Analysis of Fabrication Errors and Study of Calibration Method for Multilevel Diffractive Optical Element

ＡｎａｌｙｓｉｓｏｆＦａｂｒｉｃａｔｉｏｎＥｒｏｒｓａｎｄＳｔｕｄｙｏｆＣａｌｉｂｒａｔｉｏｎＭｅｔｈｏｄｆｏｒＭｕｌｔｉｌｅｖｅｌＤｉｆｒａｃｔｉｖｅＯｐｔｉｃａｌＥｌｅｍｅｎｔＦＵＹｏｎｇｑｉ（ＣｈａｎｇｃｈｕｎＩｎｓｔｉｔｕｔｅｏｆＯｐｔｉｃｓ...     

ON THE ACCURACY OF THE LEAST SQUARES AND THE TOTAL LEAST SQUARES METHODS

Consider solving an overdetermined system of linear algebraic equations by both the least squares method (LS) and the total least squares method (TLS). Extensive published computational evidence shows that when the original system is consistent. one often obtains more accurate solutions by using the TLS method rather than the LS method. These numerical observations contrast with existing analytic perturbation theories for the LS and TLS methods which show that the upper bounds for the LS solution are always smaller than the corresponding upper bounds for the TLS solutions. In this paper we derive a new upper bound for the TLS solution and indicate when the TLS method can be more accurate than the LS method.Many applied problems in signal processing lead to overdetermined systems of linear equations where the matrix and right hand side are determined by the experimental observations (usually in the form of a lime series). It often happens that as the number of columns of the matrix becomes larger, the ra     

Accuracy of Computed Eigenvectors Via Optimizing a Rayleigh Quotient

This paper establishes converses to the well-known result: for any vector such that the sine of the angle sin(u, )=O(), we have

Sequential Identification of Linear Dynamic Systems with Memory

This paper presents a sequential estimator for some unknown parameters in stochastic linear systems with memory. As examples stochastic differential equations with time delayed drift are considered. Based on the maximum likelihood method, we construct an estimation procedure with given accuracy in the sense of the L _p -norm (p 2). It is shown, that this procedure works also in certain cases, when the normalized information matrix of the observed process is asymptotically degenerated. The almost surely consistency of the proposed estimators and the asymptotic behavior of the length of observations are derived.     

A numerical study of bifurcation in laminar flow in curved ducts

The bifurcation phenomenon whereby multiple-vortex secondary flow occurs in place of the normal two-vortex flow in laminar flow in curved ducts has previously been studied numerically by several researchers. However, the various results have been conflicting on many points. The present paper describes a set of numerical experiments conducted to study the effect of numerical accuracy on the solution. The results show that the transition from two- to four-vortex structure depends strongly on the differencing scheme and to a lesser extent on the grid size. The study also shows that as the Reynolds number of the flow increases, a two-vortex structure is re-established via a path which involves strongly asymmetric secondary flow patterns. These results are in agreement, at least qualitatively, with recent experimental theoretical and numerical results.     

The second kind Chebyshev–Newton–Cotes quadrature rule (open type) and its numerical improvement

The weighted Newton–Cotes quadrature rules of open type are denoted by

where w(x) is a positive function and is the step size. Various cases can be selected for the weight function of the above formula. In this paper, we consider as the main weight function and study the general formula:

The precision degree of the above formula is n + 1 for even n’s and is n for odd n’s but if one considers its upper and lower bounds as two additional variables, a nonlinear system will be derived whose solution improves the precision degree of above formula up to degree n + 2 numerically. In this way, some examples are given to show the numerical superiority of our idea.     

Calculation of turbulent combustion of propane in furnaces

An evaluation of some numerical methods for turbulent reacting flows in furnace-like geometries is carried out. The Reynolds averaged Navier–Stokes equations and the two-equation k–? model together with either finite-rate or infinite-rate reaction models are solved numerically. Either single- or multiple-step reactions together with the ‘eddy dissipation concept’ (EDC) are used to model reacting flows with finite reaction rates. The numerical scheme is finite difference based, together with a multi-grid method and a local grid refinement technique. These methods have been used to calculate the combustion of propane in a single- and multiple-burner configurations. In the former case, the sensitivity of the solution to variations in some model parameters (determining the reaction rate) and numerical parameters (mesh spacing) has been studied. It is noted that different dependent variables exhibit different levels of sensitivity to the variation in model parameters. Thus, calibration and validation of models for reacting flows require that one compares the most sensitive variables. For engineering purposes, on the other hand, one may calibrate and validate models with respect to the most relevant variables. Our conclusion is that since sensitivity of the temperature distribution is relatively mild, one can still use EDC-like methods in engineering applications where details of the temperature field are of minor importance.     

Wave parameter identification problem for ocean test structure data,part 1, continuous formulation

This paper deals with the solution of the wave parameter identification problem for ocean test structure data. A continuous formulation is assumed. An ocean test structure is considered, and wave elevation and velocities are assumed to be measured with a number of sensors. Within the frame of linear wave theory, a Fourier series model is chosen for the wave elevation and velocities. Then, the following problem is posed: Find the amplitudes of the various wave components of specified frequency and direction, so that the assumed model of wave elevation and velocities provides the best fit to the measured data. Here, the term best fit is employed in the least-square sense over a given time interval.At each time instant, the wave representation involves three indexes (frequency, direction, instrument); hence, three-dimensional arrays are required. This formal difficulty can be avoided by switching to an alternative representation involving only two indexes (frequency-direction, instrument); hence, standard vector-matrix notation can be used. Within this frame, optimality conditions are derived for the amplitudes of the assumed wave model.Numerical results are presented. The effect of various system parameters (number of frequencies, number of directions, sampling time, number of sensors, and location of sensors) is investigated in connection with global or strong accuracy, local or weak accuracy, integral accuracy, and condition number of the system matrix.From the numerical experiments, it appears that the identification problem has a unique solution if the number of directions is smaller than or equal to the number of sensors; it has an infinite number of solutions otherwise. In the case where a unique solution exists, the condition number of the system matrix increases as the size of the system increases, and this has a detrimental effect on the accuracy. However, the accuracy can be improved by proper selection of the sampling time and by proper choice of the number and location of the sensors.This work was supported by Exxon Production Research Company, Houston, Texas. This paper is based partly on Ref. 1–4.     

Practical method validation: Validation sufficient for an analysis method

Validation of an analysis method depends on the purpose of the method, the chosen technique and the procedure in question. Methods are used for different research, product development, process control and quality control purposes. The human and economical importance of results vary. Each of the techniques used, such as chromatography-(HPLC, HRGC, TLC), capillary electrophoresis-(CE), spectrophotometry-(UV/VIS, IR, fluorescence, AAS, ICP) or spectrometric techniques (NMR, MS) as well as the hyphenated methods, have their own special features and deficiencies which must be considered. The method can include a simple pretreatment or it may include many demanding steps, it can use automation and data processing in various ways, it can have an official status, it can be a thoroughly verified or less studied one. How should these differences be accounted for during the validation? What would be a sufficient certainty that the method does what is expected, that the method fits for the purpose it was intended? The client (or authority) decides the required timetable, cost and quality level. This is why within a laboratory different quality levels and associated levels of validation exist. This paper tries to outline a practical test frame for validation efforts to assist the analyst when planning validation of a method.     

A comparison of integration and interpolation Eulerian-Lagrangian methods

Selected finite element Eulerian-Lagrangian methods for the solution of the transport equation are compared systematically in the relatively simple context of 1D, constant coefficient, conservative problems. A combination of formal analysis and numerical experimentation is used to characterize the stability and accuracy that results from alternative treatments of the concentrations at the feet of the characteristic lines. Within the methods analyzed, those that approach such treatment with the perspective of ‘integration’ rather than ‘interpolation’ tend to have superior accuracy. Exact integration leads to unconditional stability and excellent accuracy. Quadrature integration leads only to conditional stability, but newly derived criteria show that stability restrictions are relatively mild and should not preclude the usefulness of quadrature integration methods in a range of practical applications. While conclusions cannot be extended directly to multiple dimensions and complex flows and geometries, results should provide useful insight to the development and behaviour of specific Eulerian-Lagrangian transport models.