Data processing: measuring – calibrating – or testing – Measurement system – Performance or efficiency evaluation
Reexamination Certificate
2001-10-16
2004-03-09
Barlow, John (Department: 2863)
Data processing: measuring, calibrating, or testing
Measurement system
Performance or efficiency evaluation
C702S176000, C702S187000, C708S139000, C708S142000
Reexamination Certificate
active
06704688
ABSTRACT:
This invention relates to turbines and generator machinery. More particularly, it relates to a digital data entry method for recording observed and measured condition of machine components during inspection.
BACKGROUND OF THE INVENTION
The structural integrity of critical engineering components is often dependent on the capability and reliability of the inspection procedures used in acceptance after manufacture and in life-cycle management while in service. Inspection reliability is often expressed in terms of the probability of detection. Although the probability of detection is an end to end measure of the performance of an inspection process at a specific time, it reflects the capability of a procedure and can only be viewed as a process reliability measure if the inspection process is under control.
Inspection process control is, in turn, widely viewed as an issue of human skills and skills variables. It is commonly assumed that automation of a process produces a reliable process. Indeed, variances in the performance of any automated process is a function of the variances of all critical process parameters that are inherent to the process. Inspection of engine components after periods of operation is important not only to effectively service the worn components but also to redesign or modify the components to increase component. Such an inspection procedure, in the past, has been largely conducted by presenting an inspector with paper schematics of parts to manually mark up the damaged sections of the inspected parts.
Manually reviewing the inspection charts for a machine having thousands of components is not only time consuming but also prone to errors. Further, “free field” text entries by an inspector to describe a part allows an inspector to describe a component using any language or terminology of inspector's choice. A database search conducted on the assembled inspection data would be reduced to a so-called “string search” as a query would be specifically searching for familiar words used by the inspector.
When the search results obtained by the reduced result set are compared with other components/parts inspected by a second inspector who chose to use a different set of words to record a similar fault in a component, the outcome of the comparison process would lead to erroneous results; i.e., the same observed condition would be reported using different text by one inspector than it would by another inspector. In fact, the challenge of mastering such a lexicon in a digital database would only be a first challenge. The second would be to assure that the inspection was as complete as necessary.
Finally, even resolving differences in textual language would still not provide the quantitative measures of those part conditions. Engineers evaluating part conditions need fine discrimination among intensities of part condition. For example, choices among “usual wear”, “badly worn”, and “hardly worn” are not as valuable as “wear depth” and “30 mils”. The need to report measures of conditions as such “continuous” values is highly valued. Moreover, such values are more easily searched in a database and distributions of conditions like “wear depth” can be described mathematically in histograms showing counts of how many parts exhibited wear to various wear depths.
Also, since the inspection charts are not stored in a searchable database, there appears to be no method to productively search for identified defects of machine components. Even if the inspected charts are scanned into bit-mapped images and stored in a database, currently there appears to be no current capability to search the bit-mapped images for identified defects.
Another problem with the one-by-one manual inspection on schematic forms is that the inspector must manually recreate defects that are identical from part-to-part in a set. Thus, there is a need to digitize the data entry procedure and provide a capability to search the recorded information for future use.
Automating the inspection process and eliminating factors leading to human error would produce consistent inspection results. Further, automating the inspection process and storing the status of the inspected components in a database system which is capable of being searched would eliminate the manual exercise of analyzing the observed information by a knowledgeable reader. Greater confidence can be established by automating the entry of inspected information and digitally recording the same while providing a capability to search the recorded information for future uses.
BRIEF SUMMARY OF THE INVENTION
Accordingly, the present invention provides a method to obviate manually capturing inspection data of machine components by facilitating the capture of inspection data digitally so that captured inspection data may be analyzed as a complete set, and also in conjunction with data from similar sets.
The present invention relates to a method which facilitates digitization of data entry procedure by providing an inspector with a computer system having a user interface with a drop-down menu. The drop-down menu comprises a plurality of data entry selection tables listing various machine components, and respective areas/features to be inspected on each component, and conditions to be measured and the units of value they are to be reported in. Specifically, a first table includes a list having various machine components/parts of a machine under inspection, and the inspector may select a specific component class (referred to herein as part-type) from the list of provided part-types. For example, for a given model of gas turbine engine, a combustion liner and a first stage bucket would each be selectable from a menu list of distinct part-types. The user interface provides another dropdown menu list for selecting a predefined area, or feature, to be inspected. Finally the inspector enters an observed condition of the component as a continuous value in units specified in the condition.
For a given part type, the maximum set of combinations is approximately the product of the total number of prescribed areas or features and the number of allowed conditions. The prescribed set of combinations includes those that are identified, preferably by an engineering group, to have physical significance. A subset of the prescribed set is a “mandatory” list of combinations that engineers prefer to be reported for every inspected part.
Each individual continuous data entry, for an individual serialized part of a given part-type, is preferred to have an associated predefined part area (e.g., “liner body section”) or predefined feature (e.g., “circumferential weld on aft section” and a predefined condition (e.g., “maximum crack length (in″)”) associated with the reported value (e.g., “5.5”) Conditions may be reported in units of length, counts, color codes (to specify color appearance), etc.
The inspection information is stored as digitized data and is searchable by a database engine to quantify issues on a part set or on a fleet-wide basis. The approach of the present method also improves inspection productivity where many defects are indeed similar on parallel parts of a given part-type exposed to the same in-service operating history.
In one aspect, the present invention provides a method of inspecting and recording inspection data of machine components, the method comprising the steps of selecting a machine component among a plurality of machine components for inspection; identifying predefined areas of the selected machine component for inspection; identifying a fixed set of conditions allowed to be reported for the select machine component; inspecting the predefined areas of the select machine component; and recording inspection data as continuous numerical values to quantify entry for subsequent querying over a chosen set of machine components.
The inspection step includes creating benchmark inspection values for the selected machine component by manually entering information into data fields; and comparing inspection data of other individual machine components wit
Aslam Sami
Woodmansee Donald Ernest
Barlow John
General Electric Company
Vo Hien
LandOfFree
Method for inspecting and recording machine component... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for inspecting and recording machine component..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for inspecting and recording machine component... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3224945