Inspection and sorting system and method for part repair

• US 6,701,615 B2
• Filed: 03/08/2002
• Issued: 03/09/2004
• Est. Priority Date: 03/08/2002
• Status: Active Grant

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Devitt, John William, Harding, Kevin George, Corby, Nelson Raymond Jr., Hedengren, Kristina Helena Valborg
• Primary Examiner(s): Bryant, David P.
• Application Number:
  US09/683,981
• Publication Number:
  US 20030167616A1
• Time in Patent Office: 732 Days
• Field of Search: 298/891, 298/897, 294/021.8, 294/021.9, 294/020.5, 294/020.6, 294/020.7, 294/070.1, 294/070.4, 297/02, 297/03, 700/110
• US Class Current: 29/889.1
• CPC Class Codes: B23P 6/002 Repairing turbine components, e.g. moving or stationary blades, rotors, B23P6/045 takes precedence
• CPC Class Codes: B23P 6/007 using only additive methods, e.g. build-up welding
• CPC Class Codes: F01D 17/02 Arrangement of sensing elements
• CPC Class Codes: F01D 21/003 Arrangements for testing or measuring for measuring vibrations G01H
• CPC Class Codes: F01D 5/005 Repairing methods or devices
• CPC Class Codes: Y10T 29/49318 Repairing or disassembling
• CPC Class Codes: Y10T 29/49728 and by a metallurgical operation, e.g., welding, diffusion bonding, casting
• CPC Class Codes: Y10T 29/49746 by applying fluent material, e.g., coating, casting
• CPC Class Codes: Y10T 29/49748 by shaping, e.g., bending, extruding, turning, etc.
• CPC Class Codes: Y10T 29/49764 with testing or indicating
• CPC Class Codes: Y10T 29/49769 Using optical instrument [excludes mere human eyeballing]
• CPC Class Codes: Y10T 29/49771 Quantitative measuring or gauging
• CPC Class Codes: Y10T 29/53009 with comparator
• CPC Class Codes: Y10T 29/53022 with means to test work or product

Abstract

An inspection and sorting system for part repair includes at least one sensor for inspecting a part. The sensor is configured to obtain inspection data for the part. A comparison module is configured to receive the inspection data, to generate a repair profile for the part using the inspection data, and to compare the repair profile with a baseline to arrive at a repair recommendation for the part. A method includes inspecting a part with at least one sensor to obtain preliminary inspection data for the part. The method further includes generating a preliminary repair profile from the preliminary inspection data, comparing the preliminary repair profile with a baseline, and arriving at a repair recommendation for the part based on the comparison.

Claims

1. An inspection and sorting system for part repair comprising:

at least one sensor for inspecting a part, said sensor being configured to obtain a plurality of inspection data for the part; and

a comparison module configured to receive the inspection data, to implement a predictive repair model comprising a predictive algorithm to generate a baseline, to generate a repair profile for the part using the inspection data, and to compare the repair profile with the baseline to arrive at a repair recommendation for the part.

2. The inspection and sorting system of claim 1, wherein said comparison module is further configured to access a plurality of updated repair data and to modify the baseline by application of the predictive repair model to the updated repair data.

3. The inspection and sorting system of claim 2, wherein said comparison module is linked in real-time to a current cost database and the updated repair data comprises current cost data.

4. The inspection and sorting system of claim 1, wherein said comparison module comprises a neural network configured to apply the predictive algorithm to a plurality of historical repair data to generate the baseline and to compare the repair profile with the baseline to arrive at the repair recommendation for the part.

5. The inspection and sorting system of claim 4, wherein said neural network is further configured to receive and add the inspection data to the historical repair data to provide a dynamic historical repair database, and wherein said neural network is further configured to apply the predictive algorithm to the historical repair data from the dynamic historical repair database to generate the baseline.

6. The inspection and sorting system of claim 1, wherein said comparison module comprises a computer.

7. The inspection and sorting system of claim 6, wherein said at least one sensor comprises a machine vision sensor configured to obtain at least one visual image of the part, wherein the inspection data includes the visual image, and wherein the repair profile comprises a visual profile generated from the visual image.

8. The inspection and sorting system of claim 7, wherein said computer is configured to analyze the visual image to obtain the visual profile for the part.

9. The inspection and sorting system of claim 7, wherein said at least one sensor further comprises an x-ray imager configured to obtain at least one x-ray image of the part, wherein the repair profile further comprises an x-ray profile for the part generated from the x-ray image.

10. The inspection and sorting system of claim 9, wherein said x-ray imager comprises a digital x-ray imager, and the x-ray image comprises a digital x-ray image, wherein the inspection data includes the digital x-ray image, and wherein said computer is configured to analyze the x-ray image to generate the x-ray profile of the part.

11. The inspection and sorting system of claim 7, wherein said at least one sensor further comprises an infrared camera configured to obtain an infrared image of the part, and wherein the repair profile further comprises an infrared profile generated from the infrared image.

12. The inspection and sorting system of claim 11, wherein the infrared image is a digital infrared image, wherein the inspection data includes the infrared image, and wherein said computer is configured to analyze the infrared image to generate the infrared profile of the part.

13. The inspection and sorting system of claim 11, further comprising an infrared image converter configured to digitize the infrared image to supply a digital infrared image to said computer.

14. The inspection and sorting system of claim 7, wherein said at least one sensor further comprises a fluorescent penetrant inspection (FPI) module configured to obtain a plurality of FPI data for the part, wherein the repair profile further includes an FPI profile generated from the FPI data.

15. The inspection and sorting system of claim 14, further comprising quantification means for generating the FPI profile.

16. The inspection and sorting system of claim 7, wherein said at least one sensor further comprises an eddy current inspection module configured to obtain a plurality of eddy current data for the part, wherein the repair profile further includes an eddy current profile generated from the eddy current data.

17. The inspection and sorting system of claim 16, further comprising an eddy current converter configured to digitize the eddy current data to supply digital eddy current data to said computer, wherein said computer is further configured to analyze the digital eddy current data to generate the eddy current profile.

18. An inspection and sorting system for part repair comprising:

at least one sensor for inspecting a part, said sensor being configured to obtain a plurality of inspection data for the part; and

a comparison module configured to receive the inspection data, to generate a repair profile for the part using the inspection data, and to compare the repair profile with a plurality of repair data from a historical repair database to arrive at a repair recommendation for the part.

19. The inspection and sorting system of claim 18, wherein said comparison module is further configured to access repair facility data and to revise the repair recommendation based on the repair facility data.

20. The inspection and sorting system of claim 19, wherein said comparison module is linked in real-time to at least one numerical controller for the repair facility.

21. The inspection and sorting system of claim 18, wherein the part comprises an airfoil.

22. The inspection and sorting system of claim 21, wherein the part is a turbine blade.

23. The inspection and sorting system of claim 21, wherein said comparison module is further configured to arrive at a stripping and recoating recommendation for the part, if the repair recommendation is to repair the part.

24. The inspection and sorting system of claim 23, wherein said comparison module is further configured arrive at a crack repair recommendation for the part, if the repair recommendation is to repair the part.

25. The inspection and sorting system of claim 24, wherein the airfoil has a tip, and wherein said comparison module is further configured to determine a length of the tip to be removed, if the crack repair recommendation is to perform a crack repair process on the part.

26. A method comprising:

inspecting a part with at least one sensor to obtain a plurality of preliminary inspection data for the part;

generating a preliminary repair profile from the preliminary inspection data;

comparing the preliminary repair profile with a baseline; and

arriving at a repair recommendation for the part based on said comparison, said arrival at the repair recommendation comprising;

estimating a probability of repair P_est and a cost of repair C_est, based on said comparison, and

evaluating the cost and probability of repair C_est, P_est, to determine whether the preliminary inspection data indicates a likelihood of satisfying a predetermined repair value V_pre.

27. The method of claim 26, wherein the part comprises an airfoil.

28. The method of claim 27, wherein the part is a turbine blade.

29. The method of claim 26, wherein said generation, said comparison, and said arrival at the repair recommendation are performed using a computer.

30. The method of claim 29, further comprising:

grinding the part, if the repair recommendation is to repair the part;

inspecting the part with at least one sensor to obtain a plurality of grinding inspection data for the part;

generating a grinding repair profile from the grinding inspection data;

comparing the grinding repair profile with a grinding baseline; and

refining the repair recommendation based on said comparison with the grinding baseline.

31. The method of claim 30 further comprising:

building up the part by welding, if the repair recommendation is to continue repairing the part;

inspecting the part with at least one sensor to obtain a plurality of welding inspection data for the part;

generating a welding repair profile from the welding inspection data;

comparing the welding repair profile with a welding baseline; and

refining the repair recommendation based on said comparison with the welding baseline.

32. The method of claim 31, further comprising reshaping the part, if the repair recommendation is to continue repairing the part.

33. The method of claim 32, further comprising:

inspecting the part with at least one sensor to obtain a plurality of reshaping inspection data for the part;

generating a reshaping repair profile from the reshaping inspection data;

comparing the reshaping repair profile with a reshaping baseline; and

arriving at a recoating recommendation for the part based on said comparison with the reshaping baseline.

34. The method of claim 33, further comprising recoating the part to obtain a repaired part, if the recoating recommendation is to recoat the part.

35. The method of claim 34, further comprising:

inspecting the part with at least one of an x-ray imager, an infrared radiometer, and an eddy current inspection module to obtain a plurality of final inspection data for the part, said inspection being performed after said recoating operation; and

determining whether the part satisfies a repaired part profile based on the final inspection data.

36. The method of claim 33, wherein the baseline, the grinding baseline, the welding baseline and the reshaping baseline comprise a historical repair database comprising historical inspection data.

37. The method of claim 36, further comprising:

accessing repair facility data; and

revising the repair recommendation using the repair facility data.

38. The method of claim 36, further comprising:

accessing updated repair data; and

revising the repair recommendation based on the updated repair data.

39. The method of claim 33, further comprising:

obtaining the baseline using a predictive repair model;

obtaining the grinding baseline using a grinding predictive model;

obtaining the welding baseline using a welding predictive model; and

obtaining the reshaping baseline using a reshaping predictive model.

40. The method of claim 39, further comprising:

accessing repair facility data; and

revising the repair recommendation using the repair facility data.

41. The method of claim 39, further comprising:

accessing updated repair data; and

revising the repair recommendation based on the updated repair data.

42. The method of claim 33, further comprising generating the baseline, the grinding baseline, the welding baseline, and the reshaping baseline from a plurality of historical repair data using a neural network.

43. The method of claim 33, further comprising:

determining whether at least one of said grinding the part, said build up of the part, and said reshaping of the part should be modified to produce a satisfactory repaired part, if the preliminary repair profile does not satisfy the baseline;

determining whether at least one of said build up and said reshaping of the part should be modified to produce the satisfactory repaired part, if the grinding repair profile does not satisfy the grinding baseline; and

determining whether said reshaping of the part should be modified to produce the satisfactory repaired part, if the welding repair profile does not satisfy the welding baseline.

44. The method of claim 33, wherein said inspection to obtain the preliminary inspection data comprises:

inspecting the part with a machine vision sensor to obtain at least one preliminary visual image of the part, and

inspecting the part with at least one of an x-ray imager and an infrared camera to obtain a plurality of preliminary internal inspection data for the part,

wherein the preliminary inspection data comprises the preliminary visual image and the preliminary internal inspection data.

45. The method of claim 44, wherein said inspection to obtain the grinding inspection data is performed after said grinding operation and comprises:

obtaining at least one grinding visual image of the part, and

inspecting the part with at least one of the x-ray imager, the infrared camera, an eddy current inspection module, and a fluorescent penetrant inspection (FPI) module to obtain a plurality of grinding internal inspection data for the part,

wherein said grinding inspection data comprises the grinding visual image and the grinding internal inspection data.

46. The method of claim 44, wherein said inspection to obtain the grinding inspection data is performed during said grinding operation and comprises inspecting the part with one of the infrared camera and an eddy current inspection module to obtain the grinding inspection data.

47. The method of claim 44, wherein said inspection to obtain the welding inspection data is performed during said welding operation and comprises obtaining at least one welding visual image of the part, the welding inspection data comprising the welding visual image.

48. The method of claim 44, wherein said inspection to obtain the welding inspection data is performed after said welding operation and comprises:

obtaining at least one welding visual image of the part, and

inspecting the part with at least one of the x-ray imager, the infrared camera, an eddy current inspection module, and a fluorescent penetrant inspection (FPI) module to obtain a plurality of welding internal inspection data for the part,

wherein said welding inspection data comprises the welding visual image and the welding internal inspection data.

49. The method of claim 44, wherein said inspection to obtain the reshaping inspection data is performed during said reshaping operation and comprises obtaining at least one reshaping visual image of the part, the reshaping inspection data comprising the reshaping visual image.

50. The method of claim 44, wherein said inspection to obtain the reshaping inspection data is performed after said reshaping operation and comprises:

obtaining at least one reshaping visual image of the part, and

inspecting the part with at least one of the x-ray imager, the infrared camera, an eddy current inspection module, and a fluorescent penetrant inspection (FPI) module to obtain a plurality of reshaping internal inspection data for the part,

wherein said reshaping inspection data comprises the reshaping visual image and the reshaping internal inspection data.

51. An inspection and sorting system for turbine blade repair comprising:

a machine vision sensor configured to obtain at least one visual image of a turbine blade, the turbine blade comprising an airfoil;

a computer configured to receive the visual image, to generate a repair profile for the part using the visual image, to access repair facility data and updated repair data, to revise a baseline using the repair facility data and the updated repair data, and to compare the repair profile with the baseline to arrive at a repair recommendation based on the repair facility data and the updated repair data.

52. The inspection and sorting system of claim 51, further comprising:

a digital x-ray imager configured to obtain at least one digital x-ray image of the turbine blade, wherein said computer is configured to analyze the digital x-ray image to generate the repair profile using the visual and digital x-ray images.

53. The inspection and sorting system of claim 51, further comprising:

an infrared camera configured to obtain an infrared image of the turbine blade; and

an infrared image converter configured to digitize the infrared image to supply a digital infrared image to said computer, wherein said computer is configured to analyze the digital infrared image to generate the repair profile using the visual and digital infrared images.

54. The inspection and sorting system of claim 51, further comprising:

an eddy current inspection module configured to obtain a plurality of eddy current data for the turbine blade; and

an eddy current converter configured to digitize the eddy current data to supply digital eddy current data to said computer, wherein said computer is configured to analyze the digital eddy current data to generate the repair profile using the visual and digital eddy current data.

55. The inspection and sorting system of claim 51, wherein the baseline comprises a historical repair database.

56. The inspection and sorting system of claim 51, wherein the baseline is obtained using a predictive repair model.