摘要:

abstractThis paper discusses the robustness of a flexible ann nonlinear control law. Using a ftnite element description of the ann, we get a nonlinear model. The control of a set of state variables is obtained using a nonlinear decoupling method that is performed on the inverse dynamic model. We study the sensitivity of the result with respect to variations on the model parameters, and give some simulation results.

ISSN:0228-6203    

DOI:10.1080/02286203.1992.11760139

出版商:Taylor&Francis    

年代:1992

数据来源: Taylor

摘要:

abstractThe authors present an adaptive identification method (AIM) primarily for adaptive control systems with insufficient persistent excitation. In this paper, the matrix P(t-l) resetting, a new forgelling factorλ(1), remembering units, and a weighted filtering algorithm are proposed. The AIM technique, using a modified least squares algorithm, is presented in terms of the previously mentioned procedures. Theoretical analyses and simulation results demonstrate that the proposed methodology can overcome the bursting problems thaI sometimes exisl in adaptive control systems.

ISSN:0228-6203    

DOI:10.1080/02286203.1992.11760140

出版商:Taylor&Francis    

年代:1992

数据来源: Taylor

摘要:

abstractDuring the design of a VLSI system, some kind of simulation is of vital importance. Dynamic switch-level simulation is one of the most elegant techniques for evaluating the performence of the MOS circuits. This simulalOr is based on a linear switch–level model of MOS transistors, in which each transislOr is modeled as a voltage–controlled switch in series with a resislOr and an associated gate capacitance. It supports three levels of logic values–0,1 and X. Modular styles and unique data structures are utilized 10 implement the tree–walk algorithms based on charge sharing and final value computations. An easy–to–comprehend command language for the simulator is developed and provided to the user. The simulator supports both NMOS and CMOS technologies.

ISSN:0228-6203    

DOI:10.1080/02286203.1992.11760141

出版商:Taylor&Francis    

年代:1992

数据来源: Taylor

摘要:

abstractStochastic models for many processes that arise in the life, physical, and engineering sciences must be able to accommodate skew. Among the more frequently used models of this sort are the extreme-value, gamma, inverse Gaussian, lognormal, and WeibuJl densities. Goodness–of–fit test generally have little ability W distinguish between these competing models, and the decision to adopt one model to the exclusion of the others may hinge on the inferential analysis procedures available for the adopted model. Among the previously listed models, the inverse Gaussian density offers a variety of analyses that are comparable to those available for the normal dislribution, such as regression, t-tests, confidence intervals, and an alternative to the analysis of variance (ANOYA) for skewed data. This alternative to ANOYA is refeITed to as analysis of reciprocals (ANORE) and provides the topic for this paper.

ISSN:0228-6203    

DOI:10.1080/02286203.1992.11760142

出版商:Taylor&Francis    

年代:1992

数据来源: Taylor

摘要:

abstractA new multi-threshold Perceptron capable of handling both binary and analog input is presented and discussed. The modified Perceptron replaces the sigmoid function with sinusoidal function. A computer program has been developed to simulate behavior of a network utilizing the modified Perceptron. Both XOR and Parity Check problems were solved using a single-layer network utilizing this modified Percept ron. Based on the results obtained from the simulation. the modified Perceptron is capable of solving problems (such as XOR) that can not be solved using a single-layer of the classical Perceptron. Also. a network utilizing this modified Perceptron requires fewer number of iterations to converge to a solution than that of a multi–layer Perceptron network using back propagation.

ISSN:0228-6203    

DOI:10.1080/02286203.1992.11760143

出版商:Taylor&Francis    

年代:1992

数据来源: Taylor