Adaptation Of Model Parameters Research Articles

Stable indirect adaptive control based on discrete-time T–S fuzzy model

This paper presents an indirect adaptive fuzzy control scheme for uncertain nonlinear asymptotically stable plants. A discrete-time T–S fuzzy input–output model is employed to approximate the unknown plant dynamics. The T–S fuzzy model is fed with its own states, which are indeed its past outputs, rather than the measurements from the plants. Entirely based on this model, a feedback linearization control law is designed by using the model structure, model parameters and model states and then applied to control the model and the plant. Premise and consequent parameters of the model are updated on-line by gradient descent algorithm and recursive least square estimation method using plant output measurements. Stability analysis shows that if there the model structure is accurate the adaptive controller achieves the bounded tracking error and boundedness of all the closed-loop signals. In the ideal case where the signals are persistently exciting during the model parameter adaptation a perfect asymptotic tracking is ensured. The effectiveness of the method is verified by simulation application to SISO and MIMO example plants.

Environmental Independent ASR Model Adaptation/Compensation by Bayesian Parametric Representation

The mismatch between system training and operating conditions can seriously deteriorate the performance of automatic speech recognition (ASR) systems. Various techniques have been proposed to solve this problem in a specified speech environment. Employment of these techniques often involves modification on the ASR system structure. In this paper, we propose an environment-independent (EI) ASR model parameter adaptation approach based on Bayesian parametric representation (BPR), which is able to adapt ASR models to new environments without changing the structure of an ASR system. The parameter set of BPR is optimized by a maximum joint likelihood criterion which is consistent with that of the hidden Markov model (HMM)-based ASR model through an independent expectation-maximization (EM) procedure. Variations of the proposed approach are investigated in the experiments designed in two different speech environments: one is the noisy environment provided by the AURORA 2 database, and the other is the network environment provided by the NTIMIT database. Performances of the proposed EI ASR model compensation approach are compared to those of the cepstral mean normalization (CMN) approach, which is one of the standard techniques for additive noise compensation. The experimental results show that performances of ASR models in different speech environments are significantly improved after being adapted by the proposed BPR model compensation approach

A Study of Minimum Classification Error (MCE) Linear Regression for Supervised Adaptation of MCE-Trained Continuous-Density Hidden Markov Models

In this paper, we present a formulation of minimum classification error linear regression (MCELR) for the adaptation of Gaussian mixture continuous-density hidden Markov model (CDHMM) parameters. Two optimization approaches, namely generalized probabilistic descent (GPD) and Quickprop are studied and compared for the optimization of the MCELR objective function. The effectiveness of the proposed MCELR technique is confirmed via a series of supervised speaker adaptation experiments on a task of continuous Putonghua (Mandarin Chinese) speech recognition

Development of a dynamic model for cleaning ultra filtration membranes fouled by surface water

In this paper, a dynamic model for cleaning ultra filtration membranes fouled by surface water is proposed. A model that captures the dynamics well is valuable for the optimization of the cleaning process. The proposed model is based on component balances and contains three parameters that can be determined by a simple experimental protocol that facilitates the possibility of online adaptation of model parameters at frequent intervals. This may be required when process conditions change and/or water quality varies with time. Experiments were performed to test and validate the model.

H 2/N 2 gaseous mixture separation in dense Pd/α-Al 2O 3 hollow fiber membranes: Experimental and simulation studies

Thin Pd/α-Al 2O 3 hollow fiber (HF) membranes were prepared by an improved electroless plating technique. Pure gas permeation experiments showed that the membranes were defect free and had high hydrogen permeance. Feeding an equimolar mixture of H 2/N 2, the total diffusion flow rate of hydrogen through the composite membranes was measured in dependence of the total feed flow rate at the inlet of the shell/retentate chamber, of the temperature and of the total trans-membrane pressure difference. Furthermore, a mathematical model was derived to describe the hydrogen diffusion in pure and mixture gas experiments. The pure gas experiments were used to determine the membrane parameters. Applying the adapted model parameters, the fluxes of hydrogen occurring in the mixture gas experiments were simulated. The comparison between experimental and calculated data showed a very good agreement in a broad parameter range.

Evaluation of ellipsometric measurements using complex strategies

Results are presented of a study of the inverse problem of ellipsometry. In the lack of inverse equations, iterative methods (search) are needed to find the optimal parameters (global minimum) of the adaptive optical model. Different kinds of optical models and search strategies are demonstrated and compared for evaluating real ellipsometric measurements. It is also demonstrated that the well-known gradient search alone is useless if the search space is large (few a-priori information) or the error surface is hilly. The applied algorithms were built in a software, which is freely available at polgaro@mfa.kfki.hu.

Rapid test and non-linear model characterisation of solid-state lithium-ion batteries

This paper describes a rapid test-procedure that can be used to derive parameters of a proposed battery model. The battery model is a non-linear dynamic equivalent circuit model, which is based on Randle’s model for electrochemical impedance [J. Power Sources 54 (1995) 393]. The level of sophistication has been selected such that it gives a satisfactory prediction of battery performance, but simple enough to enable on-line identification and adaptation of model parameters based on measurements of terminal voltage, current and temperature during usage. The paper also presents test data for a commercial 100 Ah battery including ageing effects.

Adaptive modeling and classification of the secondary settling tank

In biological wastewater treatment plants the biomass is separated from the treated wastewater in the secondary settler; thus, efficient operation of the secondary settler is crucial to achieving satisfactory effluent quality in the wastewater treatment process (WWTP). In the present work, system identification and soft-computing techniques were used to formulate a model for predicting the solid volume index (SVI) and classification of the sludge bulking phenomenon in the settler. An adaptive time series model was applied to predict the SVI of the secondary settler; this model uses the recursive least square (RLS) method to update the model parameters. The method for classifying the current state of the secondary settler is based on the strong correlation that was observed between the settler state and the values of the time series model parameters, which enabled the time series model parameters to be used as effective features for monitoring the secondary settler. To classify the current state of the secondary settler, a neural network (NN) was used to classify the adaptive time series model parameters, where a hybrid Genetic Algorithm (GA) was used to decide the number of hidden nodes of the NN classifier. Application of the proposed method to a full-scale WWTP demonstrated the utility of the method for simultaneously predicting the SVI value of the secondary settler and classifying the current state of the settler.

Hybrid adaptive fuzzy identification and control of nonlinear systems

We present a combined direct and indirect adaptive control scheme for adjusting an adaptive fuzzy controller, and adaptive fuzzy identification model parameters. First, using adaptive fuzzy building blocks, with a common set of parameters, we design and study an adaptive controller and an adaptive identification model that have been proposed for a general class of uncertain structure nonlinear dynamic systems. We then propose a hybrid adaptive (HA) law for adjusting the parameters. The HA law utilizes two types of errors in the adaptive system, the tracking error and the modeling error. Performance analysis using a Lyapunov synthesis approach proves the superiority of the HA law over the direct adaptive (DA) method in terms of faster and improved tracking and parameter convergence. Furthermore, this is achieved at negligible increased implementation cost or computational complexity. We prove a theorem that shows the properties of this hybrid adaptive fuzzy control system, i.e., bounds for the integral of the squared errors, and the conditions under which these errors converge asymptotically to zero are obtained. Finally, we apply the hybrid adaptive fuzzy controller to control a chaotic system, and the inverted pendulum system.

Using WOFOST crop model for data base derivation of tritium and terrestrial food chain modules in RODOS

The European Commission Project RODOS is developing a coherent methodology for a Real-time On-line Decision Support System for Nuclear Emergency across Europe. Within this system there is a special module to model the transfer of triated water from releases to terrestrial foods. In order to model the transfer of tritiated water from air to various plants, the conversion to organically bound tritium, and the partition to edible parts of the plant, both the mean dynamics of leaf area index and a physiological description of canopy photosynthesis are required. The WOFOST crop growth model has been selected as a basis for deriving tritium transfer dynamics to plants. Its ability to reproduce site-specific biomass growth of various plants (not only from Europe) is demonstrated in this paper, as well as its compatibility to other photosynthesis models. We have tested that this model can simulate limited fertilisation situations via the adaptation of two important parameters. After adaptation of model parameters to site- specific plant growth data, multi-annual mean dynamics can be obtained using meteorological data for subsequent years.

On adaptive decision rules and decision parameter adaptation for automatic speech recognition

Recent advances in automatic speech recognition are accomplished by designing a plug-in maximum a posteriori decision rule such that the forms of the acoustic and language model distributions are specified and the parameters of the assumed distributions are estimated from a collection of speech and language training corpora. Maximum-likelihood point estimation is by far the most prevailing training method. However, due to the problems of unknown speech distributions, sparse training data, high spectral and temporal variabilities in speech, and possible mismatch between training and testing conditions, a dynamic training strategy is needed. To cope with the changing speakers and speaking conditions in real operational conditions for high-performance speech recognition, such paradigms incorporate a small amount of speaker and environment specific adaptation data into the training process. Bayesian adaptive learning is an optimal way to combine prior knowledge in an existing collection of general models with a new set of condition-specific adaptation data. In this paper, the mathematical framework for Bayesian adaptation of acoustic and language model parameters is first described. Maximum a posteriori point estimation is then developed for hidden Markov models and a number of useful parameters densities commonly used in automatic speech recognition and natural language processing.

RIE process time optimization from real time sensory data

This paper addresses the issue of end point detection and etch time control for a reactive ion etch process. Our approach involves the use of neural networks to model functional relationship between an end point detection signal as well as various in situ measurements, and the resulting film thickness remaining. An optimization algorithm is then employed to determine the optimal etch time based on the neural network model in order to achieve the desired level of film thickness remaining. This circumvents the need for monitoring and operating on noisy end point detection signal typically associated with conventional detection schemes. Furthermore, a self-organizing map is incorporated as a process classifier for identifying the current operating condition of the reactive ion etch process, which provides an alternative for tracking process variations and uncertainties, without the need tor the on-line adaptation of model or controller parameters. Tested on data from 77 randomly selected wafers, our controller yields a film thickness distribution with the standard deviation of 6.54, a 50% improvement over the scheme currently implemented in production.

Speech recognition in noisy environments using first-order vector Taylor series

In this paper, we generalize relations between clean and noisy speech signal using vector Taylor series (VTS) expansion for noise-robust speech recognition. We use it for both the noisy data compensation and hidden Markov model (HMM) parameter adaptation, and apply it for the cepstral domain directly, while Moreno used it to estimate the log-spectral parameters. Also, we develop a detailed procedure to estimate environmental variables in the cepstral domain using the expectation and maximization (EM) algorithms based on the maximum likelihood (ML) sense. To evaluate the developed method, we conduct speaker-independent isolated word and continuous speech recognition experiments. White Gaussian and driving car noises added to clean speech at various SNR are used as disturbing sources. Using only noise statistics obtained from three frames of silence and noisy speech to be recognized, we achieve significant performance improvement. Especially, HMM parameter adaptation with VTS is more effective than the parallel model combination (PMC) based on the log-normal assumption.

A Two-Compartment Convective-Diffusion Model for Slurry Bubble Column Reactors

Bubble columns and slurry bubble column reactors (SBCRs), due to their superior heat transfer characteristics, are the contactors of choice for conversion of syngas to fuels and chemicals. The multiphase fluid dynamics in these systems determines the mass and heat transfer and greatly affects reactor volumetric productivity and selectivity. Here we show how the recently collected data for liquid velocity, turbulence parameters, and holdup profiles can be used in guiding model development for liquid backmixing and in providing the needed model parameters. Predictions of the two-compartment convective-diffusion model for the residence time distribution of a nonvolatile tracer, which are based on measured liquid recirculation driven by the gas holdup profile and turbulence caused by the rising gas bubbles, are shown to be in good agreement with tracer data. The adaptation of model parameters needed to achieve predictability for industrial tracer data is discussed. The ability of the model to predict the liqu...

Exposure Pathways and Dose Calculations in RODOS: Improvement of Predictions by Measured Data

All radiation exposure pathways which are considered in the RODOS food chain and dose modules are described. For each of them, the principal processes controlling the exposure are discussed, showing up the input data requirements. Emphasis is put on those data which vary from region to region and thus require the adaptation of model parameters of RODOS is to be used for regions with different conditions. Calculation of individual doses as well as estimation of collective doses are reflected. Possibilities for improving the dose estimations by consideration of measured data are discussed. The deposition onto different kinds of surfaces (e.g. soil, plant canopies, urban areas) can be improved by means of gamma dose rate measurements, vegetation samples, in situ gamma spectrometry etc. Measurements of concentrations of activity in feed and foodstuffs, whole-body burdens, personal dosemeters etc. can help to improve the predicted internal and external doses.

Optimization of a Servo System With a Referent Model and Inner Parameter Adaptation Loop

Abstract A rather slow adaptation to parameter variations is achieved in those servosystems which contain a PI controlled inner control loop and the outer control loop with a referent model and parameter adaptation. This paper is concerned with involvement of the referent model and parameter adaptation in the inner control loop of the servo system, while keeping the PI controller in an outer control loop. A very fast adaptation was achieved by optimization of parameter adaptation algorithm coefficients for a given range of variations of a drive moment of inertia An impact of a sampling interval and state variable estimator on the maximum value of a referent model tracking error during transient response has investigated.

Constrained Generalized Predictive Control of Unstable Nonlinear Processes

Aconstrained generalized predictive linear control strategy is extended to constrained control of open-loop unstable nonlinear processes. In this strategy, process nonlinearities are accounted through adaptation of model parameters while taking care of input and output constraints acting on the open-loop unstable nonlinear process. The usefulness of this control strategy is evaluated by applying it to a chemical reactor and a biochemical reactor. The results show that this strategy is effective for control of open-loop unstable plants.

Reliable on-line adaption of models with variable time delays

A new technique is presented for on-line adaption of model parameters and time delays. The general case of continuous-time nonlinear models is considered, and the time delays are estimated to a fraction of the sampling period. The technique has the advantage that it requires user specification of only one parameter that determines the covered range of variability in the model parameters and time delays, whereas conventional techniques require a different user-specified parameter value corresponding to each particular variability. Prior knowledge of negligible or very high variability of certain model parameters or time delays can nonetheless readily be incorporated into the proposed strategy. Performance of the technique is demonstrated on simulation runs for a single-input single-output system, with two parameters and one time delay, and comparisons are made with a conventional limited-memory estimator.

ROBUST NONLINEAR PREDICTIVE CONTROL USING A DISTURBANCE ESTIMATOR

A robust nonlinear predictive control strategy using a disturbance estimator is presented. The disturbance estimator is comprised of two parts: one is the disturbance model parameter adaptation and the other is future disturbance prediction. A linear discrete model is proposed as a disturbance model which is formulated by using process inputs and available process measurements. The recursive least square (RLS) method with exponential forgetting is used to determine the uncertain disturbance model parameters and for the future disturbance prediction, future disturbances projected by the future process inputs are used. Two illustrative examples: a jacketed CSTR as a SISO system: an adiabatic CSTR as a MIMO system, and experimental results of the distillation column control are presented. The results indicate that a substantial improvement in nonlinear predictive control performance is possible using the disturbance estimator.

GENERIC MODEL ADAPTIVE CONTROL

Abstract Generic Model Control (GMC) is a process model based control algorithm incorporating a process model directly within the control structure. It has been shown to produce excellent control, despite reasonable modelling errors. In this paper an algorithm is developed within a GMC framework which reduces the effect of larger modelling errors by regularly updating the model parameters. This new adaptive algorithm is capable of adapting model parameters in a nonlinear model, where the parameters appear in a nonlinear manner. Several examples are presented to illustrate the principles of the technique.