Abnormality diagnosis filter generator
First Claim
Patent Images
1. An abnormality diagnosis filter generator determining values of N and K for an abnormality diagnostic logic which makes a diagnosis N (an integer of 2 or greater) times with respect to each diagnosis target by using observation values collected from the diagnosis targets, and generates a diagnosis result showing that the diagnosis target is abnormal if the diagnosis target is judged to be abnormal K (an integer of N or less) or more times when making the diagnosis N times, comprising:
- a first storage configured to store diagnosis result data showing whether each of the diagnosis targets is abnormal or normal with respect to each of a plurality of first times;
a second storage configured to store inspection result data showing whether each of the diagnosis targets is abnormal or normal, the inspection result data being obtained by performing maintenance inspection on the diagnosis targets with respect to at least one second time;
a diagnostic logic bias calculator configured to;
(A-1) generate false detection reliability feedback data showing a probability that each diagnosis result of the diagnosis result data is abnormal due to false detection, and overlooking reliability feedback data showing a probability that each diagnosis result of the diagnosis result data is abnormal due to overlooking, the false detection reliability feedback data and the overlooking reliability feedback data being generated depending on a time difference between each of the first times of the diagnosis result data and the second time of the inspection result data;
(A-2) update, based on the false detection reliability feedback data, hyperparameters of a statistical model which includes, as the hyperparameters, a common diagnosis performance β
fp concerning false detection, and a standard deviation σ
fp of an individual diagnosis performance γ
fp concerning false detection, and calculate, from the updated hyperparameters, an average false detection rate PFP showing an average probability of false detection, and a bias level of false detection MFP; and
(A-3) update, based on the overlooking reliability feedback data, hyperparameters of a statistical model which includes, as the hyperparameters, a common diagnosis performance β
fn concerning overlooking, and a standard deviation σ
fn of an individual diagnosis performance γ
fn concerning overlooking, and calculate, from the updated hyperparameters, an average overlooking rate PFN showing an average probability of overlooking, and a bias level of overlooking MFN; and
an operational parameter determiner configured to;
(B-1) calculate an expected average false detection rate of the abnormality diagnostic logic from a heterogeneity statistical model defined by the average false detection rate, the bias level of false detection, and the values of N and K, and calculate an expected average overlooking rate of the abnormality diagnostic logic from a heterogeneity statistical model defined by the average overlooking rate, the bias level of overlooking, and the values of N and K, each calculation being performed with respect to each combination of the values of N and K;
(B-2) calculate an optimization metric from the expected average false detection rate and the expected average overlooking rate; and
(B-3) select a pair of N and K so that the optimization metric becomes a predetermined one of minimum or maximum.
1 Assignment
0 Petitions
Accused Products
Abstract
Provided is an apparatus determining values of N and K for an abnormality diagnostic logic which makes a diagnosis N times for each diagnosis target by using observation values collected therefrom, and generates a diagnosis result showing that the diagnosis target is abnormal if the diagnosis target is judged to be abnormal K or more times. A calculator calculates average false detection rate PFP, average overlooking rate PFN, bias level of false detection MFP, and bias level of overlooking MFN, based on diagnosis result data and inspection result data. A determiner calculates an optimization metric for each combination of values N and K by using the average false detection rate, the bias level of false detection, the average overlooking rate, and the bias level of overlooking, and selects a pair of N and K by which the optimization metric becomes minimum or a threshold value or less.
10 Citations
5 Claims
-
1. An abnormality diagnosis filter generator determining values of N and K for an abnormality diagnostic logic which makes a diagnosis N (an integer of 2 or greater) times with respect to each diagnosis target by using observation values collected from the diagnosis targets, and generates a diagnosis result showing that the diagnosis target is abnormal if the diagnosis target is judged to be abnormal K (an integer of N or less) or more times when making the diagnosis N times, comprising:
-
a first storage configured to store diagnosis result data showing whether each of the diagnosis targets is abnormal or normal with respect to each of a plurality of first times; a second storage configured to store inspection result data showing whether each of the diagnosis targets is abnormal or normal, the inspection result data being obtained by performing maintenance inspection on the diagnosis targets with respect to at least one second time; a diagnostic logic bias calculator configured to; (A-1) generate false detection reliability feedback data showing a probability that each diagnosis result of the diagnosis result data is abnormal due to false detection, and overlooking reliability feedback data showing a probability that each diagnosis result of the diagnosis result data is abnormal due to overlooking, the false detection reliability feedback data and the overlooking reliability feedback data being generated depending on a time difference between each of the first times of the diagnosis result data and the second time of the inspection result data; (A-2) update, based on the false detection reliability feedback data, hyperparameters of a statistical model which includes, as the hyperparameters, a common diagnosis performance β
fp concerning false detection, and a standard deviation σ
fp of an individual diagnosis performance γ
fp concerning false detection, and calculate, from the updated hyperparameters, an average false detection rate PFP showing an average probability of false detection, and a bias level of false detection MFP; and(A-3) update, based on the overlooking reliability feedback data, hyperparameters of a statistical model which includes, as the hyperparameters, a common diagnosis performance β
fn concerning overlooking, and a standard deviation σ
fn of an individual diagnosis performance γ
fn concerning overlooking, and calculate, from the updated hyperparameters, an average overlooking rate PFN showing an average probability of overlooking, and a bias level of overlooking MFN; andan operational parameter determiner configured to; (B-1) calculate an expected average false detection rate of the abnormality diagnostic logic from a heterogeneity statistical model defined by the average false detection rate, the bias level of false detection, and the values of N and K, and calculate an expected average overlooking rate of the abnormality diagnostic logic from a heterogeneity statistical model defined by the average overlooking rate, the bias level of overlooking, and the values of N and K, each calculation being performed with respect to each combination of the values of N and K; (B-2) calculate an optimization metric from the expected average false detection rate and the expected average overlooking rate; and (B-3) select a pair of N and K so that the optimization metric becomes a predetermined one of minimum or maximum.
-
-
2. An abnormality diagnosis filter generator determining values of N and K for an abnormality diagnostic logic which makes a diagnosis N (an integer of 2 or greater) times with respect to each diagnosis target by using observation values collected from the diagnosis targets, and generates a diagnosis result showing that the diagnosis target is abnormal if the diagnosis target is judged to be abnormal K (an integer of N or less) or more times when making the diagnosis N times, comprising:
-
a first storage configured to store diagnosis result data showing whether each of the diagnosis targets is abnormal or normal with respect to each of a plurality of first times; a second storage configured to store inspection result data showing whether each of the diagnosis targets is abnormal or normal, the inspection result data being obtained by performing maintenance inspection on the diagnosis targets with respect to at least one second time; a feedback data generator configured to generate false detection reliability feedback data showing a probability that each diagnosis result of the diagnosis result data is abnormal due to false detection, and overlooking reliability feedback data showing a probability that each diagnosis result of the diagnosis result data is abnormal due to overlooking, the false detection reliability feedback data and the overlooking reliability feedback data being generated depending on a time difference between each of the first times of the diagnosis result data and the second time of the inspection result data, based on the diagnosis result data and the inspection result data; a diagnosis performance calculator configured to; obtain β
fp, σ
fp, and an expected value E(γ
fp) of the γ
fp by solving a statistical model based on maximum likelihood estimation or Markov chain Monte Carlo methods using the false detection reliability feedback data, the statistical model defining posterior distribution of the β
fp, the γ
fp, and the σ
fp which are hyperparameters, the β
fp being a common diagnosis performance concerning false detection, and the σ
fp being a standard deviation of an individual diagnosis performance γ
fp concerning false detection; andobtain β
fn, σ
fn, and an expected value E(γ
fn) of the σ
fn by solving a statistical model based on the maximum likelihood estimation or the Markov chain Monte Carlo methods using the overlooking reliability feedback data, the statistical model defining posterior distribution of the β
fn, the γ
fn, and the σ
fn which are hyperparameters, the β
fn being a common diagnosis performance concerning overlooking, and the σ
fn being a standard deviation of an individual diagnosis performance γ
fn concerning overlooking;a sample data generator configured to generate false detection sample data concerning the false detection based on the β
fp, the σ
fp, and the E(γ
fp), the false detection sample data showing a frequency of false diagnosis when the diagnosis target is diagnosed N times, and to generate, overlooking sample data concerning the overlooking based on the β
fn, the σ
fn, and the E(γ
fn), the overlooking sample data showing a frequency of false diagnosis when the diagnosis target is diagnosed N times;a bias calculator configured to calculate a average false detection rate PFP showing an average probability of false detection, an average overlooking rate PFN showing an average probability of overlooking, bias level of the false detection MFP, and bias level of the overlooking MFN, the calculation being performed based on the false detection sample data and the overlooking sample data; a rate calculator configured to calculate an expected average false detection rate of the abnormality diagnostic logic from a heterogeneity statistical model defined by the average false detection rate, the bias level of false detection, and the values of N and K, and to calculate an expected average overlooking rate of the abnormality diagnostic logic from a heterogeneity statistical model defined by the average overlooking rate, the bias level of overlooking, and the values of N and K, each calculation being performed with respect to each combination of the values of N and K; a metric calculator configured to calculate an optimization metric from the expected average false detection rate and the expected average overlooking rate; and an operational parameter optimization unit configured to select the pair of N and K so that the optimization metric becomes a predetermined one of minimum or maximum. - View Dependent Claims (3, 4, 5)
-
Specification
-
- Resources
Litigation Campaign Assessment
Thank you for your request. You will receive a custom alert email when the Litigation Campaign Assessment is available.
×
-
Current AssigneeKabushiki Kaisha Toshiba (Toshiba Corporation)
-
Original AssigneeKabushiki Kaisha Toshiba (Toshiba Corporation)
-
InventorsSuyama, Akihiro, Sato, Makoto, Yonezawa, Minoru
-
Primary Examiner(s)Chaki, Kakali
-
Assistant Examiner(s)Wu, Fuming
-
Application NumberUS13/071,064Publication NumberTime in Patent Office1,090 DaysField of SearchNoneUS Class Current706/12CPC Class CodesG05B 23/0254 based on a quantitative mod...
