Jeffrey S. Vipperman

Radiation modal expansion: Application to active structural acoustic control

This paper demonstrates active structural acoustic control using multiple input/output adaptive sensoriactuators combined with radiation filters and a feedback control paradigm. A new method of reduced order modeling/design of radiation filters termed radiation modal expansion (RME) is presented. For the experiments detailed in this paper, the RME technique reduced the modeling of the radiation matrix from 400 transfer functions to 6 transfer functions (multiplied by a constant transformation matrix). Experimental results demonstrate reductions of radiated sound power on the order of 5 dB over the bandwidth of 0-800 Hz.

Experimental Active Control of a Typical Section Using a Trailing-Edge Flap

This paper presents an experimental implementation of an active control system used to suppress flutter in a typical section airfoil. The H2 optimal control system design is based on experimental system identifications of the transfer functions between three measured system variables - pitch, plunge, and flap position - and a single control signal that commands the flap of the airfoil. Closed-loop response of the airfoil demonstrated gust alleviation below the open-loop flutter boundary. In addition, the flutter boundary was extended by 12.4% through the application of active control. Cursory robustness tests demonstrate stable control for variations in flow speed of ± 10%.

Implementation of an adaptive piezoelectric sensoriactuator

An adaptive algorithm implemented on a digital signal processor is used in conjunction with an analog multiplier circuit to compensate for the feedthrough capacitance of a piezoelectric sensoriactuator. The mechanical response of the piezoelectric sensoriactuator is resolved from the electrical response by adaptively altering the gain imposed on the electrical circuit used for compensation. For broadband, stochastic input disturhances, the feedthrough dynamic capacitance of the sensoriactuator can be identified on-line, providing a means of implementing direct-rate-feedback control in analog hardware. Rate feedback control is demonstrated on a cantilevered beam using the hybrid/adaptive circuit. The ability of the circuit to readapt to a change in capacitance is demonstrated by a simple experiment.

Multivariable feedback active structural acoustic control using adaptive piezoelectric sensoriactuators

An experimental implementation of a multivariable feedback active structural acoustic control system is demonstrated on a piezostructure plate with pinned boundary conditions. Four adaptive piezoelectric sensoriactuators provide an array of truly colocated actuator/sensor pairs to be used as control transducers. Radiation filters are developed based on the self- and mutual-radiation efficiencies of the structure and are included into the performance cost of an H2 control law which minimizes total radiated sound power. In the cost function, control effort is balanced with reductions in radiated sound power. A similarity transform which produces generalized velocity states that are required as inputs to the radiation filters is presented. Up to 15 dB of attenuation in radiated sound power was observed at the resonant frequencies of the piezostructure.

Causality analysis of feedforward-controlled systems with broadband inputs

In recent years adaptive feedforward control algorithms have been successfully implemented to attenuate the response of systems under persistent disturbances such as single and multiple tones as well as random inputs. System causality is not an issue when the excitation is sinusoidal because of its deterministic nature. However, causality is a very important factor in broadband control. Though significant deterioration in the performance of noncausal control systems have been reported in the literature, analytical tools are virtually nonexistent to predict the behavior of broadband controllers. The main objective of this research is to develop an approach to investigate system causality. A formulation is presented to address the effectiveness of a control configuration as a function of the filter size, delay time, and dynamic properties of the structure. The technique is illustrated in a simple numerical example and the results are also corroborated experimentally.

Comparison of π- and H 2 -Synthesis Controllers on an Experimental Typical Section

An experimental comparison of H2- and π-synthesized e utter suppression control systems was performed. A simpleparametric uncertainty can beused to track changesin system dynamics asa function of dynamicpressure. The control system was implemented experimentally on a NACA 0012 test model of a typical section mounted in a low-speed wind tunnel. The pitching angle, e ap angle, and plunge dee ection of the airfoil were measured with sensorsand fed back through the control compensator to generatea single control signal commanding the trailingedge e ap of the airfoil. The model of the aeroelastic system, including the dynamics of the sensors and actuators in the bandwidth of interest, was obtained using system identie cation techniques. For comparison purposes, an H2 control system with standard linear quadratic Gaussian weightings also was designed and implemented. When compared to the H2 control system, the π-synthesis controller provided better disturbance rejection in the bandwidth of the unsteady aeroelastic dynamics. In addition, the π controller required less control energy than the H2 control system. The e nal advantage of π-synthesis is the ability to design an aggressive π control system that is stabilizing across the range of operating dynamic pressures.

A principal component feedforward algorithm for active noise control: flight test results

An in-flight evaluation of a principal component algorithm for feedforward active noise control is discussed. Cabin noise at the first three harmonics of the blade passage frequency (103 Hz) of a Raytheon-Beech 1900D twin turboprop aircraft was controlled using 21 pairs of inertial force actuators bolted to the ring frames of the aircraft; 32 microphones provided error feedback inside the aircraft cabin. In a single frequency noise control test, the blade passage frequency was reduced by 15 dB averaged across the microphone array. When controlling the first three harmonics simultaneously, reductions of 11 dB at 103 Hz, 1.5 dB at 206 Hz, and 2.8 dB at 309 Hz were obtained. For single frequency feedforward control problems, the principal component algorithm is shown to be useful for reducing the computational burden and simplifying the implementation of control effort penalties in high channel count control systems. Good agreement was found between the in-flight behavior of the controller and the predicted optimal control solution.

Implications of using colocated strain-based transducers for output active structural acoustic control

Piezoceramic transducers have become popular elements of smart structures that are used for active vibration control and active structural acoustic control. A spatial differentiation is performed by the piezoceramic transducers since they couple into the strain field of the piezostructure. This differentiation causes higher-frequency modes to be emphasized more heavily, causing the effective compliance of the structure as seen by the piezoceramic transducer to increase with frequency. This nonuniform compliance has significant impact on the performance that can be achieved through colocated direct rate feedback control. It is shown that the rectangular piezoceramic transducer is a low-pass wave number filter with a cutoff frequency inversely proportional to the aperture size. Thus DRFB performance can be greatly improved simply by making the size of the piezoceramic transducer large relative to the size of the structure. The resulting increase in coupling to the lower-frequency modes, which are generally targeted by the control system, results in a much reduced control effort. In the event that a large aperture is not practical, it is shown that dynamic compensation can be used to obtain good performance at the cost of much increased computational complexity. Analytical open and closed loop results for an acoustically radiating simply supported plate piezostructure are presented.

INVESTIGATION OF THE EXPERIMENTAL ACTIVE CONTROL OF A TYPICAL SECTION AIRFOIL USING A TRAILING EDGE FLAP

This paper presents an experimental implementation of an active control system used to suppress flutter in a typical section airfoil. The H? optimal control system design is based upon experimental system identifications of the transfer functions between three measured system variables: pitch, plunge, and flap position and a single control signal which commands the flap of the airfoil. Closed-loop response of the airfoil demonstrated gust-alleviation below the open-loop flutter-boundary. In addition, the flutter boundary was extended by 12.4% through the application of active control. Cursory robustness tests demonstrate stable control for variations in flow-speed of ±10

Computational study of human head response to primary blast waves of five levels from three directions.

Human exposure to blast waves without any fragment impacts can still result in primary blast-induced traumatic brain injury (bTBI). To investigate the mechanical response of human brain to primary blast waves and to identify the injury mechanisms of bTBI, a three-dimensional finite element head model consisting of the scalp, skull, cerebrospinal fluid, nasal cavity, and brain was developed from the imaging data set of a human female. The finite element head model was partially validated and was subjected to the blast waves of five blast intensities from the anterior, right lateral, and posterior directions at a stand-off distance of one meter from the detonation center. Simulation results show that the blast wave directly transmits into the head and causes a pressure wave propagating through the brain tissue. Intracranial pressure (ICP) is predicted to have the highest magnitude from a posterior blast wave in comparison with a blast wave from any of the other two directions with same blast intensity. The brain model predicts higher positive pressure at the site proximal to blast wave than that at the distal site. The intracranial pressure wave invariably travels into the posterior fossa and vertebral column, causing high pressures in these regions. The severities of cerebral contusions at different cerebral locations are estimated using an ICP based injury criterion. Von Mises stress prevails in the cortex with a much higher magnitude than in the internal parenchyma. According to an axonal injury criterion based on von Mises stress, axonal injury is not predicted to be a cause of primary brain injury from blasts.