This paper presents an industrial case study examining the application of inferential estimation to a lubricant production plant. Three estimation algorithms are studied, First, an existing plant inference model is co...
详细信息
This paper presents an industrial case study examining the application of inferential estimation to a lubricant production plant. Three estimation algorithms are studied, First, an existing plant inference model is compared to a static estimator designed using information obtained from plant experiments. Next, the case for a dynamic and adaptive algorithm is presented, The procedures required to initialise the adaptive estimator are described, and implementation issues are discussed. This paper demonstrates that (provided such issues are carefully considered) an adaptive inferential measurement algorithm may be used to provide timely and accurate inferences.
The availability of satellite-derived rainfall products to hydrologists and other natural scientists has increased enormously in the last five years. The purpose of this paper is to review concisely the current state-...
详细信息
The availability of satellite-derived rainfall products to hydrologists and other natural scientists has increased enormously in the last five years. The purpose of this paper is to review concisely the current state-of-the-art of satellite precipitation estimation over land and the availability of standard products, and to highlight some of the strengths and limitations of satellite-derived rainfall data for hydrological applications. Methods based on infrared, visible and passive microwave radiation measurements are discussed. Results of several international activities aimed at evaluating the performance of the estimation algorithms are briefly summarized.
Using data from extensive vibrational tests of the new Saab 2000 aircraft, a combined method for vibration analysis is studied. The method is based on a realization algorihm followed by standard prediction error metho...
详细信息
Using data from extensive vibrational tests of the new Saab 2000 aircraft, a combined method for vibration analysis is studied. The method is based on a realization algorihm followed by standard prediction error methods (PEM). We find that the realization algorithm gives good initial model parameter estimates that can be further improved by the use of PEM. We use the method to get insights,into the vibrational eigenmodes. Copyright (C) 1996 Elsevier Science Ltd.
The volumetric efficiency (eta(v)) represents a measure of the effectiveness of an air pumping system, and is one of the most commonly used parameters in the characterization and control of four-stroke internal combus...
详细信息
The volumetric efficiency (eta(v)) represents a measure of the effectiveness of an air pumping system, and is one of the most commonly used parameters in the characterization and control of four-stroke internal combustion engines. Physical models of eta(v) require the knowledge of some quantities usually not available in normal operating conditions, Hence, a purely black-box approach is often used to determine the dependence of eta(v) upon the main engine variables, like the crankshaft speed and the intake manifold pressure. Various black-box approaches for the estimation of eta(v) are reviewed, from parametric (polynomial-type) models, to non-parametric and neural techniques, like additive models, radial basis function neural networks and multi-layer perceptrons. The benefits and limitations of these approaches are examined and compared. The problem considered here can be viewed as a realistic benchmark for different estimation techniques.
Consider the problem of loss function minimization when only (possibly noisy) measurements of the loss function are available. In particular, no measurements of the gradient of the loss function are assumed available ...
详细信息
Consider the problem of loss function minimization when only (possibly noisy) measurements of the loss function are available. In particular, no measurements of the gradient of the loss function are assumed available (as required in the steepest descent or Newton-Raphson algorithms). Stochastic approximation (SA) algorithms of the multivariate Kiefer-Wolfowitz (finite-difference) form have long been considered for such problems, but with only limited success. The simultaneous perturbation SA (SPSA) algorithm has successfully addressed one of the major shortcomings of those finite-difference SA algorithms by significantly reducing the number of measurements required in many multivariate problems of practical interest. This paper presents a second-order SPSA algorithm that is based on estimating both the loss function gradient and inverse Hessian matrix at each iteration. The aim of this approach is to emulate the acceleration properties associated with deterministic algorithms of Newton-Raphson form, particularly in the terminal phase where the first-order SPSA algorithm slows down in its convergence. This second-order SPSA algorithm requires only three loss function measurements at each iteration, independent of the problem dimension. This paper includes a formal convergence result for this second-order approach.
In an attempt to robustify the celebrated recursive least-squares algorithm, a H∞-norm bounded recursive least-squares algorithm for parameter estimation of linear regression models is presented in this paper. This a...
详细信息
In an attempt to robustify the celebrated recursive least-squares algorithm, a H∞-norm bounded recursive least-squares algorithm for parameter estimation of linear regression models is presented in this paper. This algorithm is a special case of the H∞ filter algorithm developed recently, and guarantees estimates with the smallest possible estimation error energy, over all possible modelling errors of fixed energy. Governed by a robust criterion function, it is shown that the proposed recursive algorithm makes cautious information updates leading to active and accelerated estimation. Connections with the ordinary recursive least-squares algorithm are pointed out. Given prior bounds on modelling errors, formulae are derived to compute ellipsoidal parameter error bounds thus providing for deterministic robust estimation.
A control strategy is developed in order to maximize the production of lipase for a process operating in fed batch mode. At first, an original method based on the mathematical properties of the model is proposed to de...
详细信息
A control strategy is developed in order to maximize the production of lipase for a process operating in fed batch mode. At first, an original method based on the mathematical properties of the model is proposed to define a closed loop control strategy. The strategy elaborated consists in regulating the substrate/biomass ratio at a given set point, which requires the design of a specific controller able to cope with the process non linearities. A linearizing controller is proposed. Then, to implement it on the process, an adaptive version of the controller is developped by coupling it with an estimation algorithm. The control system is tested and proved performant even under realistic conditions with noise on the output and disturbances on the kinetic parameters.
Teleoperation and intervention robotics involve more and more mobile or movable robots which must get new references to be at work. Two cases are presented, in which references are measured by contact, either with the...
详细信息
Teleoperation and intervention robotics involve more and more mobile or movable robots which must get new references to be at work. Two cases are presented, in which references are measured by contact, either with the environment (medical robot working at a fixed place with tactile references recording), or with other robots (cooperation with contacts like a “handshaking”), and information is processed in order to obtain transformation matrix between robots and/or environment coordinates for either work in a new place, or move an object with several robots together or one after another.
A unified statistical performance analysis using subspace perturbation expansions is applied to subspace-based algorithms for direction-of-arrival (DOA) estimation in the presence of sensor errors. In particular, the ...
详细信息
A unified statistical performance analysis using subspace perturbation expansions is applied to subspace-based algorithms for direction-of-arrival (DOA) estimation in the presence of sensor errors. In particular, the multiple signal classification (MUSIC), min-norm, state-space realization (TAM and DDA) and estimation of signal parameters via rotational invariance techniques (ESPRIT) algorithms are analyzed. This analysis assumes that only a finite amount of data is available. An analytical expression for the mean-squared error of the DOA estimates is developed for theoretical comparison in a simple and self-contained fashion. The tractable formulas provide insight into the algorithms. Simulation results verify the analysis.
Many statistical methods rely on numerical optimization to estimate a model's parameters. Unfortunately, conventional algorithms sometimes fail. Even when they do converge, there is no assurance that they have fou...
详细信息
Many statistical methods rely on numerical optimization to estimate a model's parameters. Unfortunately, conventional algorithms sometimes fail. Even when they do converge, there is no assurance that they have found the global, rather than a local, optimum. We test a new optimization algorithm, simulated annealing, on four econometric problems and compare it to three common conventional algorithms. Not only can simulated annealing find the global optimum, it is also less likely to fail on difficult functions because it is a very robust algorithm. The promise of simulated annealing is demonstrated on the four econometric problems.
暂无评论