• ETRI Journal
• /
• 제28권1호
• /
• pp.59-66
• /
• 2006

Learning behaviors of a radial basis function network (RBFN) using a singular value decomposition (SVD) and stochastic gradient (SG) algorithm, together named RBF-SVD-SG, for odor sensing systems are analyzed, and a fast training method is proposed. RBF input data is from a conducting polymer sensor array. It is revealed in this paper that the SG algorithm for the fine-tuning of centers and widths still shows ill-behaving learning results when a sufficiently small convergence coefficient is not used. Since the tuning of centers in RBFN plays a dominant role in the performance of RBFN odor sensing systems, our analysis is focused on the center-gradient variance of the RBFN-SVD-SG algorithm. We found analytically that the steadystate weight fluctuation and large values of a convergence coefficient can lead to an increase in variance of the center-gradient estimate. Based on this analysis, we propose to use the least mean square algorithm instead of SVD in adjusting the weight for stable steady-state weight behavior. Experimental results of the proposed algorithm have shown faster learning speed and better classification performance.

• 정보저장시스템학회논문집
• /
• 제1권2호
• /
• pp.121-126
• /
• 2005

Recently, an optical disc system has been proposed using blue laser diode, high NA objective lens and groove only disc structure. A new method is needed to readout CD and DVD in this Blu-Ray Disc Recorder. In order to readout CD and DVD in Blu-Ray Disc Recorder, we adopted dual-lenses actuator in consideration of optical utilization efficiency, optical performance and insurance of sufficient W.D(working distance). This dual-lenses actuator has two objectives in radial direction, one is for CD/DVD and the other is for BD. We had to solve the induced problems of collision between disc and lens, DC tilt increase, AC tilt increase, 2nd resonance deterioration and AC sensitivity drop caused by disposing two lenses in an actuator. Also we developed high response twin- objective actuator with radial tilt function to apply high-speed BD Recorder or BD Player. Consequently we presented that two kinds of dual-lenses actuators for BD Recorder have been possessed good performance in 3 axis directions. And we measured eye patterns of CD, DVD and BD by using one BD optical pick-up with dual-lenses actuator.

• 전력전자학회논문지
• /
• 제9권5호
• /
• pp.467-477
• /
• 2004

본 논문에서는 극저속 영역에서 서보 전동기 구동 장치의 관성을 추정하는 기법을 제안한다. 대개의 서보 시스템에서 저속 영역의 성능 향상을 위해 사용되는 순시 속도 관측기는 관성계수와 같은 기계계 파라미터의 변동에 민감하다. 이와 같은 문제를 해결하기 위해 방사형 기저 함수망(Radial Basis Function Networks)을 이용하는 전동기관성을 추정기법을 제안한다. 실험을 통하여 제안한 추정기법의 타당성을 검증한다.

• International Journal of Computer Science & Network Security
• /
• 제24권1호
• /
• pp.31-44
• /
• 2024

People have been using more energy in the last years. Several research studies were conducted to develop sustainable energy sources that can produce clean energy to fulfill our energy requirements. Using renewable energy sources helps to decrease the harm to the environment caused by conventional power plants. Choosing the right location and capacity for DG-RESs can greatly impact the performance of Radial Distribution Systems. It is beneficial to have a good and stable electrical power supply with low energy waste and high effectiveness because it improves the performance and reliability of the system. This research investigates the ideal location and size for solar and wind power systems, which are popular methods for producing clean electricity. A new artificial intelligent algorithm called Nutcracker Optimization Algorithm (NOA) is used to find the best solution in two common electrical systems named IEEE 33 and 69 bus systems to examine the improvement in the efficiency & reliability of power system network by reducing power losses, making voltage deviation smaller, and improving voltage stability. Finally, the NOA method is compared with another method called PSO and developed Hybrid Algorithm (NOA+PSO) to validate the proposed algorithm effectiveness and enhancement of both efficiency and reliability aspects.

• 한국유체기계학회 논문집
• /
• 제18권5호
• /
• pp.42-48
• /
• 2015

To improve the core technology of the micro gas turbine, the performance test facility was developed. This paper is focusing on the explanation of the characteristics of micro gas turbine and its assist devices. Major part of micro gas turbine were radial type of compressor, annular type of combustor, radial type of turbine, thrust foil bearing, radial foil bearing and generator. The assist devices were consist of exhaust duct, inverter, data acquisition system, load bank and test cell. Before building up the test facility, the component test was previously conducted to confirm the component performance. After the test facility was prepared, the motoring test was conducted to investigate the rotor dynamic characteristics of the micro gas turbine. Also, the part load performance test was performed. With a developed micro gas turbine test facility, the improved core technology about the micro gas turbine can be suggested to the related industries.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 2008년도 정기 학술대회
• /
• pp.8-13
• /
• 2008

In this study, an adaptive node generation procedure in the radial point interpolation method is proposed. Since we set the initial configuration of nodes by subdivision of background cell, abrupt changes of inter-nodal distance between higher and lower error regions are unavoidable. This unpreferable nodal spacing induces additional errors. To obtain the smoothy nodal configuration, it's regenerated by local Delaunay triangulation algorithm This technique was originally developed to generate a set of well-shaped triangles and tetrahedra. To demonstrate the performance of proposed scheme, the results of making optimal nodal configuration with adaptive refinement method are investigated for stress concentration problems.

• Journal of Information Processing Systems
• /
• 제13권2호
• /
• pp.215-235
• /
• 2017

Foraging is a biological process, where a bacterium moves to search for nutriments, and avoids harmful substances. This paper proposes a hybrid approach integrating the bacterial foraging optimization algorithm (BFOA) in a radial basis function neural network, applied to image classification, in order to improve the classification rate and the objective function value. At the beginning, the proposed approach is presented and described. Then its performance is studied with an accent on the variation of the number of bacteria in the population, the number of reproduction steps, the number of elimination-dispersal steps and the number of chemotactic steps of bacteria. By using various values of BFOA parameters, and after different tests, it is found that the proposed hybrid approach is very robust and efficient for several-image classification.

• 센서학회지
• /
• 제20권3호
• /
• pp.151-155
• /
• 2011

A variety of pattern recognition algorithms including neural networks may be applicable to the identification of odors. In this paper, an identification technique for an electronic odor sensing system applicable to wound state monitoring is presented. The performance of the radial basis function(RBF) network is highly dependent on the choice of centers and widths in basis function. For the fine tuning of centers and widths, those parameters are initialized by an ill-conditioned genetic fuzzy c-means algorithm, and the distribution of input patterns in the very first stage, the stochastic gradient(SG), is adapted. The adaptive RBF network with singular value decomposition(SVD), which provides additional adaptation capabilities to the RBF network, is used to process data from array-based gas sensors for early detection of wound infection in burn patients. The primary results indicate that infected patients can be distinguished from uninfected patients.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제13권2호
• /
• pp.116-123
• /
• 2013

Currently, most automobiles have automatic transmission systems. The gear-shifting strategy used to generate shift patterns in transmission systems plays an important role in improving the performance of vehicles. However, conventional transmission systems have a fixed type of shift map, so it may not be enough to provide an efficient gear-shifting pattern to satisfy the demands of driver. In this study, we developed an intelligent strategy to handle these problems. This approach is based on a normalized radial basis function neural network, which can generate a flexible gear-shift pattern to satisfy the demands of drivers, including comfortable travel and fuel consumption. The method was verified through simulations.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2009년도 IWAIT
• /
• pp.303-306
• /
• 2009

Registration of microscopic section images from an organism is of importance in analyzing and understanding the function of an organism. Microscopes usually suffer from the radial distortion due to the spherical aberration. In this paper, a correction scheme for the intra-section registration is proposed. The correction scheme uses two corresponding feature points under the radial distortion model. Proposing several variations of the proposed scheme, we extensively conducted experiments for real microscopic images. Iterative versions of the correction from multiple feature points provide good performance for the registration of the optical and scanning electron microscopic images.