Modeling and optimization of semiconductor manufacturing process with neural networks

A neural network-based process model is proposed to optimize the semiconductor manufacturing process. Being different from some works in several research groups which developed neural network-based models to predict process quality with a set of process variables of only single manufacturing step, we applied this model to wafer fabrication parameters control and wafer lot yield optimization. The original data are collected from a wafer fabrication line, including technological parameters and wafer test results. The wafer lot yield is taken as the optimization target. Learning from historical technological records and wafer test results, the model can predict the wafer yield. To eliminate the "bad" or noisy samples from the sample set, an experimental method was used to determine the number of hidden units so that both good learning ability and prediction capability can be obtained.

Dynamic mechanical properties of interpenetrating polymer networks based on polyacrylates and epoxy

Interpenetrating polymer networks (IPNs) based on polyacrylate (poly(polyethylene glycol diacrylate), PEGDA) and epoxy(diglycidyl ether of bisphenol A, DGEBA) were prepared simultaneously Dynamic mechanical properties of the SINs (simultaneous interpenetrating networks) with various compositions were studied. Enhanced mechanical properties were found in this case. From the point of view of pre-swollen networks, all of the PEGDA/DGEBA IPNs were composed of the individual pre-swollen networks. A micro-phase segregation system was produced in the SIN. Glass transition temperatures shifted inward, which was attributed to molecular packing effects or mutual-entanglements of molecular segments among the individual pre-swollen networks. In accordance with the additivity of properties, namely the parallel model, the entanglement density between the two polymer networks reached its maximum at 50/50 PEGDA/DGEBA IPN.

Parameters Inversion of Fluid-Saturated Porous Media Based on Neural Networks

The multi-layers feedforward neural network is used for inversion of material constants of fluid-saturated porous media. The direct analysis of fluid-saturated porous media is carried out with the boundary element method. The dynamic displacement responses obtained from direct analysis for prescribed material parameters constitute the sample sets training neural network. By virtue of the effective L-M training algorithm and the Tikhonov regularization method as well as the GCV method for an appropriate selection of regularization parameter, the inverse mapping from dynamic displacement responses to material constants is performed. Numerical examples demonstrate the validity of the neural network method.

SOC dynamic power management using artificial neural network

Dynamic Power Management (DPM) is a technique to reduce power consumption of electronic system by selectively shutting down idle components. In this article we try to introduce back propagation network and radial basis network into the research of the system-level power management policies. We proposed two PM policies-Back propagation Power Management (BPPM) and Radial Basis Function Power Management (RBFPM) which are based on Artificial Neural Networks (ANN). Our experiments show that the two power management policies greatly lowered the system-level power consumption and have higher performance than traditional Power Management(PM) techniques-BPPM is 1.09-competitive and RBFPM is 1.08-competitive vs. 1.79, 1.45, 1.18-competitive separately for traditional timeout PM, adaptive predictive PM and stochastic PM.

Dynamic power management approaches based on neural network

Dynamic Power Management (DPM) is a technique to reduce power consumption of electronic system. by selectively shutting down idle components. In this article we try to introduce back propagation network and radial basis network into the research of the system-level policies. We proposed two PAY policies-Back propagation Power Management (BPPM) and Radial Basis Function Power management (RBFPM) which are based on Artificial Neural Networks (ANN). Our experiments show that the two power management policies greatly lowered the system-level power consumption and have higher performance than traditional Power Management(PM) techniques-BPPM is 1.09-competitive and RBFPM is 1.08-competitive vs. 1.79,145,1.18-competitive separately for traditional timeout PM, adaptive predictive PM and stochastic PM.

SOC dynamic power management using artificial neural network

Dynamic Power Management (DPM) is a technique to reduce power consumption of electronic system by selectively shutting down idle components. In this article we try to introduce back propagation network and radial basis network into the research of the system-level power management policies. We proposed two PM policies-Back propagation Power Management (BPPM) and Radial Basis Function Power Management (RBFPM) which are based on Artificial Neural Networks (ANN). Our experiments show that the two power management policies greatly lowered the system-level power consumption and have higher performance than traditional Power Management(PM) techniques-BPPM is 1.09-competitive and RBFPM is 1.08-competitive vs. 1.79 . 1.45 . 1.18-competitive separately for traditional timeout PM . adaptive predictive PM and stochastic PM.

Efficiency Dynamics on Two Coupled Small-World Networks

We investigate the effect of clusters in complex networks on efficiency dynamics by studying a simple efficiency model in two coupled small-world networks. It is shown that the critical network randomness corresponding to transition from a stagnant phase to a growing one decreases to zero as the connection strength of clusters increases. It is also shown for fixed randomness that the state of clusters transits from a stagnant phase to a growing one as the connection strength of clusters increases. This work can be useful for understanding the critical transition appearing in many dynamic processes on the cluster networks.

Fabrication and characterization of microstructured and pH sensitive interpenetrating networks hydrogel films and application in drug delivery field

Novel microstructured and pH sensitive poly(acryliac acid-co-2-hydroxyethyl methacrylate)/poly(vinyl alcohol) (P(AA-co-HEMA)/PVA) interpenetrating network (IPN) hydrogel films were prepared by radical precipitation copolymerization and sequential IPN technology. The first P(AA-co-HEMA) network was synthesized in the present of IPN aqueous solution by radical initiating, then followed by condensation reaction (Glutaraldehyde as crosslinking agent) within the resultant latex, it formed multiple IPN microstructured hydrogel film. The film samples were characterized by IR, SEM and DSC. Swelling and deswelling behaviors and mechanical property showed the novel multiple IPN nanostuctured film had rapid response and good mechanical property. The IPN films were studied as controlled drug delivery material in different pH buffer solution using cationic compound, crystal violet as a model drug.

Dynamic mechanical studies of the sulfonated butyl rubber ionomers and their blends

Dynamic mechanical properties of sulfonated butyl rubber ionomers neutralized with different amine or metallic ion (zinc or barium) and their blends with polypropylene (PP), high-density polyethylene (HDPE), or styrene-butadiene-styrene (SBS) triblock copolymer were studied using viscoelastometry. The results showed that glass transition temperatures of ion pair-containing matrix and ionic domains (T-g1 and T-g2, respectively) of amine-neutralized ionomers were lower than those of ionomers neutralized with metallic ions, and the temperature range of the rubbery plateau on the storage modulus plot for amine-neutralized ionomers was narrower. The modulus of the rubbery plateau for amine-neutralized ionomers was lower than that of ionomers neutralized with zinc or barium ion. With increasing size of the amine, the temperature range for the rubbery plateau decreased, and the height of the loss peak at higher temperature increased. Dynamic mechanical properties of blends of the zinc ionomer with PP or HDPE showed that, with decreasing ionomer content, the T-m of PP or HDPE increased and T-g1 decreased, whereas T-g2 or the upper loss peak temperature changed only slightly. The T-g1 for the blend with SBS also decreased with decreasing ionomer content. The decrease of T-g1 is attributed to the enhanced compatibilization of the matrix of the ionomer-containing ion pairs with amorphous regions of PP or HDPE or the continuous phase of SBS due to the formation of thermoplastic interpenetrating polymer networks by ionic domains and crystalline or glassy domains.

MORPHOLOGY AND DYNAMIC MECHANICAL-PROPERTIES OF AB CROSS-LINKED POLYMERS BASED ON POLYURETHANES

Polyoxypropylene glycol (PPG) (or castor oil) and toluene diisocyanate (TDI) were mixed, and the prepolymer polyurethane (PU) (I) was formed. Vinyl-terminated polyurethane (II) was prepared from (I), and hydroxyethyl acrylate, AB crosslinked polymers (ABCPs) were synthesized from (II) and vinyl monomers such as styrene, methyl methacrylate, vinyl acetate, etc. The dynamic mechanical properties and morphology of ABCPs were measured. The ABCPs based on PPG have double glass transition temperatures (T(g)) on the sigma-vs. temperature curves. They display a two-phase morphology with plastic components forming the continous phase and PU-rich domains forming the separated phase on the electron micrographs. Irregular shapes and a highly polydisperse distribution of PU-rich domain sizes were observed. The crosslink density of ABCPs has a notable effect on the morphology and properties. The average diameter of the PU-rich domains depends on the molecular weight of prepolymer PPG. The highly crosslinked structures will produce large numbers of very small domains. ABCPs based on castor oil show a single T(g) relaxation on the dynamic mechanical spectra. The compatibility between the two components is much better in ABCPs based on castor oil than in those based on PPG, because there is a high crosslink density in the former. Comparison of the dynamic mechanical spectra of ABCP and interpenetrating networks (IPN) based on castor oil with similar crosslink density and composition imply that the two components in ABCP are compatible whereas microphase separation occurs in IPN. An improvement in the compatibility is achieved by the crosslinking between the two networks.

自动生产线和柔性制造系统的建模、分析与控制

自动生产线和柔性制造系统是典型的离散事件动态系统,近几年在控制理论界受到极大重视。本文首先对自动生产线的建模与分析的研究概况做了综述,重点放在排队网络模型。然后介绍了柔性制造系统的管理与控制问题。