Browsing by Author "Chang, CL"

Now showing 1 - 18 of 18

A least-squares finite element method for incompressible flow in stress-velocity-pressure version
(1995-12-01)
Chang, CL
;
Yang, SY
;
Hsu, CH
;
交大名義發表
;
National Chiao Tung University
In this paper we are concerned with the incompressible flow in 2-D. Introducing additional variables of derivatives of velocity, which are called stresses here, the second-order dynamic equations are reduced into a first-order system with variables of stress, velocity and pressure. Combining the compatibility condition and the divergence free condition, we have a system with six first-order equations and six unknowns. Least-squares method is performed over this extended system. The analysis shows that this method achieves optimal rates of convergence in the H-1-norm as the h approaches to zero. Numerical experiences are also available.
A least-squares finite element method for incompressible flow in stress-velocity-pressure version
(1995-12-01)
Chang, CL
;
Yang, SY
;
Hsu, CH
;
交大名義發表
;
National Chiao Tung University
In this paper we are concerned with the incompressible flow in 2-D. Introducing additional variables of derivatives of velocity, which are called stresses here, the second-order dynamic equations are reduced into a first-order system with variables of stress, velocity and pressure. Combining the compatibility condition and the divergence free condition, we have a system with six first-order equations and six unknowns. Least-squares method is performed over this extended system. The analysis shows that this method achieves optimal rates of convergence in the H-1-norm as the h approaches to zero. Numerical experiences are also available.
A recurrent fuzzy cellular neural network system with automatic structure and template learning
(2004-05-01)
Lin, CT
;
Chang, CL
;
Cheng, WC
;
電控工程研究所
;
Institute of Electrical and Control Engineering
It is widely accepted that using a set of cellular neural networks (CNNs) in parallel can achieve higher level information processing and reasoning functions either from application or biologics points of views. Such an integrated CNN system can solve more complex intelligent problems. In this paper, we propose a novel framework for automatically constructing a multiple-CNN integrated neural system in the form of a recurrent fuzzy neural network. This system, called recurrent fuzzy CNN (RFCNN), can automatically learn its proper network structure and parameters simultaneously. The structure learning includes the fuzzy division of the problem domain and the creation of fuzzy rules and CNNs. The parameter learning includes the tuning of fuzzy membership functions and CNN templates. In the RFCNN, each learned fuzzy rule corresponds to a CNN. Hence, each CNN takes care of a fuzzily separated problem region, and the functions of all CNNs are integrated through the fuzzy inference mechanism. A new online adaptive independent component analysis mixture-model technique is proposed for the structure learning of RFCNN, and the ordered-derivative calculus is applied to derive the recurrent learning rules of CNN templates in the parameter-learning phase. The proposed RFCNN provides a solution to the current dilemma on the decision of templates and/or fuzzy rules in the existing integrated (fuzzy) CNN systems. The capability of the proposed RFCNN is demonstrated on the real-world defect inspection problems. Experimental results show that the proposed scheme is effective and promising.
A two-stage least-squares finite element method for the stress-pressure-displacement elasticity equations
(1998-05-01)
Yang, SY
;
Chang, CL
;
應用數學系
;
Department of Applied Mathematics
A new stress-pressure-displacement formulation for the planar elasticity equations is proposed by introducing the auxiliary variables, stresses, and pressure. The resulting first-order system involves a nonnegative parameter that measures the material compressibility for the elastic body. A two-stage least-squares finite element procedure is introduced for approximating the solution to this system with appropriate boundary conditions. It is shown that the two-stage least-squares scheme is stable and, with respect to the order of approximation for smooth exact solutions, the rates of convergence of the approximations for all the unknowns are optimal both in the H-1-norm and in the L-2-norm. Numerical experiments with various values of the parameter are examined, which demonstrate the theoretical estimates. Among other things, computational results indicate that the behavior of convergence is uniform in the nonnegative parameter. (C) 1998 John Wiley & Sons, Inc.
Almost automatic method for reconstruction 3D geometric model of the left ventricle from 3D+1D precordial echocardiogram
(2001-01-01)
Ching, YT
;
Liu, YH
;
Chang, CL
;
Chen, JSJ
;
資訊工程學系
Echocardiography is the most convenient means for both physicians and patients for heart disease diagnosis. The 3D+1D echocardiogram. provides important information for evaluation of the 3D heart function such as the ejection fraction or wall motion. The most basic task to evaluate such functions of a heart is to segment left ventricles and reconstruct the the 3D geometric model of left ventricle from a set of echocardiographic images. Since there are many images involved, the method should not need too many user assists. In this work, we design a method for reconstructing the left ventricles with very few user assists.
An experimental study on the performance of a single discharge wire-plate electrostatic precipitator with back corona
(1999-03-01)
Chang, CL
;
Bai, HL
;
環境工程研究所
;
Institute of Environmental Engineering
The phenomenon of back corona is well known in that it may reduce the collection efficiency of an electrostatic precipitator. In this study, experimental results are presented for the influence of back corona on the performance of a laboratory scale single-discharge-wire ESP system. The effect of back corona under two power controlled methods, constant voltage operation and constant current operation, has been evaluated. The experimental results showed that as back corona occurs, the output current and the power consumption are increased under constant voltage operation. The mass collection efficiency is proportional to the applied voltage. On the other hand, for constant current operation the voltage and the power consumption are decreased under back corona. The mass collection efficiency is independent of the applied current. A lateral mixing model was also applied to assist in the discussion of corona power and particle collection efficiency under normal and stable back corona conditions. Both experimental and theoretical results indicated that the back corona has a large effect on submicron particles. The grade efficiency can be reduced by over 60% for submicron particles. The large reduction of efficiency for submicron particles is due to significant reduction in both particle charge and migration velocity. Since particles in the submicron region are more sensitive to light scattering, the opacity problem may be aggravated under back corona. (C) 1999 Elsevier Science Ltd. All rights reserved.
Applying hexagonal nanostructured zeolite particles for acetone removal
(2005-06-01)
Lin, YC
;
Bai, HL
;
Chang, CL
;
環境工程研究所
;
Institute of Environmental Engineering
This study examines the performance of a new adsorbent, hexagonal nanostructured zeolite particles (HNZP) for acetone adsorption and compares the results with that of commercial mobil synthetic zeolite-5 (ZSM-5) type zeolite. The HNZP is a pure siliceous adsorbent with different values of pore diameter and surface area being adjustable by the manufacturing condition. The results indicate that a slight increase in the average pore diameter (d) of HNZP from 2 to 2.5 nm leads to an increase in the acetone adsorption capacity, although its surface area is decreased, in which case (d = 2.5 nm) the adsorption capacity of fresh HNZP is better than that of ZSM-5 zeolite. Even for the fresh HNZP (d = 2 nm) of which the adsorption capacity is less than that of the ZSM-5 zeolite at relative humidity (RH) of 0%, its adsorption capacity is not deteriorated after repeated regeneration, but the adsorption capacity of regenerated ZSM-5 zeolite decays markedly. Thus, after only one regeneration, the adsorption capacity of HNZP (d = 2 nm) becomes better than that of the ZSM-5 zeolite. The decrease in the adsorption capacity of regenerated ZSM-5 zeolite might be because of its aluminum content that catalyzes the acetone into coke and, thus, blocks the adsorption sites. Furthermore, result on the moisture effect shows that because the pure siliceous HNZP was more hydrophobic than the ZSM-5 zeolite, the acetone adsorption efficiency of fresh HNZP (d = 2 nm) is better than that of ZSM-5 zeolite at RH = 50%.
Cardiac boundary extraction on echocardiographic images using directed graph
(2000-01-01)
Ching, YT
;
Chang, CL
;
資訊工程學系
;
Department of Computer Science
Cardiac boundary extraction on echocardiographic images is essential for quantification of cardiac function. Tracing the endocardial boundaries on the end-diastolic and end-systolic images allows the computation of clinically important measures such as ejection rate. It is a clinical need for automatically detecting the borders. In this paper, we proposed a new approach for cardiac boundary extraction on echocardiographic images by directed graph. In this approach, we spread the cardiac image in the circular direction. The spread image is mapping to a directed graph. The shortest path is found by the dynamic programming algorithm. From the implemented results, we can obtain pretty good approximation for cardiac boundary extraction.
Characteristics of copper films deposited on H(2)-plasma-treated TaN substrate by chemical vapor deposition
(2002-09-01)
Lin, CL
;
Chen, PS
;
Chang, CL
;
Chen, MC
;
電子工程學系及電子研究所
This work investigates the chemical vapor deposition Cu films deposited on the TaN substrate with and without an H(2)-plasma treatment prior to Cu film deposition and the effect of postdeposition thermal annealing. The Cu films deposited on the H(2)-plasma-treated TaN substrate have a number of favorable properties over the films deposited on the TaN substrate without the plasma treatment. These include an increased (111)-preferred orientation, smoother film surface, and a larger effective deposition rate. However, the Cu films deposited on the H(2)-plasma-treated substrate have a higher electrical resistivity, presumably due to the smaller, grain size. The postdeposition thermal annealing enhanced the (111)-preferred orientation and decreased the resistivity of the as-deposited Cu films. We presume that the H(2)-plasma treatment, resulted in a dense and uniform distribution of hydrogen radicals adsorbed on the substrate surface, leading to the shortening of incubation time and the formation of Cu films with a smoother surface and enhanced (100)-preferred orientation. A combined process including H(2)-plasma substrate treatment prior to Cu film deposition and postdeposition thermal annealing at an appropriate temperature in N(2) ambient, is proposed for the advantage of low-resistivity and high (111)-oriented Cu film deposition. (C) 2002 American Vacuum Society.
Comparison of the electronic structures of AIN nanotips grown on p- and n-type Si substrates
(2005-12-07)
Chiou, JW
;
Tsai, HM
;
Pao, CW
;
Dong, CL
;
Chang, CL
At and N K-edge x-ray absorption near-edge structure (XANES), scanning photoelectron microscopy (SPEM) and x-ray emission measurements were performed on AIN nanotips grown on p- and n-type Si substrates (p-AIN and n-AIN). Features and intensities in the Al and N K-edge XANES spectra of these AIN nanotips overall are similar. In contrast, the intensities of the valence-band SPEM spectra of p-AIN are apparently larger than those of n-AIN, which indicates that the valence-band density of states of p-AIN exceeds that of n-AIN. This result may be related to the observed enhancement of field-emission intensity of AIN nanotips grown on the p-type Si substrate.
Design of optical head with holographic optical element for small form factor drive systems
(2005-02-01)
Shih, HF
;
Chang, CL
;
Lee, KJ
;
Chang, CS
;
機械工程學系
An optical head composed of an optical module and a biaxial actuator is designed for small form factor drives. This system adopts a finite-conjugate objective lens with numerical aperture 0.65 for 654-nm wavelength. A holographic optical element is used for simplifying the optical configuration which provides a better approach for alignment. The actuator is of a spider-like suspension for the focusing and is of a swing arm for the tracking.
Destruction of styrene in an air stream by packed dielectric barrier discharge reactors
(2005-12-01)
Chang, CL
;
Bai, HL
;
Lu, SJ
;
環境工程研究所
;
Institute of Environmental Engineering
This study presents the decomposition rates of styrene vapors with non-packed and packed bed dielectric barrier discharge reactors. The concentrations of intermediate byproducts at various plasma operation conditions were evaluated. The results showed that although styrene vapors could be almost completely removed at low styrene inlet concentration of 132 ppm, the selectivity of CO2 as the major product was rather low in a non-packed bed reactor. It was found that solid carbon containing compound was the major byproduct. An increase in the styrene inlet concentration tended to reduce the styrene removal efficiency, it also led to increase in the solid byproduct. The reactors that packed with glass, Al2O3 or Pt-Pd /Al2O3 pellets could improve the styrene decomposition efficiency and reduce the formation of intermediate products, of which the best oxidation of styrene to CO2 could be achieved with a Pt-Pd /Al2O3 packed bed reactor. The carbon byproducts could also be reduced if the rector length was increased. The concentrations of ozone formed during the plasma process were also evaluated for the non-packed and packed bed reactors. The plasma reactor that packed with Pt-Pd /Al2O3 pellets was proved to have the lowest O-3 concentration.
Effect of TiN substrate plasma treatment on copper chemical vapor deposition
(2004-05-01)
Chang, CL
;
Lin, CL
;
Chen, MC
;
電子工程學系及電子研究所
;
Department of Electronics Engineering and Institute of Electronics
In this study, we investigated copper chemical vapor deposition (Cu CVD) on TiN substrates with respect to the effects of various plasma treatments. The Cu films deposited on the Ar-, H(2)-, and Ar+H(2)-plasma-treated TiN substrates all exhibit favorable properties over films deposited on the as-deposited TiN substrate. These include a smaller wetting angle of Cu nucleation and thus a smoother film surface, and an increased (111)-preferred orientation. However, the Cu films deposited on the plasma-treated substrates all exhibit a slightly higher resistivity, presumably due to a small grain size. With postdeposition thermal annealing at 400degreesC, the surface roughness of the Cu films and the Cu(111)/Cu(200) reflection ratio were both improved. We consider that a combined process including an Ar+H(2) plasma substrate treatment prior to Cu film deposition and a postdeposition thermal annealing is favorable for achieving a low surface roughness and high (111)-oriented Cu film deposition.
Effects of some geometric parameters on the electrostatic precipitator efficiency at different operation indexes
(2000-09-01)
Chang, CL
;
Bai, HL
;
環境工程研究所
;
Institute of Environmental Engineering
The performance of electrostatic precipitators (ESPs) was usually obtained under different geometric parameter values and operation indexes. Hence different conclusions were made in the literature and similar improvements may not be obtained for other applications. In this paper the effects of some geometric parameters on the performance of a wire-plate ESP at different operation indexes are evaluated theoretically. The geometric design parameters include discharge wire diameter, wire-to-wire distance, and plate-to-plate spacing. And the operation indexes considered are the electric field strength, the average current density at plates, and the corona power ratio. The results show that the collection efficiency is increased with a decrease in the wire diameter under the same electric field strength. But opposite results are observed under the same average current density at plates or the same corona power ratio applying to the ESP. The optimal wire-to-wire distance and plate-to-plate spacing vary as the electric field strength is changed. On the other hand, high collection efficiencies are always obtained with decreases in the wire-to-wire distance and the plate-to-plate spacing under the same average current density at plates or the same corona power ratio.
Finding the mitral annular lines from 2-D+1-D precordial echocardiogram using graph-search technique
(2004-03-01)
Ching, YT
;
Chen, SJ
;
Chang, CL
;
Lin, CY
;
Liu, YH
The apical four-chamber view echocardiogram collected by a transthoracic transducer can be used to evaluate the left ventricle volume. In the diastole, the left ventricle and left atrium become one chamber. In this case, the left ventricle and left atrium need to be separated using a "mitral annular line" so the volume of the left ventricle can be estimated. In this paper, a nearly automatic method for identifying the mitral annular lines from two-dimensional (2-D) + one-dimensional (1-D) precordial four-chamber view echocardiogram is presented. This method employs the optical flow technique and graph-search approach. The mitral annular line sequence is found by finding the shortest path in a weighted directed graph. The vertices in the graph are candidates for the mitral annular lines. The weights on the directed edges are determined using the optical flow technique. The proposed method requires only a physician to provide, a point that is always in the left ventricular chamber. Experimental results show that the average error for the left ventricle volume obtained based on the computed mitral annular lines is 3%.
Fuzzy Hopfield neural network with fixed weight for medical image segmentation
(2002-02-01)
Chang, CL
;
Ching, YT
;
資訊工程學系
;
Department of Computer Science
Image segmentation is a process for dividing a given image into meaningful regions with homogeneous properties. A new two step approach is proposed for medical image segmentation using a fuzzy Hopfield neural network based on both global and local gray-level information. The membership function simulated with neuron outputs is determined using a fuzzy set, and the synaptic connection weights between the neurons are predetermined and fixed to improve the efficiency of the neural network. The proposed method needs initial cluster centers. The initial centers can be obtained from the global information about the distribution of the intensities in the image, or from prior knowledge of the intensity of the region of interest. It is shown by experiments that the proposed fuzzy Hopfield neural network approach is better than most previous approaches. We also show that the global information can be used by applying the hard c-means to estimate the initial cluster centers. (C) 2002 Society of Photo-Optical Instrumentation Engineers.
H-bonded effects on supramolecular liquid crystalline trimers containing photoluminescent cores
(2001)
Lin, HC
;
Sheu, HY
;
Chang, CL
;
Tsai, CT
;
材料科學與工程學系
Several series of hydrogen-bonded (H-bonded) liquid crystalline trimers are constructed by complexation of two complementary components containing various bifunctional photoluminescent (PL) acceptor cores and monofunctional proton donors (in a 1 : 2 molar ratio). These supramolecular liquid crystalline trimers (i.e. H-bonded trimers 1/3, 1/4, 1/5, 1/6, 2/3, 2/4, 2/5 and 2/6) are, respectively, obtained from bifunctional bis-pyridyl acceptors (containing conjugated benzene and thiophene centers) (1 and 2) complexed with monofunctional carboxylic acids (containing benzene, thiophene or naphthalene) (3-6) in a 1 : 2 acceptor-donor group stoichiometry. Though the PL bis-pyridyl acceptors (1 and 2) do not possess any mesophases, the distinct mesomorphism and supramolecular architecture of these H-bonded trimers are confirmed by polarizing optical microscopy (POM), DSC, and powder X-ray diffraction (XRD) experiments. Moreover, the PL properties of the photoluminescent bis-pyridyl cores can be adjusted not only by the central structures of the cores but also by their surrounding non-photoluminescent proton donors. In general, redder shifts occur in PL spectra of H-bonded trimers when proton donors of smaller pK(a) values are H-bonded to the photoluminescent cores. Significantly, different wavelengths and polarized light of PL emission can be obtained in these supramolecular structures possessing both liquid crstalline and photoluminescent properties.
Monodisperse nanoparticle synthesis by an atmospheric pressure plasma process: An example of a visible light photocatalyst
(2004-10-27)
Bai, HL
;
Chen, CC
;
Lin, CS
;
Den, W
;
Chang, CL
An atmospheric pressure plasma-enhanced nanoparticle synthesis (APPENS) process was proposed to produce a titania-based visible light photocatalyst. The titanium tetraisopropoxide (TTIP) precursor was vaporized with a N-2 carrier gas and entered a nonthermal AC plasma system. The N-2 gas was dissociated into N atoms by the plasma energy, and then they were doped into the precursor vapors. The produced TiO2-xNx nanoparticles were highly uniform and adjustable in size. The titania-based particle sizes ranged from 25 to 40 nm, and they depended on the precursor ratio of TTIP/H-2O. In addition to producing the photocatalytic particles, the APPENS process has a potential application as a continuous-flow nanosize monodisperse aerosol generator.