Sharif Digital Repository

Hydrodynamic instabilities at the interface of stratified shear layers could occur in various modes and have an important role in the mixing process. In this work, the linear stability analysis in the temporal framework is used to study the stability characteristics of a particle-laden stratified two-layer flow for two different background density profiles: smooth (hyperbolic tangent) and piecewise linear. The effect of parameters, such as bed slope, viscosity, and particle size, on the stability is also considered. The pseudospectral collocation method employing Chebyshev polynomials is used to solve two coupled eigenvalue equations. Based on the results, there are some differences in the... 

This paper presents a stabilizing controller design method for quasi resonant (QR) converters described by a class of piecewise linear (PWL) models. The generalized state-space averaging technique (GSSA) is applied for the modeling and analysis of the half-wave zero current switching quasi-resonant (HW-ZCS-QR) buck converter. The nonlinear GSSA model of the converter is reconstructed using a piecewise linearizing technique. Subsequently, the piecewise linear models are combined together, to form a unified model, using a fuzzy modeling approach. The stability of the applied method has been investigated using Lyapunov method. Finally, a linear Hinfty; controller synthesis method is applied to... 

In this paper a multi-scale image enhancement strategy using wavelets as applied to digital mammograms is presented. For multiresolution wavelet analysis redundant dyadic discrete wavelet transform is utilized to allow translation invariance and low resolution enhancement capability. Two alternative nonlinear gain adjustments, piecewise linear and Gaussian form of gain adjustment are used for coefficient modification for the enhancement of subtle details such as microcalcification and low resolution edges of different lesion types. The results of comparing these methods of gain adjustment are presented. This comparing has done by defining new parameters for measuring quality of image based... 

Large signal behavior of the Laser Diode is analyzed with the aim of finding a qualitative relationship between modulation amplitude and the change in the shape of its transfer function. It is found that, at quasi-static conditions, increasing the amplitude of the modulating signal smoothes out the original piecewise-linear transfer function of the diode. It further shifts the now smoothed knee point to lower DC currents. As the frequency increases, however, these effects become less and less pronounced. In addition, the transfer function becomes rather wavy due to the failure of current density to reach the saturation level necessary for adequate stimulated emission. At even higher... 

A composite beam with single delamination traveled by a constant amplitude moving force is modeled accounting for the Poisson's effect, shear deformation and rotary inertia. The mechanical behavior between the delaminated surfaces is modeled using a piecewise-linear spring foundation. The governing differential equations of motion for such system are derived. Primarily, eigen-solution technique is used to obtain the natural frequencies and their corresponding mode shapes of such beam. Then, the Ritz method is employed to derive the dynamic response of the beam due to the moving force. The obtained results for the free and forced vibrations of beams are verified against reported similar... 

We study the problem of approximating a function F:ℝ → ℝ by a piecewise-linear function F̄ when the values of F at {x 1,...,xn} are given by a discrete probability distribution. Thus, for each xi we are given a discrete set y i,1,..., yi,mi of possible function values with associated probabilities pi,j such that Pr[F(xi) = yi,j] = pi,j. We define the error of F̄ as error(F, F̄) = maxi=1n E[|Fxi) - F̄(xi)|]. Let m = ∑i=1nmi be the total number of potential values over all F(xi). We obtain the following two results: (i) an O(m) algorithm that, given F and a maximum error ε, computes a function F̄ with the minimum number of links such that error(F, F̄) ≤ ε; (ii) an O(n4/3+δ + mlogn) algorithm... 

This paper considers the detector design and performance analysis in a widely separated multiple input - multiple output (S-MIMO) radar. We studied this subject under some assumptions such as single point target, Swerling models for radar cross section and Gaussian interference. The target is moving in piecewise linear trajectory and multi-scan data is stored. So, the target cell maybe changed through the time. A GLRT detector is derived that employs all data at once in a central unit. The performance of detector is extracted and numerical simulations verify all results  

The design and implementation of the hybrid control method for a three-pole active magnetic bearing (AMB) is proposed in this paper. The system is inherently nonlinear and conventional nonlinear controllers are a little complicated while the proposed hybrid controller has a piecewise linear form, i.e., linear in each sub-region. A state-feedback hybrid controller is designed in this study and the unmeasurable states are estimated by an observer. The gains of the hybrid controller are obtained by the LQR method in each sub-region. To evaluate the performance, the designed controller is implemented on an experimental setup. The experimental results show that the proposed method can efficiently... 

The approximation power of general feedforward neural networks with piecewise linear activation functions is investigated. First, lower bounds on the size of a network are established in terms of the approximation error and network depth and width. These bounds improve upon state- of-the-art bounds for certain classes of functions, such as strongly convex functions. Second, an upper bound is established on the difference of two neural networks with identical weights but different activation functions. © The Author(s) 2018  

We consider a class of continuous-Time hybrid dynamical systems that correspond to subgradient flows of a piecewise linear and convex potential function with finitely many pieces, and which includes the fluid-level dynamics of the Max-Weight scheduling policy as a special case. We study the effect of an external disturbance/perturbation on the state trajectory, and establish that the magnitude of this effect can be bounded by a constant multiple of the integral of the perturbation. © 1963-2012 IEEE  

In association with multi-inhomogeneity problems, a special class of eigenstrains is discovered to give rise to disturbance stresses of interesting nature. Some previously unnoticed properties of Eshelby's tensors prove useful in this accomplishment. Consider the set of nested similar ellipsoidal domains {Ω1, Ω2,⋯,ΩN+1}, which are embedded in an infinite isotropic medium. Suppose that Ωt= {x|x ∈ ℝ3, ∑p=13 x p2/ap2 ≤ ξt 2}, which 0 ≤ ξ1 < ξ2 < ⋯ < ξN+1 and ξ t a p , p=1,2,3 are the principal half axes of Ω t . Suppose, the distribution of eigenstrain, ε ij *(x) over the regions Γ t =Ω t+1-Ω t , t=1,2,⋯,N can be expressed as εij*(x)={f ijkl⋯mt(∑p=13 x p2/ap2)xkx l⋯xm, x∈ Γ, 0, x ∈ Ω1∪(ℝ3 -... 

Due to shear layer at the interface of density current and ambient fluid, density current disturbs and entrains the surrounding fluid. Most existing analytical and numerical models for density current flows are based on the equations for single-phase flows. In this research, the density current has been modeled with two-phase flow model. The governing equations are continuity, x- momentum, and y- momentum equations for every fluid. The volume-of-fluid (VOF) interface tracking technique which uses a piecewise-linear interface calculation (PLIC) in each cell is used to determine the deformation of free surface in density current, numerically. Surface tension is implemented by the continuous... 

This paper presents a digital hardware implementation of a frequency adaptive Hopf oscillator along with investigation on systematic behavior when they are coupled in a population. The mathematical models of the oscillator are introduced and compared in sense of dynamical behavior by using system-level simulations based on which a piecewise-linear model is developed. It is shown that the model is capable to be implemented digitally with high efficiency. Behavior of the oscillators in different network structures to be used for dynamic Fourier analysis is studied and a structure with more precise operation which is also more efficient for FPGA-based implementation is implemented. Conceptual... 

Facial expression recognition is a challenging and interesting problem in computer vision and pattern recognition. Geometric variability in both emotion expression and neutral face is a fundamental challenge in facial expression recognition problem. This variability not only directly affects geometric facial expression recognition methods, but also is a critical problem in appearance methods. To overcome this problem, this paper presents an approach which eliminates geometric variability in emotion expression; thus, appearance features can be accurately used for facial expression recognition. Therefore, a fixed geometric model is used for geometric normalization of facial images. This model... 

There has been a strong push recently to examine biological scale simulations of neuromorphic algorithms to achieve stronger inference capabilities. This paper presents a set of piecewise linear spiking neuron models, which can reproduce different behaviors, similar to the biological neuron, both for a single neuron as well as a network of neurons. The proposed models are investigated, in terms of digital implementation feasibility and costs, targeting large scale hardware implementation. Hardware synthesis and physical implementations on FPGA show that the proposed models can produce precise neural behaviors with higher performance and considerably lower implementation costs compared with... 

The state-space averaging applied to switched networks generally results in nonlinear systems. It is common to perform a small signal linearization about an operating point to obtain a linear system. When the variations in signals are large, e.g., in PFC rectifiers, the small signal approximation produces results that are susceptible to instability problems. In this paper a class of piecewise linear models merged by fuzzy system are introduced for PWM converters. The necessary and sufficient and conditions for stability of fuzzy models using fuzzy state-feedback controllers are given. The results obtained are illustrated with a buck-boost converter. The simulation and experimental results... 

This study considers the detector design and performance analysis in a statistical multiple-input-multiple-output radar. The authors studied this subject under some assumptions such as single point target, Swerling scattering models for radar cross-section and Gaussian interference. The target is assumed to move in a piecewise linear trajectory and multi-scan data is stored. So, the target cell may have changed during the time. A three-dimensional data signal space is generated that involved in observations from multiple scans, then a generalised likelihood ratio test detector is derived that employs consecutive observations in a central unit. Then a sub-optimum detector is introduced in... 

Tactile sense is a key element in developing virtual reality simulators or surgical training systems. In this regard, haptic interfaces, the generator of a sense of touch, play a significant role in producing a realistic haptic feedback force. Since the majority of practical control theories are model-based, the identification of the robot dynamic model is a process of high importance and application. The main concern, accordingly, is to find a precise dynamic model for the aforementioned user interfaces. In this research, we have implemented the Lion identification method to characterize the dynamics of a parallel haptic device in actuating a surgery simulation. We chose the Novint Falcon... 

Large-scale simulation of spiking neural networks on hardware with a remarkable resemblance to their mathematical models is a key objective of the neuromorphic discipline. This issue is, however, considerably resource-intensive due to the presence of nonlinear terms in neuron models. This paper proposes a novel uniform piecewise linear segmentation approach for nonlinear function evaluations. Employing the proposed approach, we present a uniform-segmented adaptive exponential neuron model capable of accurately producing various responses exhibited by the original model and suitable for efficient large-scale implementation. In contrast to previous nonuniform-segmented neuron models, the... 

There are many depletable resource planning problems associated with convex cost function. The Greedy algorithm has been used to find the optimal solution for such problems. But exploiting that algorithm requires a large bulk of computations to find the optimal solution. Thus, in this paper we have established two piecewise linear estimators for the convex cost function: an inner intersection of the function values and an outer tangent. A practical experiment was conducted to compare the performance of the two presented estimators and it was observed that the outer tangent performs more efficiently than the inner intersection. The execution time of the Greedy Algorithm and the estimation...