Inspection method, inspection apparatus, and facility diagnosis unit

• US 20050114081A1
• Filed: 09/20/2004
• Published: 05/26/2005
• Est. Priority Date: 09/22/2003
• Status: Active Grant

• Current Assignee: Omron Corporation
• Original Assignee: Omron Corporation
• Inventors: Fukui, Ikuma
• Granted Patent: US 7,079,979 B2
• Time in Patent Office: Days
• US Class Current: 702/182
• CPC Class Codes: G01N 29/14 using acoustic emission techniques echo of particles G01N29/046; measuring mechanical vibrations or acoustic waves in solids in general G01H1/00
• CPC Class Codes: G01N 29/30 Arrangements for calibrating or comparing, e.g. with standard objects
• CPC Class Codes: G01N 29/4445 Classification of defects

Abstract

To inspect a status of an inspection object by using an inspection apparatus for extracting amount of characteristic to an inputted waveform signal and determining a status on the basis of the extracted amount of characteristic. Specifically, the inspection apparatus uses a normal knowledge that is generated on the basis of only the data of a normal status at an initial stage to determine whether or not the status of the inspection object complies with the normal status (S1, S2). The inspection apparatus generates an abnormal kind knowledge by abnormal kind on the basis of the data of an abnormal status that are collected in accordance with repeat of the determination (S3 to S6), and then, determines the status by using the normal knowledge and the abnormal kind knowledge (S7). If a defective product is eliminated due to change of design or the like, determining the status on the basis of a new normal knowledge, a mass production shifts to a stable period of mass production (S8 to S10).

Claims

1. An inspection method using an inspection apparatus for extracting amount of characteristic to an inputted waveform signal and determining a status of an inspection object on the basis of the extracted amount of characteristic, the method comprising the steps of:

determining whether or not the status complies with a normal status by using a normal knowledge that is generated on the basis of only data of the normal status at an initial stage; and

determining the status by using the normal knowledge and an abnormal kind knowledge, the abnormal kind knowledge being generated by abnormal kind on the basis of data of the abnormal status that are collected in accordance with repeat of the normal status determination.

2. The inspection method according to claim 1,

wherein, as a result of determination of the status by using the normal knowledge and the abnormal kind knowledge, if the abnormal kind is not detected on the basis of the abnormal kind knowledge, deleting the abnormal kind knowledge of the abnormal kind, the determination processing is carried out with the abnormal kind knowledge of the abnormal kind deleted.

3. An inspection apparatus for extracting amount of characteristic to an inputted waveform signal and determining a status of an inspection object on the basis of the extracted amount of characteristic, having:

a first mode for determining whether or not the status complies with a normal status by using a normal knowledge that is generated on the basis of only data of the normal status of the inspection object; and

a second mode for determining whether or not the status is normal and whether or not the status complies with a prescribed abnormal kind by using the normal knowledge that is generated on the basis of the data of the normal status of the inspection object and an abnormal kind knowledge that is generated on the basis of data of the prescribed abnormal kind;

wherein, at the initial stage when the abnormal kind is not specified, the inspection apparatus determines the status in the first mode; and

the inspection apparatus determines the status in the second mode at a prescribed timing after the initial stage.

6. The inspection apparatus according to claim 3, comprising:

a dummy abnormal data generating device for generating the dummy abnormal data by processing the data of the normal status;

wherein the inspection apparatus carries out the normal status determination processing by using the dummy abnormal data that is generated by the dummy abnormal data generating device so as to enable to evaluate the normal knowledge.

9. The inspection apparatus according to claim 3,

wherein a threshold to determine whether or not the status is the normal status in the normal knowledge is set at a value that a cost caused by discarding the inspection object that is determined not to be normal and a cost expensed to modify the inspection object that is determined not to be normal into normal are balanced.

10. The inspection apparatus according to claim 9,

wherein the threshold to determine whether or not the status is the normal status in the normal knowledge is set by a registration device having information of a quality function limit, discard cost information, and adjustment cost information registered therein;

a loss function calculating device for calculating a loss function on the basis of the information of the registration device; and

a threshold calculating device for calculating a threshold on the basis of the loss function of the loss function calculating device.

4. An inspection apparatus for extracting amount of characteristic to an inputted waveform signal and determining a status of an inspection object on the basis of the extracted amount of characteristic, comprising:

a normal status determining device for determining whether or not the status complies with a normal status by using a normal knowledge that is generated on the basis of only data of the normal status; and

an abnormal kind determining device for determining whether or not the status complies with a prescribed abnormal kind by using an abnormal kind knowledge that is generated on the basis of data of the prescribed abnormal kind;

wherein, at the initial stage when the abnormal kind is not specified, the inspection apparatus determines the status only by the normal status determining device; and

the inspection apparatus determines overall the status by operating the normal status determining device and the abnormal kind determining device at a prescribed timing after the initial stage.

5. The inspection apparatus according to claim 4,

wherein the abnormal kind determining device is additionally incorporated into the apparatus later after the initial stage.

7. The inspection apparatus according to claim 4,

wherein, the normal status determining device calculates a vector making plural amounts of characteristic into one unit to decide the status on the basis of a distance between the vectors.

8. The inspection apparatus according to claim 4,

wherein at least one of the normal status determining device and the abnormal kind determining device calculates a vector making plural amounts of characteristic into one unit to decide the status on the basis of a distance between the vectors.

11. A facility diagnosis unit for extracting amount of characteristic to an inputted waveform signal and determining a status of a facility on the basis of the extracted amount of characteristic, comprising:

a normal status determining device for determining whether or not the status complies with a normal status by using a normal knowledge that is generated on the basis of only data of the normal status of the facility that is a diagnosis object; and

an abnormal kind determining device for determining whether or not the status complies with a prescribed abnormal kind by using an abnormal kind knowledge that is generated on the basis of data of the prescribed abnormal kind;

wherein, at the initial stage when the abnormal kind is not specified, the facility diagnosis unit determines the status only by the normal status determining device; and

the facility diagnosis unit determines overall the status of the facility by operating the normal status determining device and the abnormal kind determining device at a prescribed timing after the initial stage.