Device for predictive diagnosis of the current quality of technical production from a technical installation, particularly of the current quality of welding points of a spot welding robot

• US 6,281,465 B1
• Filed: 06/02/1999
• Issued: 08/28/2001
• Est. Priority Date: 10/02/1996
• Status: Expired due to Fees

• Current Assignee: Siemens AG
• Original Assignee: Siemens AG
• Inventors: Schubert, Martina, Müller, Klaus-Dieter
• Primary Examiner(s): Shaw, Clifford C.
• Application Number:
  US09/269,829
• Time in Patent Office: 818 Days
• Field of Search: 219/109, 219/110, 700/29, 700/30, 700/31
• US Class Current: 219/109
• CPC Class Codes: B23K 11/115 by means of two electrodes placed opposite one another on both sides of the welded parts
• CPC Class Codes: B23K 11/314 Spot welding guns, e.g. mounted on robots
• CPC Class Codes: B23K 31/125 Weld quality monitoring
• CPC Class Codes: G05B 19/41875 characterised by quality surveillance of production
• CPC Class Codes: G05B 2219/32018 Adapt process as function of results of quality measuring until maximum quality
• CPC Class Codes: G05B 2219/32187 Correlation between controlling parameters for influence on quality parameters
• CPC Class Codes: G05B 2219/32194 Quality prediction
• CPC Class Codes: G05B 23/0243 model based detection method, e.g. first-principles knowledge model
• CPC Class Codes: Y02P 90/02 Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Abstract

A device for predictively diagnosing the prevailing quality of the technical work result of a technical installation, in particular, the prevailing quality of the welding spots of a welding robot. The device includes a device for cyclically acquiring sets of measured values, whose values influence the desired quality of the work result of the system ,an installation model which uses sets of measured values of the installation to simulate an actual value for the prevailing quality of the work result of the installation, and a device at least for parameterizing the installation model. The parameterizing device includes a data base for storing selected sets of measured values and associated characteristic values, which are a measure for the quality of the work result of the installation, and the device generating and/or optimizing at least the parameters of the installation model by successively evaluating the sets of measured values and the associated characteristic quality values by an iterative optimization.

Claims

1. A device for predictively diagnosing a prevailing quality of a technical work result of a technical installation, comprising:

a first device cyclically acquiring sets of measured values during operation of the technical installation, the active values influencing a desired quality of the technical work result of the technical installation;

an installation model simulating an actual value for the prevailing quality of the technical work result of the technical installation using at least one active one of the acquired sets of measured values; and

a second device parameterizing the installation model, the second device including a database for storing selected ones of the acquired sets of measured values, and characteristic quality values determined as an examination result, the characteristic quality values being a measure for the respective quality of the technical work result of the technical installation, and an evaluator successively evaluating the measured values and the characteristic quality values by an iterative optimization, the evaluator further determining parameters of the installation model.

2. The device according to claim 1, wherein the second device includes a neural network.

3. The device according to claim 1, wherein the technical installation is a spot-welding robot, the technical work result is welding spots of the spot-welding robot, and the acquired sets of measured values include at least active values of a welding current and active values of a pressure of the electrode-holder of the spot-welding robot.

4. The device according to claim 3, wherein the measured value of the pressure of the electrode-holder of the spot-welding robot is simulated by measuring an air pressure in a pneumatic drive of the electrode holder.

5. The device according to claim 3, wherein the acquired sets of measured values further include values which characterize a pressure buildup at the electrode holder of the spot-welding robot.

6. The device according to claim 3, wherein the acquired sets of measured values further include values which characterize a curve of the welding current of the spot-welding robot.

7. The device according to claim 3, wherein the acquired sets of measured values further include values which characterize a rate time in a curve of the welding current of the spot-welding robot.

8. The device according to claim 3, wherein the acquired sets of measured values further include values which characterize a time lag in a curve of the welding current of the spot-welding robot.

9. The device according to claim 3, wherein the acquired sets of measured values further include values which characterize a current time in a curve of the welding current of the spot-welding robot.

10. The device according to claim 3, wherein the acquired sets of measured values further include values which characterize a temperature of a cooling medium for welding helmets at the electrode holder of the spot-welding robot.

11. The device according to claim 3, wherein the acquired sets of measured values further includes values which characterize at least one of an inlet and a return temperature of a cooling medium for welding helmets at the electrode holder of the spot-welding robot.

12. The device according to claim 1, wherein an actual quality value is provide at an issue output of the installation model, the actual quality value being compared to a predetermined quality setpoint value in a comparator for determining a deviation in quality.

13. The device according to claim 12, further comprising:

a processing unit, the deviation in quality being fed to the processing unit for at least one i) triggering a signaling action, ii) signaling the deviation in quality, and iii) deriving actuating signals for the technical installation.

14. A method for predictively diagnosing a prevailing quality of a technical work result of a technical installation, comprising the steps of:

cyclically acquiring sets of measured values during an operation of the technical installation, the measured values including active values which influence a desired quality of the technical work result of the technical installation;

simulating an actual value for the prevailing quality of the technical work result of the technical installation using at least one of the sets of measured values; and

parameterizing an installation model, including the steps of

storing in a database selected sets of the measured values, and characteristic quality values determined as an examination result, the characteristic quality values being a measure for the respective quality of the technical work result of the technical installation, and

successively evaluating by an iterative optimization the sets of measured values and associated ones of the characteristic quality values, and determining parameters of the installation model as a function of the successive evaluations.