Browsing by Author "Wu, Chih-Hung"

Now showing 1 - 9 of 9

A Novel hybrid genetic algorithm for kernel function and parameter optimization in support vector regression
(2009-04-01)
Wu, Chih-Hung
;
Tzeng, Gwo-Hshiung
;
Lin, Rong-Ho
;
科技管理研究所
;
Institute of Management of Technology
This study developed a novel model, HCA-SVR, for type of kernel function and kernel parameter value optimization in support vector regression (SVR), which is then applied to forecast the maximum electrical daily load. A novel hybrid genetic algorithm (HGA) was adapted to search for the optimal type of kernel function and kernel parameter values of SVR to increase the accuracy of SVR. The proposed model was tested at an electricity load forecasting competition announced on the EUNITE network. The results showed that the new HGA-SVR model Outperforms the previous models. Specifically, the new HGA-SVR model can successfully identify the optimal type of kernel function and all the optimal values of the parameters of SVR with the lowest prediction error values in electricity load forecasting. Crown Copyright (C) 2008 Published by Elsevier Ltd. All rights reserved.
A real-valued genetic algorithm to optimize the parameters of support vector machine for predicting bankruptcy
(2007-02-01)
Wu, Chih-Hung
;
Tzeng, Gwo-Hshiung
;
Goo, Yeong-Jia
;
Fang, Wen-Chang
;
科技管理研究所
Two parameters, C and Q, must be carefully predetermined in establishing an efficient support vector machine (SVM) model. Therefore, the purpose of this study is to develop a genetic-based SVM (GA-SVM) model that can automatically determine the optimal parameters, C and Q, of SVM with the highest predictive accuracy and generalization ability simultaneously. This paper pioneered on employing a real-valued genetic algorithm (GA) to optimize the parameters of SVM for predicting bankruptcy. Additionally, the proposed GA-SVM model was tested on the prediction of financial crisis in Taiwan to compare the accuracy of the proposed GA-SVM model with that of other models in multivariate statistics (DA, logit, and probit) and artificial intelligence (NN and SVM). Experimental results show that the GA-SVM model performs the best predictive accuracy, implying that integrating the RGA with traditional SVM model is very successful. (C) 2005 Elsevier Ltd. All rights reserved.
Enhanced Conversion Efficiency of GaAs Solar Cells Using Ag Nanoparticles
(2010-12-01)
Shu, Gia-Wei
;
Liao, Wei-Chuan
;
Hsu, Ching-Ling
;
Lee, Ja-Yuan
;
Hsu, I-Jen
We have studied the enhancement of conversion efficiency in GaAs solar cells using silver nanoparticles with particle average sizes between 11.4 to 17.2 nm. The maximum enhancement of 20.2% in conversion efficiency is achieved for the GaAs solar cells where the Ag nanoparticles have been incorporated. The simulations according to Mie theory suggest that the increased radiative scattering and conversion efficiency is due to a combined effect of the surface plasmon resonance and the interband transition.
Enhanced Conversion Efficiency of III-V Triple-junction Solar Cells with Graphene Quantum Dots
(2016-12-16)
Lin, Tzu-Neng
;
Santiago, Svette Reina Merden S.
;
Zheng, Jie-An
;
Chao, Yu-Chiang
;
Yuan, Chi-Tsu
Graphene has been used to synthesize graphene quantum dots (GQDs) via pulsed laser ablation. By depositing the synthesized GQDs on the surface of InGaP/InGaAs/Ge triple-junction solar cells, the short-circuit current, fill factor, and conversion efficiency were enhanced remarkably. As the GQD concentration is increased, the conversion efficiency in the solar cell increases accordingly. A conversion efficiency of 33.2% for InGaP/InGaAs/Ge triple-junction solar cells has been achieved at the GQD concentration of 1.2 mg/ml, corresponding to a 35% enhancement compared to the cell without GQDs. On the basis of time-resolved photoluminescence, external quantum efficiency, and work-function measurements, we suggest that the efficiency enhancement in the InGaP/InGaAs/Ge triple-junction solar cells is primarily caused by the carrier injection from GQDs to the
Growth mechanism of CuZnInSe2 thin films grown by molecular beam epitaxy
(2013-09-01)
Tseng, Ya Hsin
;
Yang, Chu Shou
;
Wu, Chia Hsing
;
Chiu, Jai Wei
;
De Yang, Min
CuZnInSe2 (CZIS) has potential application in solar cell for absorption layer, and give an advantage to change the band gap from CuInSe2 (1.02 eV) to ZnSe (2.67 eV). Using molecular beam epitaxy technology, the CZIS thin films were grown via CuInSe (CIS) and ZnSe base. In the case of CIS, thin films were grown on Mo-coated soda lime glass with various zinc flux. CIS was transformed into chalcopyrite and sphalerite coexisting CZIS easily but it is difficult to transform into the pure sphalerite CZIS. Zn/(Zn+In+Cu) ratio has limited to approximate 36 at% and the excess-Zn played a catalyst role. In the case of ZnSe base, which was grown on GaAs (001), various In and Cu flux defined as the T-In series and T-Cu series, respectively. There are four types of compound in the T-In series and T-Cu series, including ZnSe, InxSey, ZnIn2Se4 (ZIS) and CZIS. In the T-In series under the lowest In and Cu flux, selenium (Se) were randomly combined with cations to form the CZIS. When T-In is increased in this moment, the CZIS was transformed into ZIS. In the T-Cu series, CZIS demonstrated via In-rich ZIS (Zn(In, Cu)Se) and InxSey base ((Zn, Cu)InSe). It is chalcopyrite and sphalerite coexisting structure in the medium T-Cu region. In the high T-Cu region, it is transformed into the Zn-poor and Cu-rich CZIS. (c) 2012 Elsevier B.V. All rights reserved.
Measuring the Junction Temperature of GaInP/GaInAs/Ge Multijunction Solar Cells Using Photoluminescence
(2011-09-01)
Shu, Gia-Wei
;
Tung, Chiun-Hsiang
;
Tung, Shr-Chang
;
Su, Po-Chen
;
Shen, Ji-Lin
The junction temperatures of the three individual subcells of InGaP/InGaAs/Ge solar cells were measured using photoluminescence (PL) with three different excitation lasers. With the illumination of an extra xenon-mercury lamp, the linear relationship between the PL energy and the illumination level is clearly observed and advantageously used for deriving the junction temperature. Using the Varshni relationship between the PL peak energy and the heat-sink temperature allows us to determine the junction temperature in each subcell. (C) 2011 The Japan Society of Applied Physics
Robust model matching design methodology for a stochastic synthetic gene network
(2011-03-01)
Chen, Bor-Sen
;
Chang, Chia-Hung
;
Wang, Yu-Chao
;
Wu, Chih-Hung
;
Lee, Hsiao-Ching
Synthetic biology has shown its potential and promising applications in the last decade. However, many synthetic gene networks cannot work properly and maintain their desired behaviors due to intrinsic parameter variations and extrinsic disturbances. In this study, the intrinsic parameter uncertainties and external disturbances are modeled in a non-linear stochastic gene network to mimic the real environment in the host cell. Then a non-linear stochastic robust matching design methodology is introduced to withstand the intrinsic parameter fluctuations and to attenuate the extrinsic disturbances in order to achieve a desired reference matching purpose. To avoid solving the Hamilton-Jacobi inequality (HJI) in the non-linear stochastic robust matching design, global linearization technique is used to simplify the design procedure by solving a set of linear matrix inequalities (LMIs). As a result, the proposed matching design methodology of the robust synthetic gene network can be efficiently designed with the help of LMI toolbox in Matlab. Finally, two in silico design examples of the robust synthetic gene network are given to illustrate the design procedure and to confirm the robust model matching performance to achieve the desired behavior in spite of stochastic parameter fluctuations and environmental disturbances in the host cell. (C) 2010 Elsevier Inc. All rights reserved.
Robust synthetic gene network design via library-based search method
(2011-10-01)
Wu, Chih-Hung
;
Lee, Hsiao-Ching
;
Chen, Bor-Sen
;
生物科技學系
;
Department of Biological Science and Technology
Motivation: Synthetic biology aims to develop the artificial gene networks with desirable behaviors using systematic method. These networks with desired behaviors could be constructed using diverse biological parts, which may limit the development to complex synthetic gene networks. Fortunately, some well-characterized promoter libraries for engineering gene networks are widely available. Thus, a synthetic gene network can be constructed by selecting adequate promoters from promoter libraries to achieve the desired behaviors. However, the present promoter libraries cannot be directly applied to engineer a synthetic gene network. In order to efficiently select adequate promoters from promoter libraries for a synthetic gene network, promoter libraries are needed to be redefined based on the dynamic gene regulation. Results: Based on four design specifications, a library-based search method is proposed to efficiently select the most adequate promoter set from the redefined promoter libraries by a genetic algorithm (GA) to achieve optimal reference tracking design. As the number and size of promoter libraries increase, the proposed method can play an important role in the systematic design of synthetic biology.
The reduction of antiphase boundary defects by the surfactant antimony and its application to the III-V multi-junction solar cells
(2016-01-01)
Tong, Shih-chang
;
Wang, Jyh-Shyang
;
Wu, Chii-Bin
;
Wu, Chih-Hung
;
Chang, Jung-Ting
The CuPtB ordering and the antiphase boundaries (APBs) in GaInP2 epifilms were found to be diminished by the addition of the surfactant antimony (Sb) during the epitaxy. This is evidenced by the disappearance of the APBs related deep level peak in the Raman and photoluminescence (PL) spectra, as well as the significantly increased lifetime of the minority carriers. The GaInP2/metamorphic (MM)-GaInAs/Ge multi-junction solar cells made by this method possess enhanced short circuit current density and show a narrower distribution of the open circuit voltage. We believe the use of surfactant Sb to control APBs during the growth of InGaP2 would be very useful for designing GaInP2-based devices, such as solar cells, laser diodes and light-emitting diodes. (C) 2015 Elsevier B.V. All rights reserved.