Oscillatory servo-valve monitor

• US 5,319,296 A
• Filed: 11/04/1991
• Issued: 06/07/1994
• Est. Priority Date: 11/04/1991
• Status: Expired due to Fees

• Current Assignee: The Boeing Co.
• Original Assignee: Boeing Commercial Airplane Group
• Inventors: Patel, Manubhai C.
• Primary Examiner(s): Ro, Bentsu
• Application Number:
  US07/787,168
• Time in Patent Office: 946 Days
• Field of Search: 318/560, 318/565, 318/611, 318/623, 318/629, 318/638, 318/656, 318/657, 244/75 R, 244/76 R, 244/76 A, 244/78, 244/220, 244/221, 244/226, 244/227, 244/228
• US Class Current: 318/611
• CPC Class Codes: G05B 19/4062 Monitoring servoloop, e.g. overload of servomotor, loss of feedback or reference

Abstract

An oscillatory servo-valve fault monitor is provided for identifying faults attributable to servo-control system components. In particular, both low frequency and high frequency (oscillatory) fault responses are identified. Such oscillatory fault responses are distinguished from normal position control adjustments and transient responses. As the reaction to a fault condition is to shut down the faulty actuator and corresponding servo-control, such capability avoids inadvertent shut-downs. Independent detection paths are provided for sensing low frequency responses and high frequency responses. Each path includes a respective window comparator and up-down counter. For a given path, the window comparator defines two thresholds--in effect, a window. If the monitored signal amplitude is outside the window, then a counter or count signal is incremented. If the signal amplitude is within the window, then the counter or count signal is decremented. The time constant for counting up is less than that for counting down, so that more up counts would occur in a given time period, than would down counts for the same period. As a result, faults causing an oscillatory amplitude beyond the thresholds of the window comparator are quickly identified. If the count signals (or the count in either of the up-down counters) exceed a respective value, then the monitor indicates that a fault condition is present.

Claims

1. An apparatus for monitoring a servo-control system to detect oscillations in an error signal, comprising:

a first detection path receiving the error signal for detecting oscillations of said error signal within a first frequency range; and

a second detection path receiving the error signal for detecting oscillations of said error signal within a second frequency range;

said first detection path comprising;

(a) means for periodically accumulating a first count signal at a first rate while amplitude of said error signal is outside a first amplitude window; and

(b) means for periodically de-accumulating said first count signal at a second rate while amplitude of said error signal is within said first amplitude window, wherein said first rate is greater than said second rate.

2. The apparatus of claim 1 in which said second detection path comprises:

(a) means for periodically accumulating a second count signal at a third rate while amplitude of said error signal is outside a second amplitude window;

(b) means for periodically de-accumulating said second count signal at a fourth rate while amplitude of said error signal is within said second amplitude window, wherein said third rate is greater than said fourth rate.

3. The apparatus of claim 2 in which a fault is identified when said first count signal reaches a first level or said second count signal reaches a second level.

4. The apparatus of claim 2 in which said first amplitude window comprises a first amplitude threshold and a second amplitude threshold, said first amplitude threshold greater than said second amplitude threshold;

and in which said second amplitude window comprises a third amplitude threshold and a fourth amplitude threshold, said third amplitude threshold greater than said fourth amplitude threshold.

5. The apparatus of claim 2 in which said second detection path further comprises a band-pass filter receiving said error signal and passing an error signal component within said second frequency range, said error signal component being coupled to said accumulating means and said de-accumulating means.

6. The apparatus of claim 5 in which said second detection path further comprises means for amplifying said passed error signal component, said amplified error signal component being coupled to said accumulating means and said de-accumulating means.

7. The apparatus of claim 1 in which said accumulating means and said de-accumulating means together comprise a digital up-down counting means.

8. The apparatus of claim 1 in which said accumulating means and said de-accumulating means together comprise:

means for comparing the error signal amplitude to a first threshold amplitude and a second threshold amplitude; and

digital up-down counting means;

wherein said counting means periodically increments at said first rate while the error signal amplitude is outside said first amplitude window defined by said first and second amplitude thresholds; and

wherein said counting means periodically decrements at said second rate while the error signal amplitude is within said first amplitude window, wherein said first rate is greater than said second rate.

9. The apparatus of claim 1 in which said accumulating means comprises a count-up filter and said de-accumulating means comprises a count-down filter and wherein outputs of said count-up filter and count-down filter are combined to define said first count signal.

10. An apparatus for monitoring a servo-control system having an inner feedback loop for controlling a servo-valve and an outer feedback loop for controlling an actuator, the inner loop generating a servo-valve feedback signal, the outer loop generating an actuator position feedback signal, the apparatus detecting oscillations in a servo-valve error signal, the apparatus comprising:

means for generating the servo-valve error signal from said servo-valve feedback signal and said actuator position feedback signal;

a first detection path receiving the servo-valve error signal for detecting oscillations of said servo-valve error signal within a first frequency range; and

a second detection path receiving the servo-valve error signal for detecting oscillations of said servo-valve error signal within a second frequency range;

said first detection path comprising;

(a) means for periodically accumulating a first count signal at a first rate while amplitude of said servo-valve error signal is outside a first amplitude window; and

(b) means for periodically de-accumulating said first count signal at a second rate while amplitude of said servo-valve error signal is within said first amplitude window, wherein said first rate is greater than said second rate.

11. The apparatus of claim 10 in which said second detection path comprises:

(a) means for periodically accumulating a second count signal at a third rate while amplitude of said servo-valve error signal is outside a second amplitude window;

(b) means for periodically de-accumulating said second count signal at a fourth rate while amplitude of said servo-valve error signal is within said second amplitude window, wherein said third rate is greater than said fourth rate.

12. The apparatus of claim 11 in which a fault is identified when said first count signal reaches a first level or said second count signal reaches a second level.

13. The apparatus Of claim 10 in which the servo-control system receives a desired position command which is compared with the actuator position feedback signal to generate a first actuator position error signal input to said inner loop, wherein said servo-valve feedback signal is an "actual" servo-valve feedback signal, and wherein said servo-valve error signal generating means comprises:

means for independently generating a second actuator position error signal from the desired position command and the actuator position feedback signal;

means receiving the second actuator position error signal for simulating the servo-valve to generate a "predicted" servo-valve feedback signal; and

means for combining said "predicted" servo-valve feedback signal and said "actual" servo-valve feedback signal to generate said servo-valve error signal.

14. A method for monitoring a servo-control system to detect oscillations in an error signal, comprising the steps of:

receiving the error signal at a first detection path for detecting oscillation amplitudes of said error signal within a first amplitude window;

within said first detection path, periodically accumulating a first count signal at a first rate while amplitude of said error signal is outside said first amplitude window;

within said first detection path, periodically de-accumulating said first count signal at a second rate while amplitude of said error signal is within said first amplitude window, wherein said first rate is greater than said second rate;

receiving the error signal at a second detection path for detecting oscillations of said error signal within a second frequency range;

within said second detection path, periodically accumulating a second count signal at a third rate while amplitude of said error signal is outside a second amplitude window; and

within said second detection path, periodically de-accumulating said second count signal at a fourth rate while amplitude of said error signal is within said second amplitude window, wherein said third rate is greater than said fourth rate; and

identifying a fault when said first count signal reaches a first value or said second count signal reaches a second value.

15. The method of claim 14 further comprising the step of filtering the error signal to pass an error signal component within said second frequency range, and wherein said step of accumulating the second count signal is while amplitude of said error signal component is outside said second amplitude window, and said step of de-accumulating said second count signal is while said error signal component is within said second amplitude window.