Method of monitoring a transmission assembly of a vehicle equipped with acceleration sensors, in particular a helicopter

• US 6,016,994 A
• Filed: 07/06/1998
• Issued: 01/25/2000
• Est. Priority Date: 07/04/1997
• Status: Expired due to Term

• Current Assignee: Agusta SPA (Finmeccanica SPA)
• Original Assignee: Finmeccanica SPA
• Inventors: Maino, Bruno, Bellazzi, Alberto
• Primary Examiner(s): ELDRED, JOHN W
• Application Number:
  US09/110,622
• Time in Patent Office: 568 Days
• Field of Search: 244/17.13, 244/17.11, 244/60, 244/39, 180/337, 180/338
• US Class Current: 244/17.13
• CPC Class Codes: B64C 27/14 Direct drive between power plant and rotor hub
• CPC Class Codes: F16H 57/00 General details of gearing of screw-and-nut gearing F16H25/00; of fluid gearing F16H39/00 - F16H43/00
• CPC Class Codes: G01M 13/028 Acoustic or vibration analysis

Abstract

The monitoring method includes the steps of: acquiring a first signal from an acceleration sensor; calculating a transform in the frequency domain of the first signal to obtain an initial sequence of samples, each of the samples having a respective amplitude and a respective frequency; processing the initial sequence of samples to obtain a final sequence of samples; calculating a sixth-order moment of the final sequence of samples; and comparing the sixth-order moment with at least one predetermined threshold to detect any vibrational irregularity in the helicopter transmission assembly.

Claims

1. A method of monitoring a transmission assembly of a vehicle equipped with acceleration sensors, comprising the steps of:

acquiring a signal from an acceleration sensor mounted on the transmission assembly for detecting the acceleration of a moving member of the transmission assembly;

calculating a transform in the frequency domain of said signal to obtain a sequence of samples, each of said samples having a respective amplitude and a respective frequency;

processing said initial sequence of samples to obtain a final sequence of samples;

calculating a sixth-order moment of said final sequence of samples;

comparing said sixth-order moment with at least a first predetermined threshold; and

generating an alarm signal if said sixth-order moment exceeds said at least a first predetermined threshold.

2. A method as claimed in claim 1, characterized in that said transform is a Fourier transform.

3. A method as claimed in claim 1, characterized in that said step of processing said initial sequence of samples comprises the steps of:

acquiring a first and a second reference frequency (HC0, LC0);

modifying the amplitudes of first samples, in said initial sequence of samples (S₁), having frequencies with a first predetermined relationship with said first and second reference frequencies (HC0, LC0), to obtain an intermediate sequence of samples (S₂);

assigning each sample in said intermediate sequence of samples (S₂) a respective identification value (NORDi) according to a predetermined rule of assignment;

acquiring a reference value (R); and

modifying the amplitudes of second samples, in said intermediate sequence of samples (S₂), having identification values (NORDi) with a second predetermined relationship with said reference value (R), to obtain said final sequence of samples (S₃).

4. A method as claimed in claim 3, characterized in that said step of modifying the amplitudes of first samples in said initial sequence of samples (S₁) comprises the steps of:

zeroing the amplitudes of the samples, in said initial sequence of samples (S₁), having frequencies lower than said first reference frequency (HC0); and

zeroing the amplitudes of the samples, in said initial sequence of samples (S₁), having frequencies higher than said second reference frequency (LC0).

5. A method as claimed in claim 3, characterized in that said assigning step comprises the step of:

identifying said samples in said intermediate sequence of samples (S₂) in order of decreasing amplitude, wherein said identification values (NORDi) are whole numbers, and, commencing with the highest-amplitude sample, said samples in said intermediate sequence of samples (S₂) are assigned respective said identification values (NORDi) increasing in value from a predetermined value.

6. A method as claimed in claim 3, characterized in that said step of modifying the amplitudes of second samples in said intermediate sequence of samples (S₂) comprises the steps of:

determining the amplitude of a reference sample, in said intermediate sequence of samples (S₂), having an identification value (NORDi) with a third predetermined relationship with said reference value (R); and

altering the amplitudes of said second samples in said intermediate sequence of samples (S₂) on the basis of said amplitude of said reference sample.

7. A method as claimed in claim 6, characterized in that said step of determining the amplitude of a reference sample in said intermediate sequence of samples (S₂) comprises the step of:

determining said reference sample, in said intermediate sequence of samples (S₂), having an identification value (NORDi) equal to said reference value (R) increased by a predetermined quantity.

8. A method as claimed in claim 6, characterized in that said second samples in said intermediate sequence of samples (S₂) have respective identification values (NORDi) lower than the identification value (NORDi) of said reference sample;

and in that said step of altering the amplitudes of said second samples in said intermediate sequence of samples (S₂) comprises the step of;

equalling the amplitudes of said second samples to the amplitude of said reference sample to obtain said final sequence of samples (S₃).

9. A method as claimed in claim 1, characterized in that said step of calculating a sixth-order moment (M6A) comprises the steps of:

calculating an inverse transform in the time domain of said final sequence of samples to obtain a second signal (ET_J,K (i)) defined by a number (NJ) of samples;

calculating the mean value (ET_J,K (i)) of said second signal (ET_J,K (i));

calculating the standard deviation (σ

) of said second signal (ET_J,K (i)) and

calculating said sixth-order moment (M6A) according to the equation;

##EQU5## where M6A is said sixth-order moment, ET_J,K (i) is said second signal, NJ is said number of samples of said second signal ET_J,K (i), ET_J,K (i) is said mean value of said second signal ET_J,K (i), and σ

is said standard deviation of said second signal ET_J,K (i).

10. A method as claimed in claim 9, characterized in that said inverse transform is an inverse Fourier transform.

11. A method as claimed in claim 1, characterized in that said step of calculating a transform is preceded by a step of sampling and filtering said signal.

12. A method as claimed in claim 11, characterized in that said step of sampling and filtering comprises the steps of:

acquiring a number of synchronous samples T₁ (i) for each revolution of a shaft associated with said acceleration sensor; and

calculating an average time series T_1m (i) according to the equation;

##EQU6## where T₁ (i) is said synchronous samples, and 1 is a counter for counting the number of revolutions of said shaft.

13. A method as claimed in claim 12, characterized in that said step of calculating an average time series is preceded by a step of determining the presence of significant signal values.

14. A method as claimed in claim 12, characterized in that said step of calculating an average time series is followed by a step of determining convergence of said average time series.

15. A method as claimed in claim 1, characterized in that said step of comparing said sixth-order moment (M6A) also comprises the step of comparing said sixth-order moment (M66A) with a second threshold.