Diagnostic method for an evolutionary process

• US 5,511,004 A
• Filed: 02/01/1994
• Issued: 04/23/1996
• Est. Priority Date: 06/03/1992
• Status: Expired due to Fees

• Current Assignee: Thomson-CSF
• Original Assignee: Thomson-CSF
• Inventors: Dubost, Laurent, Heude, Jean-Noel
• Primary Examiner(s): Voeltz, Emanuel T.
• Assistant Examiner(s): KEMPER, MELANIE A
• Application Number:
  US08/182,153
• Time in Patent Office: 812 Days
• Field of Search: 364/551.01, 364/551.02, 364/550, 364/468, 364/469, 364/474.16, 364/581, 364/578, 395/924.4, 395/904, 395/912, 395/3, 395/50-51, 395/61, 395/906, 395/914
• US Class Current: 702/183
• CPC Class Codes: G01D 1/14 giving a distribution function of a value, i.e. number of times the value comes within specified ranges of amplitude
• CPC Class Codes: G05B 19/4184 characterised by fault tolerance, reliability of production system
• CPC Class Codes: G05B 23/0254 based on a quantitative model, e.g. mathematical relationships between inputs and outputs; functions: observer, Kalman filter, residual calculation, Neural Networks
• CPC Class Codes: G05B 23/0283 Predictive maintenance, e.g. involving the monitoring of a system and, based on the monitoring results, taking decisions on the maintenance schedule of the monitored system; Estimating remaining useful life [RUL] preventive maintenance, i.e. planning maintenance according to the available resources without monitoring the system G06Q10/06
• CPC Class Codes: Y02P 90/02 Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
• CPC Class Codes: Y02P 90/80 Management or planning
• CPC Class Codes: Y10S 706/904 Manufacturing or machine, e.g. agricultural machinery, machine tool
• CPC Class Codes: Y10S 706/912 Manufacturing or machine, e.g. agricultural machinery, machine tool

Abstract

The method of the invention establishes a reference state and a current state for an industrial evolutionary process from physical parameters measured on all the equipment items employing the evaluating process. These two states are compared, parameter by parameter, by resorting to fuzzy logic for classifying the quantities, and a diagnosis is established using expert rules.

Claims

1. A diagnostic method for determining deviations of an evolutionary process, comprising the steps of:

measuring a plurality of physical parameters characteristic of the evolutionary process;

calculating from the measured plurality of physical parameters a current state of the evolutionary process and a corresponding reference state of the evolutionary process;

first comparing the current state and the reference state physical parameter by physical parameter to generate relative differences between the current state and reference state for each parameter;

categorizing each parameter into one of a plurality of categories indicating a relevance of each parameter to a respective deviation in the evolutionary process;

determining for each respective deviation the relative difference between the current state and the reference state parameter by parameter which generate the respective deviation; and

diagnosing each respective deviation based on the category of each parameter and the parameter by parameter determined relative differences between the current state and the reference state which generate the respective deviation.

2. The method according to claim 1, wherein the reference state is obtained by modelling the measured physical parameters.

3. The method according to claim 1, wherein the reference state is obtained from stored values.

4. The method according to claim 1, wherein the reference state is obtained by learning.

5. The method according to any one of claims 2-4, wherein the reference state is evolutionary.

6. The method according to claim 1, wherein the current state is evaluated by classifying the measured physical parameters in domains defined with respect to reference values.

7. The method according to claim 6, wherein the domains are defined from values of corresponding physical parameters measured in an absence of anomalies and by statistically determining limits of the domains from a distribution of relative differences of magnitudes with respect to the reference values.

8. The method according to either of claims 6 or 7, wherein the step of diagnosing further comprises the step of establishing an expert rule with premises which each determine membership of a measured quantity to a domain defined with respect to the reference values of the measured physical parameter, by calculating a mean of member coefficients of each premise, by weighting the mean by a weighting function of the domain, and by comparing a result with a threshold value.

9. The method according to claim 1, wherein tendency curves are established from the measured physical parameters.

10. The method according to claim 1, wherein at least one of the plurality of physical parameters belongs to two or more families of parameters and has respective likelihood coefficients for each of the families of parameters.

11. The method according to claim 10, wherein the families of parameters have a degree of overlap.