETRI Journal
• /
• v.40 no.4
• /
• pp.471-482
• /
• 2018

This paper proposes the necessity of a walking period in footstep planning and details situations in which it should be considered. An optimization-based fast footstep planner that takes the walking period into consideration is also presented. This footstep planner comprises three stages. A binary search is first used to determine the walking period. The front stride, side stride, and walking direction are then determined using the modified rapidly-exploring random tree algorithm. Finally, particle swarm optimization (PSO) is performed to ensure feasibility without departing significantly from the results determined in the two stages. The parameters determined in the previous two stages are optimized together through the PSO. Fast footstep planning is essential for coping with dynamic obstacle environments; however, optimization techniques may require a large computation time. The two stages play an important role in limiting the search space in the PSO. This framework enables fast footstep planning without compromising on the benefits of a continuous optimization approach.

• International Journal of Aeronautical and Space Sciences
• /
• v.15 no.4
• /
• pp.434-443
• /
• 2014

A turbo fan engine performance analysis and the optimization using particle swarm optimization(PSO) algorithm have been conducted to investigate the effects of major performance design parameters of an aircraft gas turbine engine. The FJ44-2C turbofan engine, which is widely used in the small business jet, CJ2 has been selected as the basic model. The design parameters consists of the bypass ratio, burner exit temperature, HP compressor ratio, fan inlet mass flow, and nozzle cooling air ratio. The sensitivity analysis of the parameters has been evaluated and the optimization of the parameters has been performed to achieve high net thrust or low specific fuel consumption.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.45 no.11
• /
• pp.45-54
• /
• 2008

In this paper, we use particle swarm optimization (PSO) to design a randomly excited and randomly spaced linear array with either the lowest side lobe level (SLL) or the narrowest beamwidth. The positions and the excitation amplitudes of the array elements are considered as variables to be controlled. The beam pattern is optimized by controlling the two variables simultaneously and randomly. The best beam patterns are obtained using PSO in the fitness function where performance is improved by the random assignment of weight coefficients to each angular sector of the beam Pattern. The weight coefficients and angles are obtained through several trial runs. Also, an extra term, ${\beta}{\ast}BW$, is added to the fitness function to account for the beamwidth as well as the SLL. Is produces the best result for the beam pattern with either the lowest SLL or the narrowest beamwidth. In the former case, the SLL and beamwidth are about -43dB and $32.2^{\circ}$, respectively, with only 10 elements. In the latter case, the SLL and beamwidth are about -26dB and $24.2^{\circ}$, respectively.