METHOD FOR MONITORING A DEGRADATION IN AN ON-BOARD DEVICE OF AN AIRCRAFT WITH AUTOMATIC DETERMINATION OF A DECISION THRESHOLD
First Claim
1. Method for monitoring a degradation in an on-board device of an aircraft, implemented by a computer, with the degree of the degradation of said on-board device being defined by an abnormality score formed on the basis of measurements of physical parameters of said on-board device, said monitoring method comprising a step of comparing an abnormality score obtained for a given flight of said aircraft with a decision threshold and a step of transmitting an alert in the event of said decision threshold being exceeded, with said decision threshold being automatically determined for a given probability of alert Pa, which corresponds to the probability of an alert being transmitted during said monitoring method when said on-board device is healthy, by means of the following steps:
- a step of computing a plurality of abnormality scores for a plurality of flights of said aircraft without degradation so as to obtain a distribution of the probability density of the abnormality score, with the distribution being specific to the physical nature of the on-board device;
a step of adjusting the distribution using a non-parametric estimator of the probability density so as to obtain a continuous adjusted distribution function;
a step of computing a continuous adjusted division function on the basis of the continuous adjusted distribution function; and
a step of reading the antecedent of the continuous adjusted division function for a given value, said antecedent corresponding to said decision threshold, wherein said given value is 1−
Peac, with Peac being the elementary probability of exceeding the threshold for confirmation per k exceedances of threshold during n consecutive flights and being obtained by the following formula;
Peac=B−
1(k,n−
k+1)(Pa)wherein B−
1(k,n−
k+1) is an inverse Beta division function of parameters k and n−
k+1 and Pa is said probability of alert.
3 Assignments
0 Petitions
Accused Products
Abstract
A method for monitoring a degradation in an on-board device of an aircraft, comprising a step of comparing an abnormality score obtained for a given flight of said aircraft with a decision threshold (S) and a step of transmitting an alert in the event of said decision threshold (S) being exceeded, said decision threshold (S) being automatically determined for a given probability of alert Pa, corresponding to the probability of an alert being transmitted during the monitoring method when said on-board device is healthy, by means of a plurality of steps, including a step of reading the antecedent of the continuous adjusted division function for the value 1−Peac, said antecedent corresponding to said decision threshold (S), with Peac being the elementary probability of exceeding the threshold for confirmation per k exceedances of threshold during n consecutive flights and being a function of the probability of alert Pa.
17 Citations
8 Claims
-
1. Method for monitoring a degradation in an on-board device of an aircraft, implemented by a computer, with the degree of the degradation of said on-board device being defined by an abnormality score formed on the basis of measurements of physical parameters of said on-board device, said monitoring method comprising a step of comparing an abnormality score obtained for a given flight of said aircraft with a decision threshold and a step of transmitting an alert in the event of said decision threshold being exceeded, with said decision threshold being automatically determined for a given probability of alert Pa, which corresponds to the probability of an alert being transmitted during said monitoring method when said on-board device is healthy, by means of the following steps:
-
a step of computing a plurality of abnormality scores for a plurality of flights of said aircraft without degradation so as to obtain a distribution of the probability density of the abnormality score, with the distribution being specific to the physical nature of the on-board device; a step of adjusting the distribution using a non-parametric estimator of the probability density so as to obtain a continuous adjusted distribution function; a step of computing a continuous adjusted division function on the basis of the continuous adjusted distribution function; and a step of reading the antecedent of the continuous adjusted division function for a given value, said antecedent corresponding to said decision threshold, wherein said given value is 1−
Peac, with Peac being the elementary probability of exceeding the threshold for confirmation per k exceedances of threshold during n consecutive flights and being obtained by the following formula;
Peac=B−
1(k,n−
k+1)(Pa)wherein B−
1(k,n−
k+1) is an inverse Beta division function of parameters k and n−
k+1 and Pa is said probability of alert.- View Dependent Claims (2, 3, 4, 5, 6, 7, 8)
-
-
4. Method according to claim 1, wherein the probability of alert Pa is obtained by the following formula:
-
5. Method according to claim 4, comprising:
-
a step of implementing said method for monitoring a degradation in said on-board device, with the previously determined decision threshold (S), for a plurality of abnormality scores formed for a plurality of flights of said aircraft with degradation so as to deduce a probability of detecting an a posteriori degradation (1−
β
)a posteriori;a step of determining a new probability of alert Pa as a function of the probability of detecting an a posteriori degradation (1−
β
)a posteriori;a step of determining the elementary probability of exceeding the threshold Peac on the basis of the new probability of alert Pa; and a step of determining a new decision threshold (S) that is refined on the basis of the new elementary probability of exceeding the threshold Peac.
-
-
6. Method according to claim 1, comprising a step of predetermining values of k and n.
-
7. Method according to claim 6, wherein the value of n is selected before that of k and the value of k is selected as a function of that of n.
-
8. Method according to claim 7, wherein the value of n is selected in order to maximise the detection probability.
Specification