Showing papers by "National Taiwan University of Science and Technology published in 1997"

On-line tuning of fuzzy-neural network for adaptive control of nonlinear dynamical systems

Abstract: The adaptive fuzzy-neural controllers tuned online for a class of unknown nonlinear dynamical systems are proposed. To approximate the unknown nonlinear dynamical systems, the fuzzy-neural approximator is established. Furthermore, the control law and update law to tune on-line both the B-spline membership functions and the weighting factors of the adaptive fuzzy-neural controller are derived. Therefore, the control performance of the controller is improved. Several examples are simulated in order to confirm the effectiveness and applicability of the proposed methods in this paper.

Function approximation using fuzzy neural networks with robust learning algorithm

Abstract: The paper describes a novel application of the B-spline membership functions (BMF's) and the fuzzy neural network to the function approximation with outliers in training data. According to the robust objective function, we use gradient descent method to derive the new learning rules of the weighting values and BMF's of the fuzzy neural network for robust function approximation. In this paper, the robust learning algorithm is derived. During the learning process, the robust objective function comes into effect and the approximated function will gradually be unaffected by the erroneous training data. As a result, the robust function approximation can rapidly converge to the desired tolerable error scope. In other words, the learning iterations will decrease greatly. We realize the function approximation not only in one dimension (curves), but also in two dimension (surfaces). Several examples are simulated in order to confirm the efficiency and feasibility of the proposed approach in this paper.

Low Dielectric Constants of Soluble Polyimides Derived from the Novel 4,9-Bis[4-(4-aminophenoxy)phenyl]diamantane

Abstract: New diamantane-based polyimides are synthesized by reacting of 4,9-bis[4-(4-aminophenoxy)phenyl]diamantane with various aromatic tetracarboxylic dianhydrides Films of these polyimides have low dielectric constants, ranging from 258 to 274, and low moisture absorptions less than 03% Four of these films have good solubilities Three nonfluorinated polyimides were soluble in o-chlorophenol, m-cresol, N-methyl-2-pyrrolidone (NMP), and chloroform The hexafluoroisopropylidene-containing polyimide was soluble in m-cresol, chloroform, and tetrahydrofuran (THF) The soluble polyimides have molecular weights (Mn) ranging from 33 000 to 96 000 All polyimides formed tough transparent films, with tensile strengths of 553−1014 MPa, elongations to break of 61−223%, and initial moduli of 19−22 GPa Dynamic mechanical analysis (DMA) reveals that diamantane-based polyimides have two transitions on the temperature scale between 0 and 400 °C Their glass relaxations, characterized by DMA, occur at high temperatu

Optical absorption of ReS2 and ReSe2 single crystals

Abstract: The optical absorption of synthetic ReS2 and ReSe2 single crystals is reported over a temperature range from 25 to 300 K. Analysis reveals that the absorption edges of ReS2 and ReSe2 are indirect allowed transitions. The indirect band gaps at various temperatures are determined and their temperature dependence is analyzed by the Varshni equation [Physica 34, 149 (1967)] and an empirical expression proposed by O’Donnel and Chen [Appl. Phys. Lett. 58, 2924 (1991)]. The parameters that describe the temperature dependence of energy gaps of these two materials are evaluated and discussed.

Comments on the Adsorption Isotherm and Determination of Adsorption Kinetics

Abstract: The equilibrium and dynamic surface tension data are widely utilized to investigate the information of intermolecular interactions between the adsorbed surfactant molecules, the adsorption kinetics, and the mass transfer coefficients In this paper, a concept is explored: using only a limited range of equilibrium surface tension data to determine the adsorption isotherm can cause a serious mistake on the determination of adsorption kinetics and/or on the evaluation of diffusion coefficient from the dynamic surface tension data This idea is illustrated theoretically for clean interface adsorption on a pendant bubble or drop using the Langmuir and Frumkin adsorption models Two types of surfactant, one with cooperative adsorption behavior and the other with anticooperative adsorption behavior, are discussed This concept is further demonstrated using the relaxation data of surface tension of 1-octanol

Enrichment of γ‐linolenic acid from borage oil via lipase‐catalyzed reactions

Abstract: Three lipase-catalyzed reactions were utilized to enrich γ-linolenic acid in borage oil: (i) selective hydrolysis in isooctane by Candida rugosa lipase immobilized on microporous polypropylene, (ii) selective esterification of free fatty acid from saponified borage oil and n-butanol by Lipozyme IM-20, and (iii) acidolysis of the products of the previous two reactions, that is, unhydrolyzed acylglycerols and unesterified free fatty acid. In the selective hydrolysis, γ-linolenic acid content could be raised from 23.6 mol% in borage oil to 51.7% in the unhydrolyzed acylglycerols. On the other hand, γ-linolenic acid content in free fatty acid could be increased to 87% after selective esterification. Products with 65% γ-linolenic acid in their acylglycerols were obtained by means of the acidolysis reaction.

Adsorption Kinetics of 1-Octanol at the Air−Water Interface

Abstract: The adsorption of 1-octanol at an air−water interface is studied theoretically and experimentally. A video-enhanced pendant bubble tensiometry is utilized for the measurement of relaxation of surface tension. Two types of processes were investigated: the adsorption onto an initially clean air−water interface and the desorption out of a suddenly compressed interface which is originally at equilibrium. A theoretical simulation using the equilibrium surface tension is performed and it is concluded that there is no shift in controlling mechanism for the adsorption of 1-octanol molecules from an aqueous phase onto an initially clean air−water surface. The adsorption of 1-octanol onto a clean air-water interface is verified experimentally to be a diffusion-controlled process. A diffusion coefficient was computed by comparing these adsorption profiles with numerical solutions of bulk surfactant diffusion equation and the generalized Frumkin adsorption model. The re-equilibration of a compressed interface agrees...

A critical investigation of the phase shift between the inner and outer modulation for the control of machine tool Chatter

Abstract: The change of spindle speed has been recognized as an effective way to control machine tool chatter. This is because the change of spindle speed can adjust the phase shift between the inner and outer modulation. Chatter can then be suppressed or enlarged by using different phase shifts. However, it still needs to be clarified as to how the change in spindle speed is correlated with the phase shift for suppressing chatter. In this paper, the relationship between the spindle speed and phase shift is clearly studied. As a result, the behavior of chatter due to the adjustment of the phase shift between the inner and outer modulation can be understood in detail. Experimental results performed in drilling and milling operations are presented to illustrate the chatter control using the change of spindle speed.

Temperature Dependence of the Direct Band Gap of Si-containing Carbon Nitride Crystalline Films

Abstract: We have measured the temperature dependence of the spectral features in the vicinity of the direct gap E{sub g}{sup d} of Si-containing carbon nitride polycrystalline films in the temperature range between 20 and 500 K using piezoreflectance (PzR). From a detailed line-shape fit to the PzR spectra, the E{sub g}{sup d} and the broadening parameter of direct band-to-band transitions at various temperatures are determined. The temperature dependence of E{sub g}{sup d} are analyzed by the Varshni equation [Physica {bold 34}, 149 (1967)] and an empirical expression proposed by O{close_quote}Donnel and Chen [Appl. Phys. Lett. {bold 58}, 2924 (1991)]. The parameters that describe the temperature dependence of the energy gap of the material are evaluated and discussed. The broadening parameter is found to be insensitive to the temperature variation. {copyright} {ital 1997} {ital The American Physical Society}

Adsorption kinetics of C12E8 at the air-water interface: Desorption from a compressed interface

Abstract: The desorption of C 12 E 8 out of an overcrowded interface due to the sudden shrinkage of a pendant bubble in a quiescent surfactant solution is studied. A video-enhanced pendant bubble tensiometry is utilized for the measurement of the relaxation in surface tension due to the desorption of surfactant. The desorption process is found to be diffusive-kinetic mixed controlled. Rate constants of adsorption/desorption are computed by comparing these tension profiles with numerical solutions, which consider both bulk diffusion and kinetic desorption processes. The values of the kinetic rate constants of C 12 E 8 obtained from the desorption experiment are nearly the same as that obtained from the clean adsorption study (Lin, S. Y., et al., Langmuir 1996, 12, 6530). The concept that there exists a shift in controlling mechanism from diffusion control at dilute concentration to mixed diffusion-kinetic control at more elevated bulk concentration for C 12 E 8 is therefore confirmed.

The barrier, impact, morphology, and rheological properties of modified polyamides and their corresponding polyethylene—modified polyamide blends

Abstract: A systematic investigation of the effects of melt indices of polyamides (PAs) on the morphology, barrier, and impact properties of blow-molded polyethylene—polyamide and polyethylene-modified polyamide (PE—MPA) containers is reported. An appropriate alkyl carboxyl-substituted polyolefin was chosen as a compatibilizer precursor to modify PAs of varying melt indexes in a twin-screw extruder by the “reactive extrusion” process. The melt shear viscosities (ϵs) of MPAs increased with decreasing melt indices of their corresponding PAs and were higher than those of their corresponding PAs, respectively. The barrier properties of MPAs were better than those of their corresponding PAs, and the level of barrier improvement of PAs and MPAs increased consistently with decreasing melt indices of their corresponding PAs. The values of total impact energy (Et) of PAs and MPAs also improved with decreasing melt indices of PAs; however, MPAs exhibited lower values of Et than their corresponding PAs. The barrier properties of PE—MPA bottles were much better than those of PE—PA bottles. The order of barrier improvement of these PE—MPA bottles corresponded approximately to the order of barrier improvement of pure MPA when the viscosity ratio of MPA to PE was about equal to or less than one. However, regardless of exhibiting the best barrier improvement in all MPAs, a significantly smaller barrier improvement of PE—MPA5 bottles was observed when the viscosity ratio of MPA to PE was significantly greater than one. Finally, it is interesting to note that PE—MPA sheets exhibited significantly higher Et than their corresponding PE—PA sheets, although each PA was associated with a higher Et than their corresponding MPAs. Possible mechanisms accounting for the interesting behaviors noted above are suggested in this study. © 1997 John Wiley & Sons, Inc. J Appl Polym Sci 66: 2517–2527, 1997

Extended block replacement policy of a system subject to shocks

Abstract: A generalization of the block replacement (BR) policy is proposed and analyzed for a system subject to shocks. Under such a policy, an operating system is preventively replaced by new ones at times i/spl middot/T (i=1,2,3,...) independently of its failure history. If the system fails in: (a) ((i-1)/spl middot/T, (i-1)/spl middot/T+T/sub 0/), it is either replaced by a new one or minimally repaired; or (b) ((i-1)/spl middot/T+T/sub 0/, i/spl middot/T), it is either minimally repaired or remains inactive until the next planned replacement. The choice of these two actions is based on some mechanism (modeled as random) which depends on the number of shocks since the latest replacement. The average cost rate is obtained using the results of renewal reward theory. The model with two variables is transformed into a model with one variable and the optimum policy is discussed. Various special cases are considered. The results extend many of the well-known results for BR policies.

A service of home security system on intelligent network

Abstract: The authors developed "a service of home security system on intelligent network (HSSIN)", which is a home security and diversified service control network architecture. It is based on the Transmission Control Protocol/Internet Protocol (TCP/IP) standard communication protocol. The network also contains intelligent network components, and was established under the aim of creating safer, faster, more effective and more economical services that can meet customized services requirements. The HSSIN was developed for the requirements of the intelligent home security system, as the network complies with the standard communication protocol, it is adaptable to the multi-vendor environments of composite components, as well as being capable of providing intelligent network services.

Selective preparation of oxacyclic and linear phosphonates and the effect of diethyl phosphonate group on spectroscopic determination of the structures

Abstract: The formation of either oxacyclic or linear phosphonates by alternating the synthetic condition of a single method is described. Structural identification of these two isomeric oxacyclic and linear phosphonates, 2 and 3, were performed with IR, MS, 1H & 13C NMR. The 1H & 13C NMR spectroscopic data (ie., the chemical shifts and coupling constants) showed that the phosphonate group was the dominant factor in the conformation of oxacyclic phosphonate, with little effect from the phenyl and p-substituted phenyl groups. The phosphonate group with two ethoxy groups hovered over the THF-ring, thus presenting a complex 1H & 13C NMR spectra. The results indicated that the phenyl group stretched out and away from the THF-ring, thus exerting little electronic influence on the oxacyclic ring. For linear phosphonate 3, McLafferty rearrangements with both C = O and P = O groups were observed in mass fragmentations. Spectroscopic data provided the information for identification of these phosphonates.

A Boolean expression-based approach for maximum incomplete subcube identification in faulty hypercubes

Abstract: An incomplete hypercube possesses virtually every advantage of complete hypercubes, including simple deadlock-free routing, a small diameter, bounded link traffic density, a good support of parallel algorithms, and so on. It is natural to reconfigure a faulty hypercube into a maximum incomplete cube so as to lower potential performance degradation, because a hypercube so reconfigured often results in a much larger system than what is attainable according to any conventional reconfiguration scheme which identifies only complete subcubes. A maximum incomplete subcube involves one maximum complete subcube, plus certain smaller complete subcubes, and, thus, may accommodate multiple jobs of different sizes simultaneously, delivering a higher performance level. This paper proposes an efficient approach for identifying all the maximum incomplete subcubes present in a faulty hypercube. The proposed approach is on the basis of manipulating Boolean expressions, with the search space reduced considerably by taking advantage of the basic properties of faulty hypercubes during expression manipulation. It is distributed, in that every healthy node executes the same identification algorithm independently, at the same time, it is confirmed by fault simulation that our approach indeed gives rise to significantly larger reconfigured systems and requires short execution times.

Speed and accuracy of eye-gaze pointing.

Abstract: The current experiment examined the speed-accuracy trade-off of saccadic movement between two targets. Ten subjects looked alternately at two targets as fast and as accurately as possible for 2 min. under different conditions of target size, distance between targets, and direction of eye movement. Saccadic movement of the left eye was tracked and recorded with an infrared eye monitoring device to compute the starting position, ending position, and duration of each saccadic movement. Eye-movement time was significantly related to target size and distance between targets, but the speed-accuracy trade-off was significantly different from that predicted by Fitts' Law. Reaction time was not significantly changed by the direction of eye movement.

Consistent symmetric axis method for robust detection of ellipses

Abstract: Ellipse detection is an interesting issue in machine vision researches and many approaches have been suggested to determine the parameters. Recently, a symmetric axis method has been proposed to determine the centre of an ellipse, but other parameters are determined otherwise. A consistent symmetric axis method is proposed that utilises the information inherent in the symmetric axes throughout the entire process to compute all of the parameters. A closed-form solution is obtained, hence the calculation of the parameters is fast. In addition, since the determination of the symmetric axes involves a set of points on the ellipse, the procedure is robust. Experiments with disjoint ellipses, interacting ellipses, defective ellipses and real objects indicate the effectiveness of the proposed method both theoretically and practically.

A parallel algorithm for constructing a labeled tree

Abstract: A tree T is labeled when the n vertices are distinguished from one another by names such as v/sub 1/, v/sub 2/...v/sub n/. Two labeled trees are considered to be distinct if they have different vertex labels even though they might be isomorphic. According to Cayley's tree formula, there are n/sup n-2/ labeled trees on n vertices. Prufer used a simple way to prove this formula and demonstrated that there exists a mapping between a labeled tree and a number sequence. From his proof, we can find a naive sequential algorithm which transfers a labeled tree to a number sequence and vice versa. However, it is hard to parallelize. In this paper, we shall propose an O(log n) time parallel algorithm for constructing a labeled tree by using O(n) processors and O(n log n) space on the EREW PRAM computational model.

Deposition of polycrystalline β-SiC films on Si substrates at room temperature

Abstract: Polycrystalline β-SiC, with grain size up to 1500 A, has been room-temperature-deposited on Si substrates by electron cyclotron resonance chemical vapor deposition. Microwave power and the hydrogen carrier gas are the key parameters to lower the deposition temperature. According to the results of the cross-sectional transmission electron microscopy, the grain size appeared to be in the same scale as that deposited at 500 °C while a large amount of plasma-induced defects were observed in the Si substrate for the room-temperature-deposited samples. Hence, a CH4-plasma treatment prior to the β-SiC film growth was adopted, forming a SiC-like interfacial layer to suppress the substrate damages.

Conductivity of composite solid polymer electrolyte made of PEO-LiClO4-fiber

Abstract: The conductivity of composite solid polymer electrolyte (SPE) made of different compositions of poly(ethylene oxide) (PEO), LiClO4 and fiber was investigated in this study Results obtained through alternating current (AC) impedance measurements demonstrated that the conductivity of the SPE was much improved by blending fiber into it Moreover, increasing the composition of fiber added leads, thereby increasing the conductivity of the composite SPE The average conductivity of the composite SPE was 10−4 S/cm at 25 °C Performance in thermal properties was also investigated by thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC) experiments Although the mechanical strength of the composite SPE was not better than that of other materials as expected, the fiber added made it more stable

Manipulating images by using Run-length Morton Codes

Abstract: In this paper, we first present a variation of the 2-dimensional run-encoding, called the run-length Morton code encoding scheme, for compressing binary images, then we present efficient algorithms for manipulating set operations and performing conversions between the proposed encoding scheme and some well-known spatial data structures. The time complexities of set operations are linearly proportional to the size (number) of the run-length Morton codes and the time complexities of conversions are linearly proportional to the number of the nodes in the corresponding quadtree/bintree with respect to the run-length Morton codes.

Effect of addition of polyethylene glycol on the moisture transfer and waterproof properties of foam-finished nylon fabrics

Abstract: The effects of the addition of polyethylene glycol (PEG) within the coated film on the water vapor permeability and waterproof properties of the foam-coated fabrics were investigated. Foaming solutions added with PEG had lower surface tension than those without, and the viscosity and stability of the foaming solution varied with PEG of different molecular weight. When PEG is of 1,000 daltons (PEG 1000), the viscosity and stability of the foaming solution was the highest among molecular weights ranging from 400 to 16,000, whereas the surface tension was the lowest. The stability would rise proportionally with viscosity if the viscosity of the foaming solution is between 13,610 cps and 13,960 cps. When PEG 1000 was added, the particle size of the foaming solution was smaller, followed by PEG 600. But when PEG 4000 or ethylene glycol (EG) was added, the particle size was bigger. The introduction of PEG 1000 to the foaming solution produced better water vapor permeability and waterproof properties. Fabrics coated with foaming solution, to which PEG 1000 was added, had smaller and more micropores in the coated film. Coated film without the addition of PEG had less micropores, and that with the addition of EG or PEG 4000 had large pores with non-homogenous pore sizes. These results were confirmed by SEM observation.

Enhanced sintering of an Fe-Ni-P coated composite powder prepared by electroless nickel plating

Abstract: An Fe-8.2 % Ni-6.0 % P powder was prepared by electroless nickel plating on a carbonyl iron powder, where phosphorous appeared as a contaminant of the plating process. Because of the high phosphorous concentration, persistent liquid phase sintering was effective at temperatures higher than 1000 °C. The sintered microstructure was dramatically different from the conventional approaches, where a low concentration of phosphorous was added in the form of Fe3P. Sintering the alloy at a temperature as low as 1050 °C for 30 min yielded a sintered density of 98.6% theoretical and rounded grains having an average grain size of 53 µm. The rounded grains were surrounded by a large volume fraction of intergranular (Fe,Ni)3P phase, arising from the high phosphorous concentration, which slightly deteriorated the magnetic saturation but significantly increased the electrical resistivity of the alloy. Generally speaking, the magnetic saturation of the sintered alloy was improved with respect to the iron-phosphorus, iron-nickel, or iron-silicon alloys fabricated by powder processing.