Mcdonnell Douglas Corporation

Peter F Lichtenwalner

Lawrence Eric Pado

James P Dunne

Erwin W Baumann

Ronald S Becker

A method and system for controlling a dynamic nonlinear plant. An input signal controls the plant and an output signal represents a state of the plant in response to the received input signal. A memory stores input and output signals corresponding to m consecutive past states of the plant. A computer neural network predicts a set of future output states representative of the output signal corresponding to the next n consecutive future states of the plant in response to a set of trial control inputs. The trial control inputs represent the input signal corresponding to the next n consecutive future states of the plant. The neural network predicts the future output states based on the past input and output signals and the future trial control inputs. A processor generates the trial control inputs and determines a performance index, indicative of plant performance over time in response to the trial control inputs, as a function of the future output states. The processor generates the input signal for controlling the plant and modifies it as a function of the trial control inputs so that the performance index reaches a desired value.

Neural network predictive control method and system

An active damage interrogation (ADI) system (and method) which utilizes an array of piezoelectric transducers attached to or embedded within the structure for both actuation and sensing. The ADI system actively interrogates the structure through broadband excitation of the transducers. The transducer (sensor) signals are digitized and the transfer function of each actuator/sensor pair is computed. The ADI system compares the computed transfer function magnitude and phase spectrum for each actuator/sensor pair to a baseline transfer function for that actuator/sensor pair which is computed by averaging several sets of data obtained with the structure in an undamaged state. The difference between the current transfer function and the baseline transfer function for each actuator/sensor pair is normalized by the standard deviation associated with that baseline transfer function. The transfer function deviation for each actuator/sensor pair is then represented in terms of the number of standard deviations, or sigmas, from the baseline. This statistic, termed the TF Delta, is then processed by a windowed local averaging function in order to reduce minor variations due to random noise, etc. The Windowed TF Delta for each actuator/sensor pair is then integrated over the entire excitation frequency spectrum, to thereby produce the Cumulative Average Delta, which provides a single metric for assessing the magnitude of change (deviation from baseline) of that particular actuator/sensor transfer function. The Cumulative Average Delta (CAD) for each actuator/sensor transfer function provides key, first-level information which is required for detecting, localizing, and quantitatively assessing damage to the structure.

Growing community of inventors

St. Charles, MO, United States of America

Peter F Lichtenwalner