Sharif Digital Repository

Surface pressure measurements were conducted for a pitch oscillation wing in a subsonic closed circuit wind tunnel. Experimental results have been used to train a multilayer perceptron network to foresee the effect of modification of oscillation amplitude and reduced frequency. Consistent results are obtained both for the training data as well as generalization to other amplitudes and reduced frequencies. This work indicates that artificial neural networks can reliably predict aerodynamic coefficients and forecast the effects of oscillation amplitude as well as reduced frequency on the wind turbine blade performance. Moreover, this study introduces a new tool for the designers to have enough... 

Neural networks have been used for process fault diagnosis. In this work, the cluster analysis is used to design a structurally optimized Probabilistic Neural Network. This network is called the Clustered-Based Design Probabilistic Neural Network (CBDPNN). The CBDPNN is capable of diagnosing the faults of nonlinear processes operating over several regions. The performance and training status of the proposed CBDPNN is compared to a conventional Multi-Layer Perceptron (MLP) that is trained on the whole operating region. Simulation results indicate that both schemes have the same performance, but, the training of CBDPNN is much easier than the conventional MLP, although it has about 50% more... 

This paper presents a model to analyze pull-in phenomenon and dynamics of multi layer microplates using coupled finite element and finite difference methods. Firstorder shear deformation theory is used to model dynamical system using finite element method, while Finite difference method is applied to solve the nonlinear Reynolds equation of squeeze film damping. Using this model, Pull-in analysis of single layer and multi layer microplates are studied. The results of pull-in analysis are in good agreement with literature. Validating our model by pull-in results, an algorithm is presented to study dynamics of microplates. These simulations have many applications in designing multi layer... 

This paper presents a scalp electroencephalogram (EEG) rhythmic pattern detection scheme based on neural networks. Rhythmic discharges detection is applicable to the majority of seizures seen in newborns, and is listed as detecting 90% of all the seizures. In this approach some features based on various methods are extracted and compared by a modified multilayer neural network in order to find rhythmic discharges. Statistical performance comparison with seizure detection schemes of Gotman et al. and Liu et al. is performed. © 2006 IEEE  

In this work, we have studied thermal stability of nanoscale Ag metallization and its contact with CoSi 2 in heat-treated Ag(50 nm)/W(10 nm)/Co(10 nm)/Si(1 0 0) multilayer fabricated by sputtering method. To evaluate thermal stability of the systems, heat-treatment was performed from 300 to 900 °C in an N 2 ambient for 30 min. All the samples were analyzed by four-point-probe sheet resistance measurement (R s ), Rutherford backscattering spectrometry (RBS), X-ray diffractometry (XRD), and atomic force microscopy (AFM). Based on our data analysis, no interdiffiusion, phase formation, and R s variation was observed up to 500 °C in which the Ag layer showed a (1 1 1) preferred crystallographic... 

A mathematical model for cold rolling of two-layer strip is proposed using the upper bound theorem to investigate the plastic deformation behavior of the strip at the roll gap. The effects of total thickness reduction on the rolling power, rolling force, thickness reductions of each layer, and bonding strength of bimetal strip are discussed. Furthermore, experiments on two-layer strip rolling are conducted by employing aluminum-tin alloy and mild steel as the layers of the two-layer strips. It is found that all of theoretical predictions are in good agreement with the experimental measurements. Through the study, it becomes clear that the proposed analytical model is applicable for... 

This paper concerns the design of a neural state observer for nonlinear dynamic systems with noisy measurement channels and in the presence of small model errors. The proposed observer consists of three feedforward neural parts, two of which are MLP universal approximators, which are being trained off-line and the last one being a Linearly Parameterized Neural Network (LPNN), which is being updated on-line. The off-line trained parts are able to generate state estimations instantly and almost accurately, if there are not catastrophic errors in the mathematical model used. The contribution of the on-line adapting part is to compensate the remainder estimation error due to uncertain parameters... 

A novel output feedback neural controller is presented for the implementation of sliding mode control of DC/DC converters. The controller, which consists of a multilayer perceptron, has been trained so as to be robust for large variations of system parameters and state variables. Fast dynamic behavior is the other main advantage of the proposed controller, which allows realization of all the beneficial features of sliding mode control technique. Other advantages of the controller are simplicity and low cost. Computer simulations are carried out to investigate the effectiveness of the controller in voltage regulation for a relatively complex topology such as a Cuk converter. Simulation... 

The cloud manufacturing (C-Manufacturing) paradigm, as an advanced form of networked manufacturing, has recently been proposed based on a combination of existing manufacturing systems and emerging technologies, such as cloud computing, virtual manufacturing, agile manufacturing, manufacturing grid, Internet-of-things (IOT), and service-oriented technologies. In this study, through investigating the main goals of C-Manufacturing and today's hyper-competitive global marketplace circumstances, a prospective conceptual model called cloud-based global supply chain (CBGSC) has been developed which can overcome or mitigate the issues and risks associated with supply chain processes on a global... 

The electrocodeposition of Ni-TiO2 nanocomposite single layers and Ni-TiO2/TiO2 multilayers from a Watts bath containing a TiO2 sol on copper substrates was investigated by different deposition techniques. Compared with direct current (DC) deposition, both pulse plating (PP) and pulse reverse plating (PRP) facilitated higher incorporations of TiO2 nanoparticles. Morphological studies conducted by scanning electron microscopy and field emission scanning electron microscopy revealed that the microstructure of the Ni-TiO2 nanocomposite coatings are affected both by pulse potentials and durations, indicating that higher incorporations of TiO2 nanoparticles refine the microstructure. The results... 

In the present work, the hot corrosion behavior of two types of multilayer plasma sprayed TBC were investigated and compared with functionally graded and conventional TBCs. These kinds of multilayer coatings consisted of nano/μ alumina as a top coat on YSZ layer, a metallic bond coat and a functionally graded intermediate layer deposited between YSZ and bond coat layers. All the layers were sprayed on the Ni-base super alloy substrate. The hot corrosion resistance of the plasma sprayed coatings was examined at 1050 °C for 40 h, using a fused mixture of 45 wt% Na2SO4 + 55wt%V2O5. Before and after hot corrosion, the microstructure and phase analysis of the coating were studied using scanning... 

Molecular dynamics simulations are employed to investigate material properties of nanocrystalline aluminum and nanocrystalline Al/Ni multilayers at low temperature. For this purpose, both single crystal and nanocrystalline multilayers with different grain sizes and grain morphology are used as the substrate. The results of the simulations show that hardness and elastic modulus decrease with refinement of grain size in nanocrystalline aluminum and refinement of grain size and layer thickness in nanocrystalline Al/Ni multilayers, regardless of grain morphology. Furthermore, the angle between two adjacent grains, which is directly connected to the grain boundary thickness, has a great influence... 

In the present work, spark plasma sintering (SPS) process was employed to prepare a nanostructured yttria-stabilized zirconia (8YSZ) coating on a nickel-based superalloy (INCONEL 738) with functionally graded structure. A stack layer of INCONEL 738/NiCrAlY powder/Al foil/NiCrAlY+YSZ powder/YSZ powder was SPSed in a graphite die at an applied pressure of 40 MPa under an vacuum atmosphere (8 Pa). The sintering temperature was ∼1040 °C. For comparison purpose, the air plasma spray (APS) technique was employed to prepare the thermal barrier coating (TBC). Microstructural studies by scanning electron microscopy showed that the SPSed coating was sound and free of interfacial cracks and large... 

In the present paper, neutron spectrum is reconstructed using the Artificial Neural Network (ANN) and Modified Least Square (MLSQR) methods. The detector's response (pulse height distribution) as a required data for unfolding of energy spectrum is calculated using the developed MCNPX-ESUT computational code (MCNPX-Energy engineering of Sharif University of Technology). Unlike the usual methods that apply inversion procedures to unfold the energy spectrum from the Fredholm integral equation, the MLSQR method uses the direct procedure. Since liquid organic scintillators like NE-213 are well suited and routinely used for spectrometry of neutron sources, the neutron pulse height distribution is... 

Zn/Ni nanocrystalline multilayer coatings were electrodeposited using single bath method and switching current densities. Effect of deposition current densities (i1 and i2) and number of layers (n) on composition, surface morphology and roughness, microhardness, phase structure and corrosion resistance of Zn/Ni multilayers were studied and compared with that of single layer. Analyzing and optimizing the influences of mentioned parameters on corrosion resistance of multilayers carried out through Response Surface Methodology. The model based on RSM results demonstrated that improvement in corrosion resistance due to increase in “difference of deposition current densities” was more effective... 

The present work developed a mixed-mode cohesive zone model (CZM) with a mode I failure criterion to predict the delamination bending loads of multilayer, composite printed circuit boards (PCBs) assembled with soldered ball grid array (BGA) components that were reinforced with an underfill epoxy adhesive. Two different delamination modes were observed in these microelectronic assemblies: delamination at the interface between the solder mask and the first conducting layer of the PCB, and PCB subsurface delamination at the interface between the epoxy and glass fibers of one of the prepreg layers. The cohesive parameters for each of the two crack paths were obtained from fracture tests of... 

The present work developed a mixed-mode cohesive zone model (CZM) with a mode I failure criterion to predict the delamination bending loads of multilayer, composite printed circuit boards (PCBs) assembled with soldered ball grid array (BGA) components that were reinforced with an underfill epoxy adhesive. Two different delamination modes were observed in these microelectronic assemblies: delamination at the interface between the solder mask and the first conducting layer of the PCB, and PCB subsurface delamination at the interface between the epoxy and glass fibers of one of the prepreg layers. The cohesive parameters for each of the two crack paths were obtained from fracture tests of... 

Analytical and molecular dynamics simulation approaches are used in this paper to study free-vibration behavior of multilayer graphene-based nanoresonators considering interlayer shear effect. According to experimental observations, the weak interlayer van der Waals interaction cannot maintain the integrity of carbon atoms in the adjacent layers. Hence, it is vital that the interlayer shear effect is taken into account to design and analyze multilayer graphene-based nanoresonators. The differential equation of motion and the general form of boundary conditions are first derived for multilayer graphene sheets with rectangular shape using the Hamilton’s principle. Then, by pursuing an... 

The application of artificial neural network to compressor performance map prediction is investigated. Different types of artificial neural networks such as general regression neural network, rotated general regression neural network proposed by the authors, radial basis function network, and multilayer perceptron network are considered. Two different models are utilized in simulating the performance map. The results indicate that while the rotated general regression neural network has the least mean error and best agreement to the experimental data; it is however, limited to interpolation application. On the other hand, if one considers a tool for interpolation as well as extrapolation... 

One of the most important challenges in target parameters estimation via model-free methods is selection of the most effective input parameters namely features selection (FS). Indeed, irrelevant features can degrade the estimation performance. In the current study, the challenge of choosing among the several plant parameters is tackled by means of the innovative FS algorithm named ranking of features with minimum deviation from the target parameter (RFMD). The selected features accompanied with the stable and the fast learning algorithm of multilayer perceptron (MLP) neural network (i.e. Levenberg-Marquardt algorithm) which is a combination of gradient descent and Gauss-newton learning...