Apparatus and method for inspecting engine components

Abstract

An apparatus for inspecting engine components and a method for inspecting the apparatus. The inspection device includes at least one controller configured to receive a set of inspection parameters based on the detection metric. A non-destructive evaluation (NDE) instrument is included for scanning a predetermined region of a surface of an engine component according to a set of inspection parameters to generate a data set. Further, the inspection apparatus includes a computer configured to apply the detection algorithm to the data set.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to U.S. Provisional Application No. 63 / 633,139, filed April 12, 2024, which is incorporated herein by reference in its entirety for all purposes. Technical Field

[0003] The present invention generally relates to an inspection apparatus and, more particularly, to a method for inspecting a surface of an engine component for anomalies. Background Art

[0004] Manufactured components require inspection before and during operation. Ultrasonic (UT) evaluation, macroetching, eddy current imaging (ECI), X-ray / computed tomography (CT), and micro X-ray fluorescence (XRF) are all used to inspect components. Each technique has limitations, making it difficult or limited in scope to detect anomalies on the surface of manufactured components. Furthermore, inspection can be a slow process that consumes significant time and resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0005] A full and enabling disclosure of the presently described technology, including the best mode thereof, to one of ordinary skill in the art is set forth in the specification with reference to the accompanying drawings, in which:

[0006] Figure 1 is a graph showing the detection metrics versus target lifetime limits for the first lifetime model and the second lifetime model.

[0007] Figure 2 is a schematic diagram of processing local datasets from the local scanning module.

[0008] Figure 3 is a flow chart illustrating a method for inspecting engine components for chemical anomalies.

[0009] Figure 4 is constructed to execute and / or instantiate Figure 1-3 A block diagram of an example programmable circuit platform illustrating example machine-readable instructions, computational structures, and / or example operations for implementing inspection apparatus and associated methods. DETAILED DESCRIPTION

[0010] Reference will now be made in detail to embodiments of the present disclosure, one or more examples of which are illustrated in the accompanying drawings. The detailed description uses numbers and letters to refer to features in the drawings. The same or similar reference numerals have been used in the drawings and the description to refer to the same or similar parts of the present disclosure.

[0011] Aspects of the present disclosure generally relate to an inspection apparatus and method for inspecting engine components, and more specifically to a micro-XRF inspection method for identifying potential chemical anomalies on the surface of a field component using the non-limiting example of a turbine engine component. The turbine engine component may be from a jet engine or a land-based heavy-duty gas turbine. For illustrative purposes, the present disclosure will be described in conjunction with the inspection of the surface of a turbine engine component. However, it should be understood that the aspects of the present disclosure described herein are not limited thereto and may have general applicability in other inspection methods and apparatuses other than turbine engine components for the manufacture of jet engines, i.e., it is contemplated that any component for a gas turbine engine may be used, whether the engine component is manufactured in the field or newly manufactured.

[0012] Some known engine inspection methods have shortcomings that are overcome by the methods and apparatus disclosed and described herein. For example, ultrasonic (UT) evaluation lacks the ability to find some chemical anomalies of interest in some engine components because the changes in particle size or other UT-detectable features within the chemical anomaly compared to the parent material are not significant enough to result in a measurable change in the speed of sound. Macroetching requires that the surface of the engine component be chemically etched, and analysis of this surface depends on the quality of the etch and the interpretation of the resulting microstructure, potentially resulting in a lower probability of detection than desired. Eddy current inspection (ECI) cannot detect some smaller chemical anomalies on very rough surfaces, such as shot-peened surfaces, and relies on the interpretation of microstructure-driven signals rather than a direct assessment of the microstructure of the inspected engine component. X-ray / computed tomography (CT) scanning can effectively find chemical anomalies with significant density differences compared to base metal. However, chemical anomalies with small density differences compared to base metal require thinner section thicknesses than those of production components, which limits its capabilities. Finally, when searching for small chemical anomalies, micro-XRF inspections have historically been very time-consuming for larger components because the pixel size must be limited to ensure sufficient signal within the potential chemical anomaly of interest. Additionally, typical micro-XRF systems are restricted to an X / Y scanning mode with limited Z motion capability, making them unsuitable for scanning complex component geometries.

[0013] The micro-XRF inspection method and apparatus discussed herein incorporate a micro-XRF head comprising an X-ray source and at least one detector as part of a robotic and turntable system, utilize computer detection algorithms to identify possible chemical anomalies on the surface of engine components, and perform a multi-scan inspection approach to reduce historically long inspection times.

[0014] Aspects of the present disclosure provide an inspection method and inspection apparatus that utilizes micro-XRF technology to reduce the inspection (scanning) time required to discover surface chemical anomalies while still meeting accuracy and recall requirements. Although the micro-XRF inspection method will be primarily utilized, it should be understood that aspects of the disclosure herein can be applied to other inspection processes.

[0015] "Precision" as used herein is a measure of the number of false positives found in a population and is given by the ratio: (true positives) / (true positives + false positives).

[0016] As used herein, a "call" is a measure of the number of false negatives found in a population and is given by the ratio: (true positives) / (true positives + false negatives).

[0017] A "life model" is a probabilistic life model that combines mission analysis, material properties, anomaly size and frequency distribution, the component region being inspected, and the probability of anomaly detection to define a life prediction for an engine component. The model is used to set a minimum required call level for the engine component. In other words, the life model is used to establish a number of acceptable false negatives.

[0018] As used herein, a "chemical anomaly" is an area where the composition of one or more elements, or any combination of elements, is statistically different from other areas. Typically, a chemical anomaly is an area that is statistically different from the intended master alloy from which the engine component will be formed. Establishing a static threshold for the variation found in the elemental composition, based on a predetermined set of inspection requirements, is a function of the detection limits associated with the scanning instrument used, and ultimately, the speed at which the component can be scanned.

[0019] As used herein, the term "high-speed scanning" means that a complete scan of an engine component surface or a predetermined portion of an engine component surface is completed as quickly as possible while still meeting a predetermined set of inspection requirements and required minimum call levels defined by a life model.

[0020] As used herein, the term "higher resolution" can refer to spatial resolution with a smaller pixel size, or to the increased signal produced by a slower scan rate. Higher resolution refers to resolution, the signal-to-noise ratio of a high-speed scan, or both resolution and the signal-to-noise ratio of a high-speed scan (higher resolution is improved by comparison).

[0021] As used herein, the term "high-resolution scan" refers to a complete local scan centered on the location of a possible chemical anomaly. A high-resolution scan has a higher resolution and is based on a set of predetermined inspection requirements. In one aspect, a high-resolution scan can be returned to after a high-speed scan is completed. In another aspect, a high-resolution scan occurs solely based on the set of inspection requirements.

[0022] As used herein, the term "inspection metrics" describes a set of inspection limits that must be met in order to successfully inspect an engine component. These include achieving the required level of accuracy and recall, detecting anomalies larger than a certain defined size, and detecting anomalies of a specific shape. Specific inspection metrics must be defined for each type of anomaly.

[0023] The term "elemental map" as used herein describes a multi-channel pixelated dataset. Each pixel represents an area at a location defined in scan coordinates and a set of values ​​that represent an integration of detector data within one or more energy windows or metadata associated with the readings occurring at a particular pixel location. Each channel represents an aggregation of detector data from one or more energy windows, which are typically centered around a specific elemental characteristic X-ray emission energy peak or combination of peaks. The energy window definitions used to create an elemental map are analysis specific and can be tuned to specific features associated with chemical anomalies of the component being inspected. An elemental map can include a dataset collected via a point or line scan.

[0024] The word "exemplary" may be used herein to mean "serving as an example, instance, or illustration." Any implementation described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other implementations. Additionally, all embodiments described herein should be considered exemplary unless expressly stated otherwise.

[0025] The singular forms "a," "an," and "the" include plural references unless the context clearly dictates otherwise. Also, as used herein, the term "group of elements" or "a group of elements" may include any number of elements, including only one.

[0026] All directional references (e.g., radial, axial, proximal, distal, up, down, upward, downward, left, right, lateral, front, back, top, bottom, above, below, vertical, horizontal, clockwise, counterclockwise, upstream, downstream, forward, backward, etc.) are used for identification purposes only to help the reader understand the present disclosure and do not create limitations, especially with respect to the position, direction, or use of the aspects of the disclosure described herein. Unless otherwise indicated, connection references (e.g., attach, couple, connect, and engage) are to be interpreted broadly and may include intermediate structural elements between sets of elements and relative movement between elements. Therefore, connection references do not necessarily infer that two elements are directly connected and in a fixed relationship to each other. The exemplary figures are for illustrative purposes only, and the sizes, positions, orders, and relative sizes reflected in the accompanying figures may vary.

[0027] Nondestructive evaluation (NDE) inspection systems are designed to inspect a variety of engine components. The engine component can be a field component that has been in service for a predetermined amount of time and is ready for inspection. The field component can be an engine component, more specifically, a jet engine component. Furthermore, the engine component can be a heavy-duty gas turbine component. Furthermore, the engine component can be a newly manufactured component, a billet, a forging, or a casting. The size and shape of the engine component dictate how the NDE technique should be manipulated to cover the intended surface area. For example, a micro-XRF-based inspection system can utilize a robotic arm to maneuver an X-ray source and X-ray detector at a constant standoff distance along the contour of a jet engine blade. The position of the robotic arm is coupled with the elemental spectrum obtained by the micro-XRF system to create an elemental map of the inspected surface (e.g., the surface of a jet engine blade, etc.). Inspection can be accomplished using a single scan speed. It is also contemplated that multiple scan speeds can be used in tandem. Different scan speeds vary inspection parameters. These inspection parameters can be micro-XRF inspection parameters, including tube excitation parameters, scan speed, spot and pixel size, standoff distance, and inspection angles limited by the component geometry. High-speed scanning is used to cover larger areas at higher speeds while achieving the necessary minimum call levels. Any potential anomalies identified can then be locally scanned with high-resolution scanning. Varying speed scans are used in conjunction to minimize or reduce inspection time while meeting the required set of detection metrics.

[0028] The predetermined set of inspection requirements R may include predetermined boundaries for normal chemical variation. This set of parameters may be input by an operator, from a database, or a combination of the operator and the database. The predetermined set of inspection requirements R may vary depending on the component being inspected. The predetermined set of inspection requirements R may include, by way of limiting example, predefined thresholds established based on historical data.

[0029] Detection Algorithm A is a computer algorithm that outputs information about the collected elemental mapping dataset. Detection Algorithm A may include the ability to assess the cleanliness of component surfaces to identify suspected chemical anomalies. For example, Detection Algorithm A may be an optimization routine in which the false positive rate is experimentally determined as a function of scanning conditions. Combinations of recall, accuracy, and inspection time can be experimentally generated. A minimum recall level is set by the lifetime model, which then provides flexibility in balancing inspection time and false positives.

[0030] Figure 1FIG2 is an example of a graph showing an inspection metric versus a target life limit, where the inspection metric is a call level. The results of a first life model 80 and a second life model 82 are represented on the call level versus target life limit graph. The first life model 80 is for a first engine component 12a, and the second life model 82 is for a second engine component 12b, which is different from the first engine component 12a. Inspection parameters for high-speed scanning are selected so that they produce an effective call level that is within the call level range required by the life model to allow the inspected engine component (as a non-limiting example, the engine components 12a and 12b described herein) to operate until its associated target life. For gas turbine engines, life models provide a method for estimating the operating life of a specific component. The life model can be updated using inspection data collected when the component is new or using inspections completed during maintenance to account for potential material anomalies. Additional information from the inspection can adjust the life prediction, increasing or decreasing the predicted life of the engine component based on the findings. In one non-limiting example, the life model is updated based on the inspection of an anomaly. Based on the known anomaly population, the life model can provide a range of call level values ​​that must be met by the inspection in order for the component to reach its target life. Using the life model that incorporates information from the known anomaly population, a high-speed scan setting is selected so that the minimum required call level is met for a scan across the entire surface of the engine component.

[0031] Each engine component 12a, 12b is associated with an original target life limit 84. During maintenance, scheduled inspections, or other necessary inspections, inspections for possible anomalies may occur. Recall levels between 0% and 100% are possible, and generally, the higher the recall level, the slower the scan speed. The results of the first and second life models 80, 82 show that there is a range of recall level values ​​that produce life predictions that exceed the target life limit 84. As a result, there is no need to inspect a component based on a recall level that exceeds the target life limit 84 because the component will be removed from service at the target life. The life models 80, 82 enable the use of inspection parameters with lower recall levels, which in turn results in faster inspection times.

[0032] The first life model 80 intersects the 100% target life limit at a call level of 52%, while the second life model 82 intersects the 100% target life limit at a call level of 90%. During the inspection of the first engine component 12a according to the first life model 80, the first minimum acceptable call level value 86 is 52%. During the inspection of the second engine component 12b according to the second life model 82, the second minimum acceptable call level value 88 is 90%. Producing inspection parameters for the 52% call level may require less runtime than producing inspection parameters for the 90% call level. The minimum acceptable call level values ​​86, 88 translate into different sets of inspection parameters S that form the scan instructions 36. 80 、S 82 The scan instructions 36 are used by the NDE instrument 24 to scan different engine components 12a, 12b. Thus, a set of scan parameters can be selected to scan a set of anomalies, and the results of the scans determine a call level for the set of scan parameters. For example, a validation study involving scanning multiple known anomalies using a selected set of scan parameters produces results that can be evaluated to determine a call level for the set of scan parameters. The call level can then be used with the set of scan parameters to inspect the engine components 12a, 12b.

[0033] Determining the inspection metrics 60 for various components, such as minimum acceptable call level values ​​86, 88, can be accomplished by conducting a validation study (also referred to as a validation study) of the inspection technique with a known set of anomalies to demonstrate that the required call level is met for each selected set of inspection parameters. The inspection rate or call level can be defined by the life model 80, 82. The methods described herein can include determining a set of inspection parameters S by minimizing the required call level and maximizing accuracy. 80 、S 82 , and reduce the amount of inspection time within the range determined by the effectiveness study and defined by the corresponding life models 80, 82. In one non-limiting example, the minimum call level value is between 70% and 100%. Therefore, the inspection metric 60 is set by the inputs of the life model. For example, in addition to the characteristic data and engine operating conditions, the inputs to the life model include values ​​defining the inspection rate or call level.

[0034] Similarly, the methods described herein can include further optimization. In addition to or in lieu of the call level, maximizing accuracy and minimizing inspection time can be pursued, as long as the call level is above the minimum acceptable call level defined by the lifespan model. The accuracy level value can be determined by optimizing between the number of false positives returned by a high-speed scan and the amount of time it takes to clear false positives by a high-resolution scan. Options for reducing the amount of inspection time can be used as a set of inspection parameters for optimization. In one non-limiting example, the optimized accuracy level value is between 60% and 90%.

[0035] The speed of the high speed scan is determined so as to achieve a critical minimum call level value, which is, as a non-limiting example, between 70% and 100%, as described herein. The critical minimum call level required for any engine component is set by the corresponding life model of the component. A second constraint on the maximum speed achievable by the high speed pass is to achieve a level of accuracy that, when combined with subsequent high resolution local scans of any potential anomalies, results in the shortest possible total scan time. As a non-limiting example, as described herein, the level of accuracy is between 60% and 90%. The required level of accuracy is determined via an iterative process. For example, the false positive rate can be determined as a function of the scanning conditions. A combination of call, accuracy, and inspection time can be generated. The life model sets the minimum call level and balances the inspection time with false positives to meet the minimum call level. The subsequent scan time for the total number of potential anomalies scanned using high resolution must be less than the incrementally slower high speed scan speed that will produce fewer false positives.

[0036] Upon completion of the high-speed scan, detection algorithm A is implemented. Anomaly detection is based on determining statistical deviations from baseline composition. In one example, detection algorithm A comprises a series of steps including: selecting a specific element map that indicates a surface or structural anomaly being screened, denoising the element map, and calculating the deviation of the intensity from the measured value of the baseline map signal to identify potential chemical anomalies. In one example, denoising uses a non-local means algorithm to reduce sensitivity to x-ray detector noise sources while maintaining spatial consistency of boundary conditions (including anomalies). The median signed deviation (MSD) is calculated for each pixel in each selected element map. The MSD results are then combined by orthogonal summation. The MSD algorithm is part of a family of M-estimator methods that provide a robust measure of variability in univariate samples. A defined threshold is then applied to the distribution in the combined sample to determine how the algorithm identifies anomalies. If a pixel or group of pixels exceeds the defined threshold, it is marked as an anomaly, otherwise it is not processed as the expected base material.

[0037] Detection algorithm A can generate readable data in the form of a first set of element maps representing the surface of the engine component. The readable data may include a first set of element maps output from a high-speed scan. In a non-limiting example, the first set of element maps may be two-dimensional. The first set of element maps may be processed to locate any possible chemical anomaly locations on the surface. In the event that a potential chemical anomaly is detected, the detection algorithm may further include marking the possible chemical anomaly locations with a marker to provide a physical record of the possible chemical anomaly locations. Detection algorithm A can utilize different detection processes. In one example, the marker is based on a direct chemical assessment detected on the surface. In another example, detection algorithm A can work by detecting differences from a baseline (as a non-limiting example, a chemical background). In another example, detection algorithm A can work by directly detecting a chemical anomaly of interest.

[0038] In one example, if the component surface satisfies a predetermined set of inspection requirements R associated with the presence of any chemical anomalies, a first elemental map without any flags is generated by the detection algorithm A. In this case, as a non-limiting example, a signal indicating that the engine component satisfies the predetermined set of inspection requirements R can be sent to an operator via a computer user interface.

[0039] Some specific examples of chemical anomalies that may occur in turbine engine components are small spots, white spots, dirty white spots, stringers, oxidation, oxycarbonitride clusters (OCNC) large enough to act as failure initiation sites, corrosion, and coating loss. Small spots are chemical anomalies that form during solidification due to interdendritic microsegregation and fluid flow of the liquid alloy. White spots, dirty white spots, and stringers are chemical anomalies caused by incomplete melting of material that falls into the liquid pool during the final melting step. Oxidation occurs when metal reacts with oxygen in the atmosphere to form oxides. Corrosion occurs when metal reacts with contaminants during combustion, causing low melting point compounds to attack the metal surface. Loss of coating occurs when chipping or flaking occurs in the applied coating, allowing the operating environment to approach the underlying substrate. All of these chemical anomalies result in a surface chemistry at the abnormal location that is statistically distinguishable from the starting baseline chemistry.

[0040] At the end of the high-speed scan, Detection Algorithm A is run. If a suspected chemical anomaly is identified, a high-resolution scan can be run locally. At the end of the high-resolution scan, Detection Algorithm A is executed again. This can be the same Detection Algorithm A used to evaluate the high-speed scan or a new algorithm. Detection Algorithm A can be used to determine if a suspected chemical anomaly is a true anomaly, or the operator can use the output of Detection Algorithm A to determine if the test is a true positive (anomaly) or a false positive (non-anomaly).

[0041] Figure 21 and 2. Example processing of a local data set 70 using detection algorithm A is shown. Detection algorithm A is applied to the local data set 70 to generate a set of element maps 72, which, in a non-limiting example, can be the readable data 44. The set of element maps 72 can be a collection of three element maps 72a, 72b, 72c, which, as a non-limiting example, are tuned to specific signatures associated with different elements. In one non-limiting example, the specific signature is specific to metallic elements. In one aspect, an enhanced map 78 is derived from the set of element maps 72, such as a denoised map of a potential chemical anomaly 64 associated with a pixel 65 determined and displayed using detection algorithm A. A specific set of detection metrics can be defined for each type of potential chemical anomaly and underlying material type.

[0042] Detection algorithm A may include determining a threshold map 74 having a flag 76 indicating the location of a group of pixels 65 (the group of pixels 65 being at least one pixel) on the enhancement map 78 exceeding a predefined threshold value based on the inspected state of the surface determined by the set of elemental maps 72. The inspected state refers to the overall chemical distribution reflected in the set of elemental maps 72 established using the surface. Possible chemical anomaly locations 64 may be associated with the presence of small spots, dirty white spots, or fine streaks on the surface. Additionally, possible chemical anomaly locations may include oxidation, corrosion, repairs, OCNCs large enough to act as failure initiation sites, loss of coatings, or localized areas that deviate from the expected base metal composition defined by the expected alloy chemical statistical signature. A subsequent determination may then be made as to whether the flag is confirmed to be a true chemical anomaly.

[0043] In one non-limiting example, the threshold map 74 can be presented alongside the set of element maps 72 associated with the possible chemical anomaly location. An operator or user can then determine whether the possible chemical anomaly location matches a potential chemical anomaly or is a false positive.

[0044] High-resolution scanning results in a slower inspection speed than that used in high-speed scanning. Limiting the area covered by the slower high-resolution scan to a predetermined area of ​​the surface reduces the total scan time. The higher resolution of the high-resolution scan enables a higher recall level and a higher precision level than can be achieved with high-speed scanning. The higher recall level and the higher precision level of the high-resolution scan provide greater confidence in determining whether the identified potential anomaly is a true positive (chemical anomaly of interest) or a false positive (chemical anomaly of interest). As described herein, detection algorithm A is tuned to minimize the number of false positives, thereby achieving a precision level between 60% and 90%, and to minimize the number of false negatives, thereby achieving a recall level between 70% and 100%.

[0045] Figure 31 is a flow chart illustrating a method 100 for inspecting an engine component (e.g., engine components 12a, 12b, etc.) as disclosed and described herein. At block 102, a set of inspection parameters is determined from a detection metric defined by a life model of the engine component. As a non-limiting example, a set of inspection parameters S is determined from the life model 80 of the engine component. 80 The life models 80, 82 and target life limits 84 of the engine components 12a, 12b are analyzed. Figure 1 As shown in the example of FIG, the intersection of the life model 80, 82 of the respective engine component 12a, 12b with the target life limit 84 provides a call level 86, 88 for the respective engine component 12a, 12b. Then, a set of inspection parameters S can be determined. 80 、S 82 To achieve the call levels of the engine components 12a, 12b, as described herein.

[0046] At block 104, based on a set of inspection parameters S 80 、S 82 Scanning at least a portion of the surface of the engine component to generate a data set, as a non-limiting example, a global data set. For example, the NDE inspection system can scan the engine component according to the inspection parameter S 80 、S 82 All or a portion of the surface of the engine component 12a, 12b is scanned at a set scan speed (e.g., based on a call level and an accuracy level associated with the call level, etc.) The data set includes scan values ​​representing image data, chemical composition, visual characteristics, etc.

[0047] At block 106, the detection algorithm A is applied to the dataset. For example, the dataset 70 generates an element map 72 from the data 44 in the dataset 70, and then a threshold map 78 is determined based on the element map 72. The element map 72 can be a set of element maps 72, including, for example, three element maps 72a, 72b, 72c, which are tuned to specific signatures associated with different elements (e.g., different metal elements).

[0048] At block 108, possible chemical anomaly locations based on the application of detection algorithm A are marked. For example, detection algorithm A may determine a threshold map 74 that includes markers 76 indicating locations in the data 44 of the element map 72 that exceed the values ​​in the threshold map 74. The marked locations 76 (e.g., threshold regions 76) may then be further inspected and / or evaluated to determine whether the anomaly is a true positive or a false positive. If the marked location 76 is a true positive, representing a chemical anomaly of interest, the engine component may be further inspected, repaired, removed, etc.

[0049] The detection algorithm A described herein may further include comparing relevant elementary signals of interest, pre-filtering signal noise, smoothing the input data, and performing statistical analysis for anomaly detection, wherein parameters are optimized based on real scan data. In another embodiment, the detection algorithm A described herein may further include comparing relevant elementary signals of interest and ground truth information to utilize machine learning or deep learning methods for anomaly detection.

[0050] Benefits associated with the inspection process disclosed herein include reducing the amount of inspection time while still locating potential anomaly locations. The goal of the inspection process for discovering potentially harmful anomalies can be defined so that it is successful only when all potential anomalies are correctly identified without false positive results. In other words, when the recall level is 100% and the accuracy level is 100%. As discussed herein, such a definition may require very time-consuming inspection parameters that involve multiple repeated inspections of the same area, or involve long duration scans to provide maximum resolution. Although this may enable all anomalies to be discovered, the time required may preclude the inspection technique from being used in an industrial environment, resulting in scrapped parts that do not contain potential anomalies. Including a life model consistent with a multi-channel scanning strategy (global high speed followed by local high resolution) allows for a reduction in the total amount of inspection time while still ensuring that the rate of anomaly detection is large enough to ensure operation to achieve the target life. The reduced amount of inspection time allows the inspection to be established so that fewer parts must be scrapped.

[0051] Figure 4 is constructed to execute and / or instantiate Figure 1-3 The block diagram of an exemplary programmable circuit platform 400 that implements exemplary machine-readable instructions, computing structures, and / or exemplary operations for implementing exemplary inspection apparatus and associated methods. The programmable circuit platform 400 may be, for example, a server, a personal computer, a workstation, a self-learning machine (e.g., a neural network), a mobile device (e.g., a cellular phone, a smart phone, an iPad, etc.) TM tablet computer), controller, Internet appliance or other computing and / or electronic device.

[0052] The programmable circuit platform 400 of the illustrated example includes programmable circuitry 412 (also referred to as processor circuitry 412). The programmable circuitry 412 of the illustrated example is hardware. For example, the programmable circuitry 412 may be implemented by one or more integrated circuits, logic circuits, FPGA microprocessors, CPUs, GPUs, DSPs, and / or microcontrollers from any desired family or manufacturer. The programmable circuitry 412 may be implemented by one or more semiconductor-based (e.g., silicon-based) devices. In this example, the programmable circuitry 412 implements the life models 80, 82 and / or the NDE inspection system, among other things.

[0053] The programmable circuit 412 of the illustrated example includes a local memory 413 (e.g., cache, registers, etc.). The programmable circuit 412 of the illustrated example communicates with a main memory including a volatile memory 414 and a non-volatile memory 416 via a bus 418. The volatile memory 414 may be comprised of synchronous dynamic random access memory (SDRAM), dynamic random access memory (DRAM), Dynamic Random Access Memory The non-volatile memory 416 may be implemented as flash memory and / or any other desired type of memory device. Access to the main memories 414, 416 of the illustrated example is controlled by a memory controller 417. In some examples, the memory controller 417 may be implemented as one or more integrated circuits, logic circuits, microcontrollers from any desired family or manufacturer, or any other type of circuitry to manage the flow of data to and from the main memories 414, 416.

[0054] The programmable circuit platform 400 of the illustrated example further includes an interface circuit 420. The interface circuit 420 may be implemented by hardware according to any type of interface standard, such as an Ethernet interface, a Universal Serial Bus (USB) interface, interface, a near field communication (NFC) interface, a peripheral component interconnect (PCI) interface, and / or a peripheral component interconnect express (PCIe) interface.

[0055] exist Figure 4 In the example shown, one or more input devices 422 are connected to the interface circuitry 420. The input devices 422 allow a user (e.g., a human user, a machine user, etc.) to input data and / or commands into the programmable circuitry 412. The input devices 422 can be implemented by, for example, an audio sensor, a microphone, a camera (still or video), a keyboard, buttons, a mouse, a touch screen, a track pad, a track ball, an isopoint device, and / or a voice recognition system.

[0056] One or more output devices 424 are also connected to the interface circuit 420 of the illustrated example. The output device 424 can be implemented, for example, by a display device (e.g., a light-emitting diode (LED), an organic light-emitting diode (OLED), a liquid crystal display (LCD), a cathode ray tube (CRT) display, an in-place switching (IPS) display, a touch screen, etc.), a tactile output device, a printer, and / or a speaker. Therefore, the interface circuit 420 of the illustrated example typically includes a graphics driver card, a graphics driver chip, and / or a graphics processor circuit (e.g., a GPU).

[0057] The interface circuitry 420 of the illustrated example also includes communication devices, such as transmitters, receivers, transceivers, modems, residential gateways, wireless access points, and / or network interfaces, to facilitate exchanging data with external machines (e.g., any type of computing device) via a network 426. Communication can be via, for example, an Ethernet connection, a digital subscriber line (DSL) connection, a telephone line connection, a coaxial cable system, a satellite system, a field-line wireless system, a cellular telephone system, an optical connection, etc.

[0058] The programmable circuit platform 400 of the illustrated example also includes one or more mass storage devices 428 for storing software and / or data. Examples of such mass storage devices 428 include magnetic storage devices (e.g., floppy disks, drives, HDDs, etc.), optical storage devices (e.g., Blu-ray discs, CDs, DVDs, etc.), RAID systems, and / or solid-state storage disks or devices such as flash memory devices and / or SSDs.

[0059] Can be Figure 1-3 The machine-executable instructions 432 implemented by the machine-readable instructions may be stored in the mass storage device 428, in the volatile memory 414, in the non-volatile memory 416, and / or on at least one removable non-transitory computer-readable storage medium (e.g., a CD or DVD).

[0060] It should be understood that any combination of geometries associated with the inspection devices discussed herein can be envisioned. The various aspects of the present disclosure discussed herein are for illustrative purposes and are not meant to be limiting. It should be understood that the disclosed designs can be used with any applicable component, including in-situ components from any engine, including jet engines, where the engine components are made of materials capable of producing chemical anomalies.

[0061] This written description uses examples to describe aspects of the invention described herein, including the best mode, and also to enable any person skilled in the art to practice aspects of the invention, including making and using any devices or systems and performing any incorporated methods. The patentable scope of the aspects of the disclosure is defined by the claims, and may include other examples that occur to those skilled in the art. Such other examples are within the scope of the claims if they have structural elements that do not differ from the literal language of the claims, or if they include equivalent structural elements with insubstantial differences from the literal language of the claims.

[0062] Further aspects are provided by the subject matter of the following clauses:

[0063] An inspection apparatus for inspecting an engine component includes: at least one controller configured to receive a set of inspection parameters based on inspection metrics determined by a life model of the engine component; a non-destructive evaluation (NDE) instrument configured to scan a predetermined area of ​​a surface of the engine component according to the set of inspection parameters to generate a data set; and a computer configured to apply an inspection algorithm to the data set and identify pixels outside a predetermined threshold with a flag, the pixels representing a potential chemical anomaly, and the flag representing a possible chemical anomaly location of the potential chemical anomaly.

[0064] Inspection device according to one or more of the preceding clauses, wherein the pixel is part of a group of pixels representing the potential chemical anomaly.

[0065] Inspection apparatus according to one or more of the preceding clauses, wherein the NDE instrument scans the predetermined area with high resolution scanning.

[0066] Inspection device according to one or more of the preceding clauses, wherein said predetermined area is defined by said marking.

[0067] Inspection apparatus according to one or more of the preceding clauses, wherein the predetermined area is a first predetermined area, and wherein the NDE instrument scans a second predetermined area with a high speed scan before scanning the first predetermined area with the high resolution scan.

[0068] Inspection apparatus according to one or more of the preceding clauses, wherein the NDE instrument is a micro-XRF instrument.

[0069] Inspection device according to one or more of the preceding clauses, wherein the detection metric is a call level.

[0070] Inspection device according to one or more of the preceding clauses, wherein the call level is between 70% and 100%.

[0071] The inspection device of one or more of the preceding clauses, wherein the computer is further configured to compare the at least one signature to a known set of chemical anomaly locations during a validation study to establish at least one of a recall level, a precision level, or an amount of inspection time.

[0072] Inspection device according to one or more of the preceding clauses, wherein said level of accuracy is between 60% and 90%.

[0073] The inspection device according to one or more of the preceding clauses, further comprising a user interface.

[0074] A method for inspecting an engine component, the method comprising: receiving a set of inspection parameters at an NDE instrument, the set of inspection parameters implementing an inspection metric determined by a life model of the engine component; scanning a predetermined area of ​​a surface of the engine component with the NDE instrument based on the set of inspection parameters to generate a data set; applying a detection algorithm to the data set with a computer; identifying a group of pixels outside a predetermined threshold using a marker, the group of pixels representing a potential chemical anomaly, and the marker representing a possible chemical anomaly location of the potential chemical anomaly; and outputting at least one of the marker, the potential chemical anomaly, or the possible chemical anomaly location.

[0075] The method of one or more of the preceding clauses, wherein receiving the set of inspection parameters comprises receiving the set of inspection parameters that achieves a minimum acceptable call level determined by the life model.

[0076] The method of one or more of the preceding clauses, wherein scanning the predetermined area comprises scanning with a high speed scanning setting or scanning with a high resolution scanning.

[0077] The method of one or more of the preceding clauses, wherein scanning with the high speed scan setting returns the data set as a partial data set.

[0078] The method of one or more of the preceding clauses, wherein scanning with the high speed scan setting returns the data set as a global data set.

[0079] The method of one or more of the preceding clauses further comprising, in the event a possible anomaly is detected, after scanning with the high speed scan setting, instructing the NDE instrument with the computer to scan only locations where potential chemical anomalies are detected with the high resolution scan.

[0080] The method of one or more of the preceding clauses, further comprising transforming the data set into a set of element maps before applying the detection algorithm.

[0081] The method of one or more of the preceding clauses, further comprising denoising the set of element maps to define an enhancement map.

[0082] The method of one or more of the preceding clauses, wherein applying the detection algorithm to the data set comprises applying the detection algorithm to the enhancement map to define a threshold map.

[0083] The method of one or more of the preceding clauses, further comprising identifying, with the detection algorithm, a set of pixels exceeding a threshold value to define the set of landmarks representing locations of the possible chemical anomaly.

[0084] The method of one or more of the preceding clauses, wherein the call level is between 70% and 100%.

[0085] The method of one or more of the preceding clauses, wherein the call level is between 50% and 100%.

[0086] The method according to one or more of the preceding clauses, wherein scanning at least a portion of the surface of the engine component comprises scanning a majority of the surface of the engine component with the high speed scanning setting.

[0087] The method of one or more of the preceding clauses, wherein determining the set of inspection parameters further comprises minimizing the call level, maximizing the level of accuracy, and minimizing the amount of inspection time while still satisfying the call level required by the life model.

[0088] At least one tangible computer-readable storage medium comprising instructions for implementing the method of any preceding clause.

[0089] At least one non-transitory computer-readable storage medium comprising instructions for implementing the method of any preceding clause.

[0090] At least one computing device configured to implement the method of any preceding clause.

[0091] At least one computing device configured to implement the inspection device described in any preceding clause.

[0092] At least one non-transitory computer-readable storage medium for use with the inspection apparatus of any preceding clause.

Claims

1. An inspection device for inspecting engine components, characterized in that: The inspection equipment includes: a controller configured to receive a set of inspection parameters based on a detection metric determined by a life model of the engine component; a non-destructive evaluation (NDE) instrument for scanning a predetermined area of ​​a surface of the engine component according to the set of inspection parameters to generate a data set; and A computer is configured to apply a detection algorithm to the data set and identify pixels outside a predefined threshold with a marker, the pixels representing a potential chemical anomaly, and the marker represents a possible chemical anomaly location of the potential chemical anomaly.

2. The inspection device according to claim 1, characterized in that in, The NDE instrument scans the predetermined area with high resolution scanning.

3. The inspection device according to claim 2, characterized in that in, The predetermined area is defined by the marking.

4. The inspection device according to claim 3, characterized in that in, The predetermined area is a first predetermined area, and wherein the NDE instrument scans a second predetermined area with a high speed scan before scanning the first predetermined area with the high resolution scan.

5. The inspection device according to claim 1, characterized in that in, The NDE instrument is a micro X-ray fluorescence (micro XRF) instrument.

6. The inspection device according to claim 1, characterized in that in, The detection metric is the call level.

7. The inspection device according to claim 6, characterized in that in, The call level is between 70% and 100%.

8. The inspection device according to claim 1, characterized in that in, The computer is further configured to compare the signature to a known set of chemical anomaly locations during a validation study to establish at least one of a call level, a precision level, or an amount of inspection time.

9. The inspection device according to claim 8, characterized in that in, The accuracy level is between 60% and 90%.

10. The inspection device according to claim 1, wherein Further included is a user interface.

Images