Phillips, Stephen M. | Search NASA

NASA NTRS2003

Multi-layer assemblies with predetermined stress profile and method for producing same

Multi-layer assemblies of polysilicon thin films having predetermined stress characteristics and techniques for forming such assemblies are disclosed. In particular, a multi-layer assembly of polysilicon thin films may be produced that has a stress level of zero, or substantially so. The multi-layer assemblies comprise at least one constituent thin film having a tensile stress and at least one constituent thin film having a compressive stress. The thin films forming the multi-layer assemblies may be disposed immediately adjacent to one another without the use of intermediate layers between the thin films. Multi-layer assemblies exhibiting selectively determinable overall bending moments are also disclosed. Selective production of overall bending moments in microstructures enables manufacture of such structures with a wide array of geometrical configurations.

Heuer, Arthur H.↗

NASA NTRS2002

Large area polysilicon films with predetermined stress characteristics and method for producing same

Multi-layer assemblies of polysilicon thin films having predetermined stress characteristics and techniques for forming such assemblies are disclosed. In particular, a multi-layer assembly of polysilicon thin films may be produced that has a stress level of zero, or substantially so. The multi-layer assemblies comprise at least one constituent thin film having a tensile stress and at least one constituent thin film having a compressive stress. The thin films forming the multi-layer assemblies may be disposed immediately adjacent to one another without the use of intermediate layers between the thin films. Multi-layer assemblies exhibiting selectively determinable overall bending moments are also disclosed. Selective production of overall bending moments in microstructures enables manufacture of such structures with a wide array of geometrical configurations.

Heuer, Arthur H.↗

NASA NTRS1994

Antiwindup analysis and design approaches for MIMO systems

Performance degradation of multiple-input multiple-output (MIMO) control systems having limited actuators is often handled by augmenting the controller with an antiwindup mechanism, which attempts to maintain system performance when limits are encountered. The goals of this paper are: (1) To develop a method to analyze antiwindup systems to determine precisely what stability and performance degradation is incurred under limited conditions. It is shown that by reformulating limited actuator commands as resulting from multiplicative perturbations to the corresponding controller requests, mu-analysis tools can be utilized to obtain quantitative measures of stability and performance degradation. (2) To propose a linear, time invariant (LTI) criterion on which to base the antiwindup design. These analysis and design methods are illustrated through the evaluation of two competing antiwindup schemes augmenting the controller of a Short Take-Off and Vertical Landing (STOVL) aircraft in transition flight.

Marcopoli, Vincent R.↗

NASA NTRS1994

Control strategies for systems with limited actuators

This work investigates the effects of actuator saturation in multi-input, multi-output (MIMO) control systems. The adverse system behavior introduced by the saturation nonlinearity is viewed here as resulting from two mechanisms: controller windup - a problem caused by the discrepancy between the limited actuator commands and the corresponding control signals, and directionality - the problem of how to use nonlimited actuators when a limited condition exists. The tracking mode and Hanus methods are two common strategies for dealing with the windup problem. It is seen that while these methods alleviate windup, performance problems remain due to plant directionality. Though high gain conventional antiwindup as well as more general linear methods have the potential to address both windup and directionality, no systematic design method for these schemes has emerged; most approaches used in practice are application driven. An alternative method of addressing the directionality problem is presented which involves the introduction of a control direction preserving nonlinearity to the Hanus antiwindup system. A nonlinearity is subsequently proposed which reduces the conservation inherent in the former direction-preserving approach, improving performance. The concept of multivariable sensitivity is seen to play a key role in the success of the new method.

Marcopoli, Vincent R.↗

NASA NTRS1993

Antiwindup analysis and design approaches for MIMO systems

Performance degradation of multiple-input multiple-output (MIMO) control systems having limited actuators is often handled by augmenting the controller with an antiwindup mechanism, which attempts to maintain system performance when limits are encountered. The goals of this paper are: 1) to develop a method to analyze antiwindup systems to determine precisely what stability and performance degradation is incurred under limited conditions. It is shown that by reformulating limited actuator commands as resulting from multiplicative perturbations to the corresponding controller requests, mu-analysis tools can be utilized to obtain quantitative measures of stability and performance degradation. 2) To propose a linear, time invariant (LTI) criterion on which to base the antiwindup design. These analysis and design methods are illustrated through the evaluation of two competing antiwindup schemes augmenting the controller of a Short Take-Off and Vertical Landing (STOVL) aircraft in transition flight.

Marcopoli, Vincent R.↗

NASA NTRS1988

An averaging analysis of discrete-time indirect adaptive control

An averaging analysis of indirect, discrete-time, adaptive control systems is presented. The analysis results in a signal-dependent stability condition and accounts for unmodeled plant dynamics as well as exogenous disturbances. This analysis is applied to two discrete-time adaptive algorithms: an unnormalized gradient algorithm and a recursive least-squares (RLS) algorithm with resetting. Since linearization and averaging are used for the gradient analysis, a local stability result valid for small adaptation gains is found. For RLS with resetting, the assumption is that there is a long time between resets. The results for the two algorithms are virtually identical, emphasizing their similarities in adaptive control.

Phillips, Stephen M.↗

NASA NTRS1988

Averaging analysis of adaptive control algorithms

The method of averaging is used to analyze discrete-time indirect adaptive control. The analysis focuses on various prediction-error-driven identification algorithms coupled with a general linear control law. The plant is not required to be in the model set of the identifier, which accounts for systems with unmodeled plant dynamics. Exogenous input signals including known command signals and unknown disturbances are also included. Both gradient and Newton-based algorithms are considered.

Phillips, Stephen M.↗