Data processing: vehicles – navigation – and relative location – Vehicle control – guidance – operation – or indication – Vehicle diagnosis or maintenance indication
Reexamination Certificate
2002-03-28
2003-11-04
Camby, Richard M. (Department: 3661)
Data processing: vehicles, navigation, and relative location
Vehicle control, guidance, operation, or indication
Vehicle diagnosis or maintenance indication
C701S031000, C701S035000
Reexamination Certificate
active
06643569
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates to methods and systems for detecting failures or performance degradation in dynamic systems such as flight vehicles.
2. Background Art
FIG. 1
is a representation of the prior art, which depicts failure detection and isolation (i.e., FDI) of flight vehicle icing. Generally, the method of this prior art teaches detection and isolation of failures in flight vehicles that result in a loss of control effectiveness. Detection and isolation of failures is accomplished via a linear state estimator or observer that continuously calculates an estimate {circumflex over (x)} of the state vector x of the flight vehicle dynamic system in question. The flight vehicle dynamic system is assumed to have sensors available for measuring some or all of the state variables. The measured values y are normally present as part of the dynamic system.
The state estimator calculates estimated values ŷ of the sensor outputs and is designed such that for no system failures the estimated values ŷ agree with measured sensor values y. Whenever system failures (as described above) occur, there is a non-zero error e difference between y and ŷ:
e=y−ŷ.
Each state estimator is designed to detect and isolate a particular hypothesized failure mode ƒ.
The feedback gains for the state estimator are chosen such that the error residuals for a given hypothesized failure are in a unique direction in output space. Isolation of the failure from other possible failures is done via the directionality of the error residuals.
The state estimator is a linear Luenberger-type observer which represents a dynamic system that has dynamics typically given by nonlinear mathematical models. The dynamics are typically a linearized model of a nonlinear system. In this case, the models can be obtained from empirical measurement of dynamic system performance through instrumented flight test or through mathematical modeling of the system. The nonlinear models may be obtained from the above data by any of a number of standard regression techniques as is known in the art.
The state estimator is an observer of the form:
{circumflex over ({dot over (x)})}=A{circumflex over (x)}+Bu+L
(
y−ŷ
)
ŷ=C{circumflex over (x)}+Du,
where A and B are the state transition and input matrices, respectively, of the nominal reference model for the system dynamics, and D is the estimator gain matrix that is chosen such that the output error residual e=y−ŷ is one dimensional. The design of the estimator gain is explained in section 2.1 of Appendix A hereto. The nominal matrices A and B are obtained by linearizing the system nonlinear models at an operating point and are approximately valid representations of the dynamic system in a neighborhood of that operating point. Other sets of matrices are required to characterize flight vehicle dynamics over the entire flight envelope.
It was previously recognized by the inventors that some procedure could be found for selecting operating points and associated neighborhoods, and representing the dynamic system within that neighborhood based on certain operating parameters for the dynamic system (e.g., flap setting angles for flight vehicles). This procedure is called “gain scheduling” in the prior art papers authored by the inventors.
However, it was not known or shown in the prior art how to select operating points or how large the neighborhoods could be to achieve acceptable error levels for the FDI.
It was also recognized in the prior art papers authored by the inventors and noted herein below that detection and isolation of hypothesized failures could be accomplished by examining the magnitude of error residuals along the direction of the output of the FDI for the hypothesized failure.
U.S. Pat. Nos. 5,615,119; 5,819,188; and 6,085,127 to Vos disclose fault tolerant automatic control systems utilizing analytic redundancy. The systems are used for controlling a dynamic device, preferably a flight vehicle. The systems include a processing structure which controls the operation of the systems. In operation, the processing structure transforms sensed dynamic device control criteria into a linear time invariant coordinate system, determines an expected response for the device according to the transformed control criteria, compares the expected response with a measured response of the device and reconfigures the control means based on the comparison.
U.S. Pat. No. 4,355,358 to Clelford et al. discloses an adaptive flight vehicle actuator fault detection system. The system, utilizing sensors to determine the position of various operating devices within a flight vehicle, compares the positions with expected positions provided by an operating model of the flight vehicle. Thereafter, the system provides fault warnings, based upon the actual device operating conditions and the expected operating conditions obtained from the model.
U.S. Pat. No. 5,919,267 to Urnes et al. discloses a neural network fault diagnostics system and method for monitoring the condition of a host system, preferably a flight vehicle including a plurality of subsystems. The system includes a neural network means for modeling the performance of each subsystem in a normal operating mode and a plurality of different failure modes. The system also includes a comparator means for comparing the actual performance of each subsystem with the modeled performance in each of the normal and possible failure modes. Finally, the system includes a processor for determining, based on the comparisons of the comparator, the operating condition of the host system.
U.S. Pat. No. 5,070,458 to Gilmore et al. discloses a method of analyzing and predicting both airplane and engine performance characteristics. In operation, the system monitors the operation of a flight vehicle during flight and stores the monitored parameters and flight circumstances in a memory. Thereafter, during subsequent flights, the system determines and/or predicts how the flight vehicle should be operating.
U.S. Pat. No. 4,312,041 to DeJonge discloses a flight performance data computer system. In operation, the system monitors the operation of various operating characteristics of a flight vehicle during flight and provides an indication of the characteristics to the flight vehicle operator. The information provided by the system assists the operator during the flight.
U.S. Pat. Nos. 5,195,046; 5,838,261; and 6,052,056 disclose various systems for monitoring the performance of dynamic flight vehicle subsystems and providing an indication of the performance to the flight vehicle operator.
U.S. Pat. No. 3,603,948 discloses fault identification, isolation, and display device for testing a flight vehicle control system. The device senses malfunctions in selected portions of the system and provides a visual display which instantaneously identifies and isolates the malfunctioning section and memorizes the fault status of the section until the device is manually or automatically reset.
U.S. Pat. No. 3,678,256 discloses a performance and failure assessment monitor which assesses overall performance of the operation of the automatic landing mode of a flight control system for a flight vehicle. The monitor is connected to various sensors throughout the flight vehicle so that it can compare what the flight control system of the flight vehicle is accomplishing during a landing maneuver against an independent model generated within the monitor of what the flight control system should be accomplishing. The resultant comparison is displayed to the pilot as a measure of relative confidence that the landing will be accomplished properly. The monitor also includes failure verification and failure reversion control for making immediate and accurate assessments of the consequence of a failure of any component in the flight vehicle which in any way affects the ability of the flight control and flight guidance instrument systems to operate properly,
Miller Robert H.
Ribbens William B.
Brooks & Kushman P.C.
Camby Richard M.
The Regents of the University of Michigan
LandOfFree
Method and system for detecting a failure or performance... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and system for detecting a failure or performance..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and system for detecting a failure or performance... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3167925