Large Parameter Variations Research Articles

Offset-trimming bit-line sensing scheme for gigabit-scale DRAM's

A new offset-trimming bit-line sensing scheme is described which is suitable for gigabit-scale DRAM's. This sensing scheme can suppress the sensitivity degradation caused by the large electrical parameter variation of deep submicron transistors. The effective offset voltage dependence on trimming time is analyzed and verified with simulation results. As compared with a conventional direct sensing scheme, the proposed scheme shows remarkable improvement on the sensitivity. A test device was fabricated with a 0.25 /spl mu/m CMOS technology and its measurement results indicate the successful operation of offset-trimming.

Changes in Vancomycin Pharmacokinetics in Critically Ill Infants

We aimed to assess the pharmacokinetics of vancomycin in critically ill infants, and to evaluate the standard recommended dose of 10 mg/kg 6 hourly. All infants admitted to the Baragwanath Hospital ICU who had arterial lines in situ, and for whom vancomycin 10 mg/kg 6 hourly was prescribed for an infective insult and who had parental consent, were included in the study. Vancomycin was infused over 60 minutes. Serum samples were taken immediately before the dose and at 30, 60, 120 and 300 minutes after the end of the vancomycin infusion, on days 2 and 8 of therapy. Extrapolated peak concentration (Cmax), trough concentration (Cmin), apparent volume of distribution (Vd), elimination half-life (t1/2el) and clearance (CL) were determined for each patient. Day 2 values were compared with those of day 8. Day 2 serum concentrations were assayed on 20 patients and day 8 concentrations in 15. The mean vancomycin Vd on day 2 (0.81 l/kg) was significantly (P = 0.007) larger than that on day 8 (0.44 l/kg). The change in Vd resulted in a significant change in mean Cmax (29.1 vs 35.5 micrograms/ml) (P = 0.02) and mean t1/2el (5.3 vs 3.4h) (P = 0.01) over the treatment period. Critically ill infants displayed a large initial volume of distribution which probably resulted from aggressive fluid resuscitation. This also results in a large variation in other pharmacokinetic parameters, namely Cmax and t1/2el. Although the routine monitoring of vancomycin serum concentrations remain controversial, we feel that in view of these large pharmacokinetic variations, the critically ill infant is a specific group where monitoring of vancomycin serum levels is indicated.

Application of H∞ Control and Lqg with Frequency-Dependent Weight Syntheses to a Multivariable Active Flutter Suppression System

This paper describes the effectiveness of three robust control synthesis techniques applied to a multivariable active flutter suppression system using numerical simulations. These techniques are the mixed-sensitivity reduction in H∞ (H∞-mix), the normalized left coprime factorization robust stabilization in H∞ (H∞-normalized), and LQG with frequency-dependent weights (FWLQG), which are proposed in this paper. Although the plant is an unstable multi-input and multi-output system with large parameter variation, the FWLQG technique is superior to both H∞ synthesis techniques from the viewpoint of the increment of the flutter velocity, the time response and the solubility of control problems.

Temperature regulation with neural networks and alternative control schemes

In this article, we compare the neuro-control algorithm to three other control algorithms: fuzzy logic control, generalized predictive control, and proportional-plus-integral control. Each of these four algorithms is implemented on a water bath temperature control system. The four systems are compared through experimental studies under identical conditions with respect to set-point regulation, the effect of unknown load disturbances, large parameter variation, and variable deadtime in the system. It is found that the neurocontrol system compares well with the other three control systems and offers encouraging advantages. From the results of the experimental studies, however, the best characteristics of each of these different classes of control systems may be combined for realizing a more efficient and intelligent control scheme.

Effect of respiration on Doppler parameters of normal tricuspid porcine bioprosthetic valves.

The Doppler indexes of tricuspid porcine bioprosthetic valves were evaluated in twelve patients without clinical and two-dimensional echocardiographic evidence of valve dysfunction. Peak and mean pressure gradients across the prostheses were measured using the simplified Bernoulli equation. All the Doppler measurements were compared during inspiration and expiration. During inspiration peak velocity, peak gradient and mean gradient (1.52 +/- 0.28 m/s; 9.7 +/- 3.05 mmHg; 4.07 +/- 1.16 mmHg) were significantly higher than during expiration (1.28 +/- 0.8 m/s; 6.58 +/- 2.7 mmHg; 2.98 +/- 1.13 mmHg; p < 0.01) but pressure half time was not significantly different (122 +/- 62 ms versus 134 +/- 75 ms; p > 0.05). Inspiratory range of peak velocities, peak gradients, mean gradients and pressure half times were respectively 0.8-2.04 m/s; 4.9-16.6 mmHg; 1.2-7.2 mmHg; 42-340 ms while expiratory range of values was 0.8-1.93 m/s; 2.6-15 mmHg; 1.1-5.7 mmHg; 46-345 ms. These data suggest that even very long pressure half times do not indicate valve dysfunction. This study demonstrates that large variation of Doppler parameters are present during respiration and could produce inaccuracy in the assessment of bioprostheses in tricuspid position if they are not taken in consideration.

HDR test series E33.1/2 for investigation of contact condensation processes in horizontal pipes during emergency core cooling injection

Within the framework of phase III of the Germany HDR Safety program, an experimental facility was planned and built to investigate and simulate contact condensation processes during emergency core cooling injection during loss-of-coolant accidents. In the pressurized water reactors of the Siemens/UB KWU, the emergency core cooling system is based on a simultaneous cold water injection into the cold leg (nozzle inlet) and into the hot leg (so called ‘Hutze’). The HDR-E33.1/2 test facility is a single-effect test facility with a pipe 8 m long and 187 mm in diameter, as well as a horizontal cold water inlet via nozzle (diameter, 65 mm). The test series E33.0/1 and E33.2 were carried out in May and September 1990, with large variation of the test parameters, water and steam mass flow rates and the system pressure. Two basically different flow patterns predominated during the tests: stratified flow and plug flow. The immediate area surrounding the nozzle region functioned as the zone of the greatest interaction between the steam and cold water, irrespective of the actual flow pattern. This paper is based on a systematization and first physical evaluation, and reviews the significant results from about 170 HDR tests.

Active Control of Chaotic Vibration in a Constrained Flexible Pipe Conveying Fluid

An active vibration control system is designed, and is numerically verified, to suppress the undesirable chaotic vibration in a constrained flexible pipe conveying fluid, which exhibits regions of flutter and chaotic motions at sufficiently high flow velocity. The four-dimensional analytical model obtained from the continuous system by Galerkin's method, that has been previously verified to represent adequately the dynamics, is utilized for designing the control law. Various control design methodologies are investigated for different situations. Firstly, optimal regulator theory is applied to obtain feedback gains to stabilize the system with full state information. The effects of the location and length of the piezoelectric actuators on the efficiency as a vibration damper are theoretically examined in this case. Secondly, a state observer is added to estimate the required state signals. To cover the situation when the states are not directly measurable, dynamic compensators are obtained to control the system with only the output feedback. Finally, a robust controller for such a system with large flow velocity variations, without sensing the flow velocity or gain-scheduling, is developed. The robust control method is based on a newly developed sensitivity-based Quantitative Feedback Theory (QFT) scheme, which allows the controller to make the closed loop system response meet quantitatively specified performance requirements even though the system has large parameter variations. Numerical simulations are carried out for the different controllers and they validate the effectiveness of the proposed control scheme. The QFT scheme yields a remarkable result in stability robustness with respect to flow velocity variations.

Model reference adaptive fuzzy control of high-order systems

The paper presents a model reference adaptive fuzzy control scheme which is suitable for high-order systems of well-known structure. It contains a hybrid fuzzy controller composed of a nonintegral fuzzy and a proportional-integral algorithm. The output of the hybrid fuzzy controller is adjusted by a fuzzy tuning algorithm. The synthesis of the hybrid fuzzy controller and the fuzzy tuning algorithm has been described for the angular speed control problem of the linearized vector controlled permanent magnet synchronous motor drive. The computer simulation results have proved the effectiveness of the proposed method, showing stable system responses with low sensitivity to large parameter variations.

Multivariable Active Flutter Suppression Tems using H∞ Control and LQG with Frequency Dependent Weight Syntheses

This paper describes the effectiveness of three robust control synthesis techniques applied to a multivariable active flutter suppression system using numerical simulations. These techniques are the mixed-sensitivity reduction in H∞(H∞-mix), the normalized left coprirne factorization robust stabilization in H∞ (H∞-normalized), and LQG with frequency-dependent weights (FWLQG), which is proposed in this paper. Although the plant is an unstable multi-input and multi-output system with large parameter variation, the FWLQG technique is superior to both H∞ synthesis techniques from viewpoints of the increment of the flutter velocity, the time response and the solubility of control problems.

Analysis of CHF location sensibility following heat flux and mass flux variations

An experimental investigation has been performed on critical heat flux (CHF) in subcooled upflow boiling with a uniformly heated vertical tube, using R12 as a coolant, to evaluate the dependence of the CHF location on the variation of the thermal hydraulic parameters affecting the thermal crisis, such as mass flux and heat flux. The test section consists of a stainless steel tube, 7.57 mm i.d., 2.1 m long, uniformly heated using d.c. current. The research showed a different behaviour of the CHF location following mass flux and heat flux variations in the low and in the high quality regions. The sensibility of the CHF location upon the variation of a thermal hydraulic parameter, such as mass flux or heat flux, decreases as the critical quality decreases, and with increasing of the mass flux. At very low critical qualities the CHF location shows a relative insensibility upon large variations of thermal hydraulic parameters. A weak influence of pressure has also been detected.

Nonlinear Adaptive Control of a Continuous Polymerization Reactor

Nonlinear control strategies are suitable for processes presenting important nonlinearities. In the present polymerization process, a wide domain of monomer conversion is reachable and the gel effect is considered, imposing large variations in the model. A precise dynamic model is thus required. In this study, an adaptive control for a multi-input multi-output linearizable system, together with an extended Kalman filter, is applied to a simulated continuous polymerization reactor to follow two output set-points (monomer conversion and reactor temperature) in the presence of model parameter uncertainties. Simulation results show that this technique is robust and the output performance can be ensured even in the presence of large model parameter errors or variations and disturbances.

Stability robustness analysis of discrete-time systems with multiple large parameter variations

A method is presented to construct the largest set of parameter variations for any asymptotically stable (convergent) matrix, whose corresponding set of perturbed matrices remains stable.

H(infinity) robust control design for linear feedback systems

This paper deals with the problem of designing, for a linear time-invariant multivariable plant, a feedback controller that minimizes the H°° norm of a quadratic combination of the sensitivity and complementary sensitivity functions. With the parameterization of the controller and quadratic cost, a design procedure based on a gain equalization test for computing the optimal controller is described in detail. A design algorithm, which is implementable in computers, is presented. The pitch-axis controller design for a digitized version of the Grumman F-14 aircraft is given. The performance robustness under the external disturbances and large aircraft parameter variations are shown to demonstrate the effectiveness of this approach.

The determination of feedback gains for uncertain linear systems using higher‐order sensitivity minimization

AbstractTo reduce the influence of large parameter variation on the response of linear systems, this paper considers the design of controllers that minimize the higher‐order sensitivities. Here the higher‐order sensitivities are defined as the second‐order, third‐order, etc. partial derivatives of the state and control vectors with respect to the uncertain parameter. The feedback gains are obtained by minimizing a quadratic cost function that includes state and control sensitivities of arbitrary order. The paper also presents an efficient numerical procedure for solving the resultant system of non‐linear matrix equations. The effectiveness of the approach is demonstrated by the design of an output feedback controller for an electromagnetic suspension. By comparison with standard design approaches it is shown that higher‐order sensitivity minimization can be used to reduce the deviation from the nominal trajectory caused by parameter variation.

ROBUST CONTROL OF A CHAOTIC VIBRATORY SYSTEM

An active robust control system is designed, and is numerically verified, to suppress the undesirable chaotic vibration in a constrained flexible pipe conveying fluid, which exhibits regions of flutter and chaotic motions at sufficiently high flow velocity. The four-dimensional analytical model obtained from the continuous system by Galerkin's method is utilized for designing the control law. Some interesting nonlinear behaviors with the four-dimensional analytical model are discussed as well. Without sensing the flow velocity or gain-scheduling, a robust controller is developed for the system under large flow rate variations by using a newly developed Quantitative Feedback Design (QFD) method. This technique allows the controller to make the closed loop system response meet quantitatively specified performance requirements even though the system has large parameter variations. Numerical simulations validate the effectiveness of the QFD method which yields a remarkable result in stability and performance robustness with respect to flow velocity variations.

Optimal input for identification versus sensitivity reduction of systems with large parameter variations

In this article, we present an algorithm leading to an optimal controller gain for the automatic regulation of a linear system (closed-loop policy) and to an optimal auxiliary input. This is used for the purpose of either system identification, resulting in the maximum sensitivity measure and thus increased accuracy (Refs. 1–25), or system sensitivity reduction, resulting in the minimum sensitivity measure and thus reliable operation (Refs. 26–46). These results, which are robust in terms of parameter variations, are developed without constraints on the input functions.

Power Spectral Density of Nonlinear System Response: The Recursion Method

Spectral densities of the response of nonlinear systems to white noise excitation are considered. By using a formal solution of the associated Fokker-Planck-Kolmogorov equation, response spectral densities are represented by formal power series expansion for large frequencies. The coefficients of the series, known as the spectral moments, are determined in terms of first-order response statistics. Alternatively, a J-fraction representation of spectral densities can be achieved by using a generalization of the Lanczos algorithm for matrix tridiagonalization, known as the “recursion method.” Sequences of rational approximations of increasing order are obtained. They are used for numerical calculations regarding the single-well and double-well Duffing oscillators, and Van der Pol type oscillators. Digital simulations demonstrate that the proposed approach can be quite reliable over large variations of the system parameters. Further, it is quite versatile as it can be used for the determination of the spectrum of the response of a broad class of randomly excited nonlinear oscillators, with the sole prerequisite being the availability, in exact or approximate form, of the stationary probability density of the response.

Emergence of orientation selective simple cells simulated in deterministic and stochastic neural networks.

The processing of visual data in area 17 of the mammalian cortex is mainly performed by cells with receptive fields which are tuned to different orientations of input stimuli. The mechanisms underlying the emergence of receptive field properties of orientation selective cells are not well understood up to now. Recently, some models for the prenatal development of the receptive fields of orientation selective simple cells have been proposed, which emerge in neural networks trained by Hebb type unsupervised learning rules. These models, however, use different network architectures and are restricted to the case of identical input neurons. In this work, a biologically motivated neural network model with a general architecture is presented. It is trained with a Hebb type updating rule and with uncorrelated input. The input neurons are identified with retinal ganglion cells and exhibit mature Mexican hat type receptive fields. If the receptive fields of the input neurons have identical properties (deterministic model), a set of parameter domains is found, which characterize different kinds of receptive field maturation behaviour of the network. Results obtained by other authors with similar models are contained in this description as special cases. In addition, the more general and rarely investigated stochastic model, where random variations of the parameters describing the receptive fields of the input neurons occur, is investigated. A high sensitivity of the network against these random variations is obtained. In case of large variations of receptive field parameters of the ganglion cells, a qualitatively new kind of maturation behaviour appears. A significant part of the synaptic connections from ganglion cells to the cortical cell is removed and small simple cell receptive fields with only few lobes emerge. The stochastic model is found to provide a better description of the size, scatter and structure of receptive fields present in biological systems, than the deterministic model.

Reconstruction of the parameters of plane-layered elastic media by generalized ray theory

This paper seeks to reconstruct the parameters of elastic layered media such as P-wave velocity, S-wave velocity, density and thickness from the multioffset seismic reflection data. Since the data are highly non-linear to the low-wavenumber components, the non-linear waveform inversion method, with the aid of generalized ray theory, is proposed to solve this problem in space–time domain. As opposed to the layer-stripping method, the present method attempts to invert all layer parameters simultaneously, thus reducing the cumulative errors resulting from the upper layers. The parameters are inverted by minimizing the weighted square error between the observed data and the calculated data of the layered model, the optimization of which is based on the quasi-Newton method. In synthetic tests, we find that the inverted results are good when the variation of parameters between layers is not too large. The modified method for large variation of parameters is first to fix those of deeper layers and neglect the signals reflected from them, then recover some other parameters simultaneously until those of upper layers attain a stable value, and finally, invert all parameters simultaneously. The results so obtained show a significant improvement. This method was tested to be stable in the presence of noise in seismograms.

SYSTEM IDENTIFICATION WITH ARTIFICIAL NEURAL NETWORKS

System identification is the term scientists and engineers use to refer to the process of building mathematical models of dynamical systems based on observed data. This paper approaches system identification as a pattern recognition problem. We use computers to simulate the system response for a variety of different mathematical models. For each distinct system model, simulated system responses tend to remain segregated in one or more amorphous regions of system response space despite (1) large variations in system parameters, (2) experimental errors, and (3) noise. The actual system response is classified with the model corresponding to the region of system response space where it is found. The classifier is an Artificial Neural Network (ANN) which implements a Generalized Vector Quantizer (GVQ). A small number of simple but powerful discriminant functions facilitate the correct classification of most of the responses in any given region. The required distribution of discriminants among the regions evolves automatically as they learn their respective functions.