Taipei, Taiwan

University of Taipei

www.utaipei.edu.tw
Taipei, Taiwan
SEARCH FILTERS
Time filter
Source Type

Chung P.-H.,University of Taipei | Li T.,University of Hawaii at Manoa
Journal of Climate | Year: 2013

The interdecadal change of the mean state and two types of El Niño was investigatedbased on the analysis of observational data from 1980 to 2010. It was found that easterlytrades and sea surface temperature (SST) gradients across the equatorial Pacific undergo a regime change in 1998/99, with enhanced trades and a significant cooling (warming)overtropical eastern (western) Pacific in the later period. Accompanying this mean state change is more frequent occurrence of central Pacific (CP) El Niño during 1999-2010.The diagnosis of air-sea feedback strength showed that atmospheric precipitation and wind responses to CP El Niño are greater than those to the eastern Pacific (EP) El Niño for givena unit SST anomaly (SSTA) forcing. The oceanic response to the same wind forcing, however, is greater in the EP El Niño than in the CP El Niño. A mixed layer heat budget analysis reveals that zonal advection (thermocline change induced vertical advection) primarily contributes to the CP (EP) El Niño growth. The role of the mean SST zonal gradient in El Niño selection was investigated through idealized numerical experiments. With the increase of the background zonal SST gradient, the anomalous wind and convection response toa specified EP or CP SSTA shift to the west. Such adifference results in a bifurcation of maximum SSTA tendency, as shown from a simple ocean model. The numerical results support the notion that a shift to the La Niño-likeinterdecadal mean state is responsible for more frequent occurrence of CP-type El Niño. © 2013 American Meteorological Society.


Fluctuations in the stock market follow the principle of volatility clustering in which changes are cataloged by similarity; as such, large changes tend to follow large changes, and small changes tend to follow small changes. This clustering is one of the major reasons why many generalized autoregression conditional heteroscedasticity (GARCH) models do not forecast the stock market well. In this paper, an adaptive Fuzzy-GARCH model with particle swarm optimization (PSO) is proposed to solve this problem. The adaptive Fuzzy-GARCH model refers to both GARCH models and the parameters of membership functions, which are determined by the characteristics of market itself. Here, we present an iterative algorithm based on PSO to estimate the parameters of the membership functions. The PSO method aims to achieve a global optimal solution with a rapid convergence rate. The three stock markets of Taiwan, Japan, and Germany were analyzed to illustrate the performance of the proposed method. © 2011 Elsevier Inc. All rights reserved.


Yin M.-S.,University of Taipei
Expert Systems with Applications | Year: 2013

The grey system theory, identified as one of the developed multiple attribute decision-making techniques, has been published by and indexed in over 300 internationally recognized refereed journals. The objective of this study is to conduct a bibliometric study on publication and citation patterns of grey system theory published from 1996 to 2010 through a systemic search using the ISI web-based databases with a specific focus on grey relational analysis (GRA) and grey prediction. Results of the study demonstrate that there has been a substantial increase in the number of peer-reviewed papers on GRA or grey prediction indexed by the ISI Web of Knowledge. Also, citation analysis was used to examine the contributions of GRA and grey prediction studies. This bibliometric analysis would provide a ready reference for scholarly works on GRA and grey prediction, and serve as an informative summary kit for future research works. © 2012 Elsevier B.V. All rights reserved.


By integrating neural networks (NNs) with turn-on transient energy analysis, this work attempts to recognize demand load, including the buyers' load on the power systems and the internal load on the cogeneration systems, thereby increasing the recognition accuracy in a non-intrusive energy management (NIEM) system. Analysis results reveal that an NIEM system and a new method that is based on genetic algorithms (GA) can effectively manage energy demand in an optimal economic dispatch for cogeneration systems with multiple cogenerators, which generate power for buyers. Furthermore, the global optimum of economic dispatch under typical environmental and operating constraints of cogeneration systems is found using the proposed approach, which is based on genetic algorithms. Moreover, the use of the proposed GA-based method for economic dispatch can substantially reduce computational time, fuel cost, power cost and air pollution. © 2010 Elsevier Ltd.


This paper proposes a novel graphical method to compute all feasible gain and phase margin specifications-oriented robust PID controllers to stabilize uncertain control systems with time-varying delay. A virtual gain-phase margin tester compensator is incorporated to guarantee the concerned system with certain robust safety margins. The complex Kharitonov theorem is used to characterize the parametric uncertainties of the considered system and is exploited as a stability criterion for the Hurwitz property of a family of polynomials with complex coefficients varying within given intervals. The coefficients of the characteristic equation are overbounded and eight vertex Kharitonov polynomials are derived to perform stability analysis. The stability equation method and the parameter plane method are exploited to portray constant gain margin and phase margin boundaries. The feasible controllers stabilizing every one of the eight vertex polynomials are identified in the parameter plane by taking the overlapped region of the plotted boundaries. The overlapped region of the useful region of each vertex polynomial is the Kharitonov region, which represents all the feasible specifications-oriented robust PID controller gain sets. Variations of the Kharitonov region with respect to variations of the derivative gain are extensively studied. The way to select representative points from the Kharitonov region for designing robust controllers is suggested. Finally, three illustrative examples with computer simulations are provided to demonstrate the effectiveness and confirm the validity of the proposed methodology. Based on the pre-specified gain and phase margin specifications, a non-conservative Kharitonov region can be graphically identified directly in the parameter plane for designing robust PID controllers. © 2011 Elsevier Ltd. All rights reserved.


Hung J.-C.,University of Taipei
Applied Soft Computing Journal | Year: 2011

This paper studies volatility forecasting in the financial stock market. In general, stock market volatility is time-varying and exhibits clustering properties. Thus, this paper presents the results of using a fuzzy system method to analyze clustering in generalized autoregressive conditional heteroskedasticity (GARCH) models. It also uses the adaptive method of recursive least-squares (RLS) to forecast stock market volatility. The fuzzy GARCH model represents a joint estimation method; the membership function parameters together with the GARCH model parameters make this problem of stock market is highly nonlinear and complicated. This study presents an iterative algorithm based on a genetic algorithm (GA) to estimate the parameters of the membership functions and the GARCH models. In this paper, the GA method is employed to identify a global optimal solution with a fast convergence rate in the context of the fuzzy GARCH model estimation problem studied here. Based on simulation results, we determined that both the estimation of in-sample and the forecasting of out-of-sample volatility performance are significantly improved when the GARCH model considers both the clustering effect and the adaptive forecast. © 2011 Elsevier B.V.


Hung J.-C.,University of Taipei
Applied Soft Computing Journal | Year: 2013

This study considers the problem of estimating the direction-of-arrival (DOA) for code-division multiple access (CDMA) signals. In this type of problem, the associated cost function of the DOA estimation is generally a computationally-expensive and highly-nonlinear optimization problem. A fast convergence of the global optimization algorithm is therefore required to attain results within a short amount of time. In this paper, we propose a new application of the modify particle swarm optimization (MPSO) structure to achieve a global optimal solution with a fast convergence rate for this type of DOA estimation problem. The MPSO uses a first-order Taylor series expansion of the objective function to address the issue of enhanced PSO search capacity for finding the global optimum leads to increased performance. The first-order Taylor series approximates the spatial scanning vector in terms of estimating deviation results in and reducing to a simple one-dimensional optimization problem and the estimating deviation has the tendency to fly toward a better search area. Thus, the estimating deviation can be used to update the velocity of the PSO. Finally, several numerical examples are presented to illustrate the design procedure and to confirm the performance of the proposed method. © 2012 Elsevier B.V.


This study investigates relationships among food safety knowledge, attitudes and hazard analysis critical control point (HACCP) practices in restaurant employees in Taiwan. The authors administered a baseline questionnaire to 542 restaurant employees to assess their food safety knowledge, attitude and HACCP practices. A total of 421 valid questionnaires were returned and used in analysis. Mean scores for each survey item were calculated and used in a structural equation model (SEM) designed to assess interrelationships between the three. Participants scored an average 84.7% correct in food safety knowledge, with highest and lowest correct scores in, respectively, the food poisoning and good hygienic practices (GHP) constructs. The highest score in the attitude section was " concern for food safety" followed by " self-improvement." With the exception of the food poisoning construct, this study found correlations among knowledge, attitude, and HACCP practices, with attitude mediating the relationship between knowledge and HACCP practices. The implications of these findings are discussed. © 2012 Elsevier Ltd.


Tsai C.-M.,University of Taipei
Pattern Recognition | Year: 2012

The study applies an intelligent region-based thresholding method for the binarization of color document images with highlighted regions. The results also indicate that the proposed method can threshold simultaneously when the background is gradually changing, reversed, or inseparable from the foreground, with efficient binarization results. Rather than the traditional method of scanning the entire document at least once, this method intelligently divides a document image into several foreground regions and decides the background range for each foreground region, in order to effectively process the detected document regions. Experimental results demonstrate the high effectiveness of the proposed method in providing promising binarization results with low computational cost. Furthermore, the results of the proposed method are more accurate than global, region-based, local, and hybrid methods. Images were analyzed using MODI OCR measurement data such as recall rate and precision rate. In particular, when test images produced under inadequate illumination are processed using the proposed method, the binarization results of this method have better visual quality and better measurable OCR performance than compared global, region-based, local, and hybrid methods. Moreover, the proposed algorithm can be run in an embedded system due to its simplicity and efficiency. © 2011 Elsevier Ltd All rights reserved.


Energy management systems strive to use energy resources efficiently, save energy, and reduce carbon output. This study proposes transient feature analyses of the transient response time and transient energy on the power signatures of non-intrusive demand monitoring and load identification to detect the power demand and load operation.This study uses the wavelet transform (WT) of the time-frequency domain to analyze and detect the transient physical behavior of loads during the load identification. The experimental results show the transient response time and transient energy are better than the steady-state features to improve the recognition accuracy and reduces computation requirements in non-intrusive load monitoring (NILM) systems. The discrete wavelet transform (DWT) is more suitable than short-time Fourier transform (STFT) for transient load analyses. © 2012 by the authors.

Loading University of Taipei collaborators
Loading University of Taipei collaborators