Diagnosis of networks of components, with stripwise modeling

• US 6,172,966 B1
• Filed: 01/14/1998
• Issued: 01/09/2001
• Est. Priority Date: 05/15/1996
• Status: Expired due to Fees

• Current Assignee: Dassault Electronique
• Original Assignee: Dassault Electronique
• Inventors: Taillibert, Patrick, Loiez, Etienne
• Primary Examiner(s): Olms, Douglas W.
• Assistant Examiner(s): HOM, SHICK C
• Application Number:
  US08/983,154
• Time in Patent Office: 1,091 Days
• Field of Search: 370/252, 370/241, 370/242, 370/243, 370/244, 370/245, 370/246, 364/500, 364/551.01, 364/551.02, 364/578, 364/579, 364/580, 371/22.1, 371/22.4, 371/22.5, 371/24, 371/25.1
• US Class Current: 370/252
• CPC Class Codes: G01R 31/2834 Automated test systems [ATE]; using microprocessors or computers G01R31/317 takes precedence; ATE for detection of defective computer hardware G06F11/2736
• CPC Class Codes: G01R 31/2846 using hard- or software simulation or using knowledge-based systems, e.g. expert systems, artificial intelligence or interactive algorithms
• CPC Class Codes: G06F 11/22 Detection or location of defective computer hardware by testing during standby operation or during idle time, e.g. start-up testing

Abstract

An n-tuple defines an envelope of a variable over a time interval. An input memory receives blocks of measurements made at known points of the network and converts them into initial n-tuples. In a base memory, a functional model comprises a list of components, (L_v), and the designation (Df, Dg) of at least one particular function of the measurements (L_p). Management will, repetitively, (a) match together available initial n-tuples, and at least one model applicable to them, (b) through computations on n-tuples, with the function designation (Df, Dg) of the model, and these n-tuples, obtain at least one result n-tuple (Nf, Ng), (c) apply the result n-tuple (Nf, Ng) to test for a standard expression, and (d) associate the output from the test with the list of components (L_v) of the model. Decision logic analyzes various outputs from the test, with a view to determining one or more defective components.

Claims

1. A device for aiding the diagnosis of a network of components the device comprising:

input memory storing variables, each variable representing a measurement of a physical quantity made at a known location of the network and stored in the form of n-tuples defining a time interval of the variable and an envelope of the variable over the time interval;

computing means capable of computing functions on the n-tuples;

base memory storing definitions of functional models relating to the network, each functional model including a list of components of the network and at least one particular function applicable to the n-tuples representing physical quantities and related to the components in the list of components;

test means receiving at least one n-tuples, and providing an output relating to the satisfying of a standard expression by the at least one n-tuples; and

processing means including;

first receive routine to receive, repetitively, from the input memory at least one initial n-tuples,

selection routine to select, repetitively, a designation of the definitions of the functional models stored in the base memory and corresponding to the at least one initial n-tuples,

deploy routine to call, repetitively, the computing means utilizing the at least one initial n-tuples and a particular function related to the selected designation,

second receive routine to receive, repetitively, from the computing means at least one result n-tuple,

application routine to apply, repetitively, the at least one result n-tuples to the test means,

associating routine to associate, repetitively, the output from the test means with the list of components of the selected designation, and

decision logic, capable of analyzing various outputs from the test means, to determine one or more defective components.

2. The device according to claim 1 wherein the functional models pertain solely to directly measurable quantities.

3. The device according to claim 2 wherein each measurable quantity is applicable at most once into each function designation of the functional model.

4. The device according to claim 1 wherein the envelope defined by each n-tuple is delimited as a trapezoid.

5. The device according to claim 1 further comprising a conversion operator able to transform input measurement values into at least one initial multiplet.

6. The device according to claim 5 wherein the conversion operator is designed to operate an optimization between the number of n-tuples and their span along the ordinate.

7. The device according to claim 1 further comprising means of formal computation on algebraic functions, their primitives and their derivatives, for the real numbers, and such that the computing means of computation on n-tuples deploy the means of formal computation.

8. The device according to claim 1 wherein the computing means of computation over n-tuples are designed to perform:

the composition of n-tuples, so as to segment them and/or concatenate them, thus making it possible to obtain synchronous n-tuples relating to one and the same time interval,

the algebraic computations over n-tuples rendered synchronous,

the integration of an n-tuple over a part at least of its time interval.

9. The device according to claim 1 further comprising means for receiving a map of a network to be diagnosed, and a generator of model-relations from the map and from physical laws regarding the components of the network.

10. The device according to claim 1 wherein the test means are assigned to a standard structure of differential equations able to receive one of first and second designations of first and second particular functions, and n-tuples relating to the variables of the first and second particular functions, and are designed to yield a logic output relating to the satisfying of the differential equation by the variables.

11. The device according to claim 10 wherein the computing means are capable, for a functional model, of yielding at least one first result n-tuple, for the first particular function designated by the first designation, and n-tuples of measurable quantities which the at least one first result n-tuple involves, and at least one second result n-tuple for the second particular function designated by the second designation, and n-tuples of measurable quantities which the at least one second result n-tuple involves.

12. The device according to claim 11 wherein the computing means yielding a first sequence of the at least one first result n-tuple, for the first particular function designated by the first designation, and a second sequence of the at least one second result n-tuple, for the second particular function designated by the second designation, and

the test means further includes:

an operator able to work on a pair of synchronous n-tuples respectively taken from the first and second sequences, and

an operator able to work globally on a first pair of synchronous n-tuples from third and fourth sequences and on a second pair of synchronous n-tuples from fifth and sixth sequences, which relates to a time interval whose bounds are respectively associated with the first and with the second pairs of synchronous n-tuples.

13. The device according to claim 1 further comprising means for constructing the map of a network to be diagnosed, and a generator of model-relations from the map and from physical laws regarding the components of the network.