The existing structural rating system in the maintenance steering group-3 is facing difficulties with developing new aircraft. There is always a lack of available data to help make proper inspection intervals. By taki...
详细信息
The existing structural rating system in the maintenance steering group-3 is facing difficulties with developing new aircraft. There is always a lack of available data to help make proper inspection intervals. By taking advantage of the powerful learning ability and data fusion capability of the backpropagation network, this paper aims to develop a new decision support system for the determination of scheduled maintenance inspection intervals for aircraft structures susceptible to accidental damage. As long as various data are available, regardless of engineering experience, experimental values, and in-service reliability data feedback, the backpropagation network has been proven able to make acceptable predictions after training, making accidental damage rating a dynamic and flexible procedure.
Factor geometric dilution of precision (GDOP) is an indicator that shows the quality of GPS positioning and has often been used for choosing suitable satellite's subset from at least 24 orbited existing satellites...
详细信息
Factor geometric dilution of precision (GDOP) is an indicator that shows the quality of GPS positioning and has often been used for choosing suitable satellite's subset from at least 24 orbited existing satellites. The calculation of GPS GDOP is a time-consuming task which can be done by solving measurement equations with complicated matrix transformation and inversion. In order to decrease this computational burden, in this research the artificial neural network (ANN) has been used. Although the basic back propagation (BP) is the most popular ANN algorithm and can be used in the estimators, detectors or classifiers, it is too slow for practical problems and its performance is not satisfactory in many cases. To overcome this problem, six algorithms, namely, BP with adaptive learning rate and momentum, Fletcher-Reeves conjugate gradient algorithm (CGA), Polak-Ribikre CGA, Powell-Beale CGA, scaled CGA, and resilient BP have been proposed to reduce the convergence time of the basic BP. The simulation results show that resilient BP, compared with other methods, has greater accuracy and calculation time. The resilient BP can improve the classification accuracy from 93.16 to 98.02 % accuracy by using the GPS GDOP measurement data.
Accurate wavelet estimation is absolutely critical to the success of any seismic inversion. The inferred shape of seismic wavelet may strongly influence the seismic inversion result and, thus, subsequent assessments o...
详细信息
ISBN:
(纸本)9781479921508
Accurate wavelet estimation is absolutely critical to the success of any seismic inversion. The inferred shape of seismic wavelet may strongly influence the seismic inversion result and, thus, subsequent assessments of the reservoir quality. Wavelet amplitude is a key factor in the process of wavelet estimation. So, this paper proposes a method with wavelet energy balancing strategy of impedance inversion. However, inversion problems are generally ill-posed. Regularization methods are essential for solving such problems, this paper we adopt regularization method to solute its intrinsic ill-posed problem. Meanwhile, we make improvements to conjugategradient search direction on the basis of normal conjugate gradient algorithm. Theoretic simulations are made and the field data are applied, it is proved that this method has advantage of high precision, robustness, and wide range of applications.
Most geometric computer vision problems involve orthogonality constraints. An important subclass of these problems is subspace estimation, which can be equivalently formulated into an optimization problem on Grassmann...
详细信息
Most geometric computer vision problems involve orthogonality constraints. An important subclass of these problems is subspace estimation, which can be equivalently formulated into an optimization problem on Grassmann manifolds. In this paper, we propose to use the conjugate gradient algorithm on Grassmann manifolds for robust subspace estimation in conjunction with the recently introduced generalized projection based M-Estimator (gpbM). The gpbM method is an elemental subset-based robust estimation algorithm that can process heteroscedastic data without any user intervention. We show that by optimizing the orthogonal parameter matrix on Grassmann manifolds, the performance of the gpbM algorithm improves significantly. Results on synthetic and real data are presented. (c) 2011 Elsevier B.V. All rights reserved.
In order to take advantage of the attractive features of Polak-RibiSre-Polyak and Fletcher-Reeves conjugategradient methods, two hybridizations of these methods are suggested, using a quadratic relaxation of a hybrid...
详细信息
In order to take advantage of the attractive features of Polak-RibiSre-Polyak and Fletcher-Reeves conjugategradient methods, two hybridizations of these methods are suggested, using a quadratic relaxation of a hybrid conjugategradient parameter proposed by Gilbert and Nocedal. In the suggested methods, the hybridization parameter is computed based on a conjugacy condition. Under proper conditions, it is shown that the proposed methods are globally convergent for general objective functions. Numerical results are reported;they demonstrate the efficiency of one of the proposed methods in the sense of the performance profile introduced by Dolan and Mor,.
We make a analysis to the surface-consistent model, and use the conjugategradient method to decompose the seismic record in time domain in terms of the surfaceconsistent precondition and the model. Application to rea...
详细信息
We make a analysis to the surface-consistent model, and use the conjugategradient method to decompose the seismic record in time domain in terms of the surfaceconsistent precondition and the model. Application to real and typical seismic data chosen from areas with complex surface conditions shows that the effect of near-surface condition on seismic records can be effectively eliminated by means of the presented method. This method has advantages of faster processing, the ability of robust to noise, and obtaining remarkable results when seismic data are processed in areas of complex surface conditions and low signal-to-noise ratio.
We deduce the finite-element acoustic wave equation in the frequency domain. In order to eliminate boundary reflection, the absorbing boundary conditions of Clayton-Engquist paraxial wave equation are introduced to th...
详细信息
We deduce the finite-element acoustic wave equation in the frequency domain. In order to eliminate boundary reflection, the absorbing boundary conditions of Clayton-Engquist paraxial wave equation are introduced to the frequency domain. The finite-element stiffness matrix and mass matrix are compressed for storage. The solutions of forward modeling are obtained by using the generalized conjugate gradient algorithm. On these bases, we deduce the Jacobi matrix representing the relation between the wavefield data residuals δ Ů and the element material-property adjusted values δ λ at a certain frequency. Using the differences δ Ů between the surface 2D shot recorded data and theoretical modelling data, the adjusted values δ λ can be obtained iteratively. Because of the limitation of the computer's storage, larger number of unknowns is not permitted. The measure of compressing and assembling the element Jacobi matrix coefficients in the same medium is also proposed. By using this method, the unknowns' number of inversion is reduced. Combined with the conjugate gradient algorithm, only a few frequencies in valid-wave domain are needed in this inversion. Some numerical examples of the modeling and inversion are given. The effectivity of the method is proved using the results.
This paper investigates a quantum neural network and discusses its application in control systems. A learning-type neural network-based controller that uses a multi-layer quantum neural network having qubit neurons as...
详细信息
This paper investigates a quantum neural network and discusses its application in control systems. A learning-type neural network-based controller that uses a multi-layer quantum neural network having qubit neurons as its information processing unit is proposed. Three learning algorithms;a back-propagation algorithm, a conjugate gradient algorithm and a real-coded genetic algorithm, are investigated to supervise the training of the multi-layer quantum neural network. To evaluate the learning performance and the capability of the quantum neural network-based controller, we conducted computational experiments for controlling a nonlinear discrete-time plant and a nonholonomic system - in this study a two-wheeled robot. The results of computational experiments confirm both the feasibility and the effectiveness of the quantum neural network-based controller and that the real-coded genetic algorithm is suitable for the learning method of the quantum neural network-based controller.
AEROELASTIC effects induced by the interaction of unsteady aerodynamics and structural dynamics are an extremely important concern in aircraft design. Among the aeroelastic effects, flutters and limit-cycle oscillatio...
详细信息
AEROELASTIC effects induced by the interaction of unsteady aerodynamics and structural dynamics are an extremely important concern in aircraft design. Among the aeroelastic effects, flutters and limit-cycle oscillations (LCO) are the critical constraints on the performance of an aircraft. Although a possible flutter or LCO can be suppressed via many conventional controls, such as classical single-input/single-output (SISO) feedback controls [1] and linear quadratic Gaussian (LQG) control [2], it is not easy to synthesize an efficient control scheme to suppress an aeroelastic instability because of parameter uncertainties caused by inaccurate computation of the aerodynamics and/or structural modes, as well as varying configurations. Therefore, it is necessary to develop an adaptive control to suppress the instability of an aeroelastic system with parameter uncertainties.
The Multidisciplinary Design Analysis and Optimization (MDAO) community has developed a multitude of algorithms and techniques, called architectures, for performing optimizations on complex engineering systems which i...
详细信息
ISBN:
(数字)9781600869372
ISBN:
(纸本)9781600869372
The Multidisciplinary Design Analysis and Optimization (MDAO) community has developed a multitude of algorithms and techniques, called architectures, for performing optimizations on complex engineering systems which involve coupling between multiple discipline analyses. These architectures seek to effciently handle optimizations with computationally expensive analyses including multiple disciplines. We propose a new testing procedure that can provide a quantitative and qualitative means of comparison among architectures. The proposed test procedure is implemented within the open source framework, OpenMDAO, and comparative results are presented for five well-known architectures: MDF, IDF, CO, BLISS, and BLISS-2000. We also demonstrate how using open source soft- ware development methods can allow the MDAO community to submit new problems and architectures to keep the test suite relevant.
暂无评论