• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    defect detection system on an object. the invention relates to a system and a process for detecting defects on an object (11), comprising the following steps: - forming an image (13) representative of that object (11), from relative signs (9) to the object, - build subdivisions (15) of this image according to self-adaptive resolutions, and - calculate differentials between different subdivisions to detect an abnormal subdivision indicative of breakdown premises.
    公开号:BR112015024826B1
    申请号:R112015024826-8
    申请日:2014-03-27
    公开日:2020-10-06
    发明作者:William BENSE；Valerio Gerez
    申请人:Snecma；
    IPC主号:
    专利说明:

    TECHNICAL FIELD
    [0001] The present invention relates to the domain of defect detection on an object and, more particularly, to the monitoring of the rotating elements of an aircraft engine. PREVIOUS TECHNICAL STATUS
    [0002] There are numerous techniques, applying optical or audible processes to detect defects in an object. The advantage of these techniques is that they can be applied in a non-intrusive way.
    [0003] For example, in the aeronautical field, when testing production or inspecting an engine's blades, different non-destructive control techniques based on the use of thermal cameras are applied. These techniques consist of using a mobile heat emitter to heat the blade and a mobile thermal camera to photograph an infrared image of the blade. Image analysis is based on comparing a very high number of image areas in order to detect defects in the blade.
    [0004] Thus, current methods need to perform calculations that can occupy a considerable amount of time, except in using particularly powerful and very expensive calculators.
    [0005] Furthermore, it is difficult to know what granularity to adopt (that is, what is the resolution of the area to be compared), while the defects sought are not known.
    [0006] The objective of the present invention is, therefore, to propose a detection process that is simple to apply without going through complex calculations and capable of quickly and accurately detecting defects on an object or element of a motor, without presenting the pre-existing drawbacks. cited. EXPOSURE OF THE INVENTION
    [0007] The present invention is defined by a defect detection process on an object, which comprises the following steps: - forming a representative image of that object, from signals related to the object; - build subdivisions of this image, according to self-adaptive resolutions, and - calculate the differentials between different subdivisions to detect an abnormal subdivision indicative of breakdown premises.
    [0008] The process makes it possible to detect defects of virtually all sizes within a reasonable time.
    [0009] Advantageously, the process comprises a confirmation phase that comprises a comparison of the differentials relating to an abnormal subdivision belonging to a last image with differentials relating to the same abnormal subdivision belonging to each of a determined number of previous images of that object.
    [0010] This allows confirming the detection of defects, avoiding false alarms.
    [0011] Advantageously, the process includes: - a high or very high alert generation, if it is found that the differentials have increased in the course of the last images; and - a generation of alerts of medium importance, if it is found that the differentials remain constant during the last images.
    [0012] This allows to assess the importance of the premises of failure or defects.
    [0013] According to a preferred embodiment of the invention, the steps of constructing the subdivisions and calculating the differentials comprise the following steps: - (a1) grid this image in a plurality of current subdivisions; - (a2) calculate first current differentials between each current subdivision and neighboring current subdivisions; - (a3) check if there is a current subdivision for which first current differentials with at least a first determined number of neighboring subdivisions are indicative of anomaly; - (a4) calculate, in the case where the previous step (a3) is confirmed, second current differentials between that current subdivision and distant current subdivisions; (a5) verify that this current subdivision, present with at least a second number of current subdivisions apart, second current differentials indicating anomaly; - (a6) declare this current subdivision to be invalid in the event that the previous step (a5) is confirmed; - (a7) requadricular an area that covers this invalid current subdivision to form new subdivisions that overlap the previous invalid subdivision, the new subdivisions being considered as the current current subdivisions; - (a8) repeat steps (a2) - (a6) for each of the new current subdivisions of that covering area; - (a9) make an ET mask in the covering area between the preceding invalid subdivisions and the new subdivisions that thus form subdivisions of reduced sizes, comprising at least one invalid subdivision, these subdivisions of reduced sizes being considered as the current current subdivisions; - (a10) check if the current subdivision size is greater than a predetermined resolution; and - (a11) reiterate, in the case where the previous step (a10) is confirmed, the previous steps (a2) - (a10) for each current current subdivision, if not, declare the current current subdivision (s) as abnormal subdivisions.
    [0014] Thus, this process is based on zooms and an optimal number of pertinent comparisons, allowing to reduce the calculation load and not to prejudge the size of the defect beyond the resolution.
    [0015] Advantageously, it is verified in step (a3) if the first current differentials are greater than a first predetermined limit and it is verified, in step (a5), whether the second current differentials are greater than a second predetermined limit.
    [0016] This allows the detection of defects that do not consider errors, as well as possible differences in contexts between remote regions.
    [0017] Advantageously, the process includes the construction of a learning database that records healthy differentials between different subdivisions of the image and calculates in step (a3) the differences between the first current differentials and corresponding sound differentials to verify if they they are higher than a predetermined level and in step (a5) the differences between the second current differentials and corresponding sound differentials are calculated to check if they are higher than a second predetermined level.
    [0018] This allows to consider inhomogeneities that may exist in the middle of the object.
    [0019] Advantageously, this object is a rotating element of an aircraft engine.
    [0020] Indeed, the numerical treatment, according to the invention, is economical in calculation and can, therefore, be easily used by means of treatment embedded in an aircraft.
    [0021] According to an embodiment, the signals relating to that object are infrared signals from the object, so that the representative image of that object is an infrared image that translates a thermal field in a transient phase, after heating the object by a thermal request.
    [0022] According to another embodiment, the signals related to that object are ultrasound signals coming from the object, so that this representative image of that object is an image that translates ultrasonic waves reflected by the object.
    [0023] The invention also aims at a defect detection system on at least one rotating element of an aircraft engine, comprising: - means of on-board excitations to cause the emission of signals by that rotating element; - embedded acquisition means for acquiring signals sent by this rotating element; and - treatment means configured to carry out the process steps, according to any of the preceding claims.
    [0024] According to a first embodiment of the system, according to the invention, the excitation means are heating means to heat that rotating element of the motor by a thermal request, and acquisition means are thermographic means to acquire an infrared image, translating a thermal field in transitory phase of this rotating element.
    [0025] According to a second embodiment of the system, according to the invention, the excitation means are means of emitting ultrasound waves, and the acquisition means are means of receiving ultrasound waves reflected by the object.
    [0026] The invention also includes an automatic defect detection system on at least one rotating element of an aircraft engine, comprising: - built-in heating means to heat that rotating element of the engine by thermal demand; - embedded thermographic means to acquire at least one infrared image, translating a thermal field in transitory phase of this rotating element; and - treatment means to calculate differentials related to a component of the thermal field between different subdivisions of this image, in order to detect variations of that component of the thermal field indicative of defects on that rotating element.
    [0027] Thus, each flight and, automatically, the rotating elements of the engine can be monitored to detect the first signs of fatigue. This makes it possible to carry out predictive maintenance and not simply preventive maintenance as the rotating elements can be replaced, when they actually suffer damage, thus increasing profitability (less parts replaced) and safety (less risk of blade loss). The analysis is performed according to differential measures that allow to get rid of the context. In particular, the fact of making comparisons between spatially close zones makes it possible to avoid problems due to the distance from the heat source or lightening by the sun.
    [0028] Advantageously, when the differential corresponding to a current subdivision is indicative of anomaly, the treatment means are configured to calculate other differentials, reorganizing the subdivisions and / or fine-tuning the current comparison subdivision, in order to locate the defect locations .
    [0029] This allows to reduce the number of subdivisions to be studied and, therefore, to reduce the calculation time and the request for a calculator.
    [0030] Advantageously, the treatment means are configured to register, for each flight, these differentials relative to the thermal fields of the different subdivisions and to analyze the evolution of these differentials from flight to flight.
    [0031] This allows to consolidate the result of the detection and to follow, in a systematic way, the health of the rotating elements from flight to flight.
    [0032] Advantageously, the detection system includes a database of signatures of degradations representing different forms of degradation and their progress, and the treatment means are configured to compare the differentials related to the thermal fields of the subdivisions that present defects in these degradation signatures.
    [0033] This allows the most likely type of defect to be determined.
    [0034] In accordance with a more advantageous embodiment of the present invention, the heating means are made up of at least one pre-heated heating element already in the engine.
    [0035] This makes it possible to reduce the mass shipped and also allows to control the heating medium itself.
    [0036] According to a variant, the heating means are intended to heat that element by thermal pulsations.
    [0037] Thus, the rotating element can be heated in a sufficiently short time, so that the material of the rotating element does not reach a constant temperature.
    [0038] According to this variant, the treatment means are configured to calculate differentials between an amplitude of the thermal field of a current subdivision and amplitudes of the thermal fields of neighboring subdivisions.
    [0039] According to another variant, the heating means are intended to heat that element by periodic thermal waves.
    [0040] According to this other variant, the treatment means are configured to calculate lags between the thermal field of a current subdivision and the thermal fields of neighboring subdivisions.
    [0041] Detection, depending on the lag, has the advantage of being little influenced by the distance from the heat source or by the sunlight, as the temperature is not measured, but the lag.
    [0042] Advantageously, the rotating element is a blade of a wheel with a blade of that motor. BRIEF DESCRIPTION OF THE DRAWINGS
    [0043] Other features and advantages of the invention will appear with the reading of preferred embodiments of the invention, made with reference to the attached figures, among which: - figure 1 illustrates, schematically, a defect detection system on a object, according to the invention; figure 2 represents an algorithm that illustrates different stages of a detection process, according to a preferred embodiment of the invention; - figures 3A-3E are grid of an image, illustrating, in a schematic way, the steps of the organization chart of figure 2; figures 4A-4D illustrate the detection of point and progressive defects on different grid lines, according to the invention; figure 5 represents a detection algorithm that comprises a confirmation phase, according to a first embodiment of the process, according to the invention; figure 6 represents a detection algorithm, comprising a confirmation phase, according to a second embodiment of the process; according to the invention, - figure 7 illustrates, schematically, a defect detection system on a rotating element of an aircraft engine, according to a first embodiment of the system of figure 1; - figure 8 represents a block diagram, illustrating the steps for detecting defects on a wheel with engine blades, according to the system in figure 7; and - figure 9 schematically illustrates a defect detection system on a rotating element of an aircraft engine, according to a second embodiment. DETAILED EXPOSURE OF PARTICULAR ACCOMPLISHMENTS.
    [0044] The concept based on the invention is based on a process of detecting defects in an object, using a representative image of the object, performing a minimum number of comparisons between zones, the size of which adapts hyperactively to the size of the defects.
    [0045] Figure 1 schematically illustrates a defect detection system on an object, according to the invention.
    [0046] The detection system 1 comprises excitation means 3, acquisition means 5, data processing means 7.
    [0047] The means of excitations are intended to cause the emission of signals 9 by the object 11, while the means of acquisition 5 are intended to acquire these signals.
    [0048] As an example, the excitation means 3 are heating means to heat the object 11 by a thermal request and the acquisition means 5 are thermographic means to acquire an infrared image, translating a thermal field in transient phase (see the figure 7).
    [0049] According to another example, the excitation means 3 are means of emitting ultrasound waves and the acquisition means 5 are means of receiving ultrasound waves reflected by the object 11 (see figure 9).
    [0050] Treatment means 7 are intended to treat signals 9 related to the object, in order to detect defects and generate alerts.
    [0051] More particularly, the treatment means 7 are configured to form a graph or image 13 representative of the object 11, from signals 9 relative to the object. Image 13 is defined here as a representation of object 11 in the numerical or mathematical sense of the term in which each point of image 13 corresponds to an antecedent of object 11. Thus, the image may correspond, for example, to an optical, thermal representation , or acoustics of object 11.
    [0052] The treatment means 7 are also configured to construct, in an iterative way, subdivisions of the image 13, according to self-adaptive resolutions, that is, according to resolutions that adapt to the extent of the defect.
    [0053] In addition, the treatment means 7 are configured to calculate, in an iterative way, differences between different subdivisions, in order to detect an abnormal subdivision indicative of breakdown premises.
    [0054] Figure 2 represents an algorithm illustrating different stages of a detection process, according to an embodiment of the invention.
    [0055] In addition, figures 3A-3E are grid of an image 13, illustrating, in a schematic manner, the steps of the organization chart of figure 2.
    [0056] In step E1 the treatment means 7 are configured to form the image 13 representative of the object 11, from signals 9, relative to that object. The image 13 can, for example, be the translation of a thermal field acquired by thermographic means or the translation of ultrasound signals acquired by sound wave pickups. It is also defined a determined resolution corresponding to a minimum size of defects, this allows, on the one hand, not to warn about acceptable defects and, on the other hand, to give a stop to the algorithm.
    [0057] In step E2 the treatment means 7 are configured to square the image 13 in a plurality of current subdivisions. The dimensions of the subdivisions are selected according to the size of the object, so that one can have close and distant neighbors.
    [0058] Figure 3A shows a grid 15 of part of the image 13 in nine subdivisions 115a-115i in the form of large squares of the same sizes. The small squares 225 represent the selected resolution. On the other hand, it will be noted that the subdivisions can also be hexagonal or triangular in shape or any other geometric shape.
    [0059] In step E3, treatment means 7 are configured to calculate first current differentials between each current subdivision 115a and neighboring current subdivisions 115b-115i. For example, the differential represents the difference in the values of a physical parameter (for example, optical, thermal or sound field) between a subdivision and a neighboring subdivision.
    [0060] More particularly, the means of treatment 7 calculate a component of the physical parameter for each subdivision and then compare the component of each subdivision to those of its neighbors.
    [0061] According to the example of figure 3A, each square is compared with its eight neighbors, calculating the differential between, on the one hand, the component of the physical parameter in a square 115a and, on the other hand, the component relative to each one of the eight bordering squares 115b-115i.
    [0062] Step E4 is a test in which the treatment means 7 are configured to check whether or not there is a current subdivision for which of the first current differentials with at least a first determined number of neighboring subdivisions are indicative of anomalies.
    [0063] An anomaly indicator can, for example, be the comparison of the differential with the predetermined limit. As a variant, the anomaly indicator can be defined by the difference between the observed differential and a measured healthy differential, during a learning phase and the comparison of this difference with the predetermined level. It will be noted that the predetermined limit or level may depend on several factors, such as the number of neighbors, the size of subdivision 115a, the measured physical parameter, the desired precision, etc.
    [0064] If the test result in step E4 is negative, then in step E5, object 11 will be considered valid.
    [0065] On the contrary, if a subdivision is found for which the first current differentials, with at least a first determined number of neighboring subdivisions, are indicative of anomaly, then this subdivision is considered to be potentially invalid and proceeds to step E6 .
    [0066] It will be noted that, if the differentials indicate an anomaly with only another neighboring subdivision, one can truly consider it to be an inaccuracy or measurement error. In other words, for the subdivision to be declared potentially invalid, there must be at least a limit number of neighboring subdivisions with which the differentials are indicative of anomaly. This limit number can also depend on the number of neighbors, the size of the subdivision, the measured physical parameter and the desired precision. In the example in figure 3A, this limit number is chosen equal to four and this figure shows that subdivision 115a in the center has, in relation to at least four of its neighbors 115b-115i, an differential indicative of anomaly.
    [0067] Thus, when the test of step E4 is confirmed, the treatment means 7 are configured to compare, in step E6, the potentially invalid subdivision with distant subdivisions. In particular, the treatment means 7 calculates second current differentials between the potentially invalid current subdivision and remote current subdivisions. For example, the incriminated central square 115a of figure 3A can be compared with only eight remote neighbors (not shown), in order to limit the calculation load. It will be noted that the neighbors are chosen differently for the subdivisions at the edges of the image, as they cannot be considered neighbors in all directions. Thus, in order to consider edge effects, different limits can be chosen according to the fact that potentially invalid subdivisions are at the edges or inside the image.
    [0068] Step E7 is a test in which the treatment means 7 are configured to verify that the current subdivision is present with at least a second number of current subdivisions apart, current differential differentials indicating anomaly are considered neighbors sufficiently distant to leave the potentially invalid zone. If the test result of step E7 is negative, then, in step 8, the incriminated subdivision is considered to be valid. Indeed, if a subdivision is different from its next neighbors, but not distant neighbors, it can be deduced that the subdivision in question is valid, but not the next neighbors. In this case, a particular status can be considered, but, in any case, the next neighbors will also be tested and detected by the algorithm.
    [0069] On the contrary, if the test result of step E7 is confirmed, then in step E9 the incriminated subdivision is considered invalid.
    [0070] As before, an anomaly is detected, when the differential is greater than a predetermined limit. In addition, for the incriminated subdivision to be declared invalid, there must be at least a second number of neighboring subdivisions with whom the differentials are indicative of anomaly. Figure 3A also shows that subdivision 115a in the center has at least four of its remote neighbors (not shown) a differential indicative of anomaly.
    [0071] The fact of comparing a given subdivision with its close neighbors and then with distant neighbors allows to confirm the invalidity of the subdivision and to adapt the resolution of the subdivisions. In fact, if the differentials between the given subdivision and the nearby neighbors are indicative of an anomaly and if the anomaly results from a real defect, then the differentials with the distant neighbors must also indicate an anomaly, considering that it moves away from the area defective. In particular, if the anomaly is due to a progressive defect, then differentials with distant neighbors are necessarily more important than with close neighbors. On the other hand, if the defect is very punctual, then the differentials with distant neighbors will be at least as important as with close neighbors.
    [0072] It will be noted that, in order to avoid false alarms, the same limits are not taken into account for near and far comparisons. In fact, the subdivisions that are spaced out are usually sufficiently spaced from the incriminated area and therefore present in relation to these very important differentials. However, the context in the remote areas can be different and, therefore, the values of the physical parameter between the two zones can present significant deviations, without necessarily having a defect. Thus, to avoid false alarms, it is advantageous to choose a higher limit for a comparison between two distant subdivisions than for a comparison between two close subdivisions.
    [0073] Then, the treatment means 7 are configured to calculate other differentials, reorganizing the subdivisions and / or fine-tuning their sizes.
    [0074] In effect, in step E10, the treatment means are configured to requadricular an area 215 covering the subdivision 115a declared invalid (see figure 3B). Thus, new subdivisions are formed that overlap the invalid subdivision. The new zone 215 is homotetic to the invalid subdivision, for example, with a strictly understood relationship between 1 and 2.
    [0075] The example in figure 3B illustrates, in a schematic way, a reorganization of the subdivisions, according to a simple lag of a horizontal squared and a vertical squared one. Thus, this example shows that four new current squares 215a - 215d cover the previous invalid current square 115a (represented in dotted lines). Each of these four new squares 215a-215d covers a portion of the preceding square 115a plus a portion of its immediate vicinity. This allows to scrutinize all the surroundings of the zone declared invalid.
    [0076] Again, the treatment means 7 calculate new differentials regarding the new cut of the comparison areas.
    [0077] In effect, the new subdivisions 215a-215b are considered as the current current subdivisions and for each of these new subdivisions, steps E11-E17 are carried out, which are equivalent to steps E3-E9 respectively.
    [0078] Thus, in step E11 the treatment means 7 are configured to calculate first current differentials between each new current subdivision 215a-215b and neighboring current subdivisions.
    [0079] Step E12 is a test in which the treatment means 7 are configured to check if there is a new current subdivision for which first current differentials with at least a first determined number of neighboring subdivisions are indicative of anomaly. If the test result of step E12 is negative, then in step E13 the subdivision is considered to be valid, otherwise, it is considered to be potentially invalid and the step E14 is passed.
    [0080] In step E14, the treatment means 7 are configured to compare the new potentially invalid subdivision with remote subdivisions.
    [0081] Step E15 is a test in which the treatment means 7 are configured to check if the new current subdivision has at least a second number of current subdivisions away from the current second differentials indicative of anomaly. If the test result from step E15 is negative, then in step E16 the incriminated subdivision is considered to be in reverse, if the test result from step E15 is confirmed, then it is considered in step E17 that the new incriminated subdivision is invalid .
    [0082] Thus, at the end of step E17, there is at least one new invalid division and one preceding invalid subdivision. The example in figure 3B shows a new invalid current square 215a and a preceding invalid square 115a. The cutout between the current and the preceding invalid squares gives more precision on the location of the defect.
    [0083] In effect, E18 the treatment means 7 are configured to make a mask, according to a logical operation ET between the previous invalid subdivisions 115a and the new subdivisions 215 to 215d in the covering zone 215. This forms reduced size subdivisions 315a-315d, comprising at least one invalid reduced size subdivision 315a (see figure 3c). These new reduced size 315a-315d subdivisions are considered to be the current current subdivisions.
    [0084] In step E19, treatment means 7 are configured to verify that the current current subdivision size 315a-315d is larger than the predetermined resolution 100. If so, treatment means 7 will be configured to reiterate the previous steps E3 -E18 for each current subdivision, and if not, the current subdivision or subdivisions are declared invalid in abnormal step in step E20.
    [0085] Figure 3C shows that the ET mask refines the zone, decreasing the length and width of the square by a factor of 2. However, the size of the invalid square 315a remains larger than the size of the smaller square 100 corresponding to the resolution and therefore , the same steps are restarted, as shown in figure 3D. Finally, figure 3E shows that resolution 100 is reached and the smaller invalid squares 100a-100d are located.
    [0086] The example of figures 3A-3E shows that the detection process, according to the invention, allows to greatly reduce the number of calculation steps.
    [0087] In fact, the image, according to the example of figures 3A-3E, contains 18x 18 = 324 small squares 100. Thus, making an abstraction of the border effects, if each square 100 was compared with its other neighboring squares, having 2592 comparisons would be made and the technique would be less effective, since only anomalies that were very localized on a 100 grid would be detected.
    [0088] With the above technique and always abstracting the edge effects, in the step of figure 5A, 9 x 8 = 72 comparisons are made, in the step of figure 5B, 4 x 8 = 32 comparisons, in the step of the figure 5C, zero comparison, and finally in the step of figure 5D, 9 x 8 = 72 comparisons, whether in total 176 comparisons only. This allows to reduce the calculation time and the request of the calculator.
    [0089] More generally, for an image of an object of 100 cm x 20 cm and a resolution of 1 mm, if each 1 mm zone were taken independently and compared with its eight neighbors, they would have, without counting edge effects 160000 comparisons and only 1 mm defects or very marked defects can be detected.
    [0090] However, applying the detection process, according to the invention, taking an initial grid of 1 cm and assuming that with only a single defect, the total number of comparisons is approximately 16000.
    [0091] Thus, the process, according to the invention, considerably the number of calculations, optimizing the number of comparisons. In addition, it allows the detection of defects, the size of which lies between the dimensions of an initial subdivision 115a and the selected resolution 100.
    [0092] In fact, figures 4A-4D illustrate the detection of point and progressive defects on different grid lines.
    [0093] Figures 4A and 4B show that a point defect 21A can be detected on a large square 425 or on a small square 525. However, figures 4C and 4D show that a progressive defect 21b can be detected on a large square 425 , but not on a small square 525. In fact, the differential between a small square 525 and its neighbors is very small and thus a progressive defect would not be detected with a classic method that considers only small squares.
    [0094] Figure 5 is a detection algorithm, according to the invention, comprising a confirmation phase, according to a first embodiment.
    [0095] The confirmation phase comprises a comparison of the differentials related to an abnormal subdivision, belonging to a last image with differentials related to the same abnormal subdivision belonging to each of a determined number of previous images of the object, the data of the previous subdivisions abnormalities being recorded in a database 17 associated with the means of treatment.
    [0096] Step E21 refers to the measurement or acquisition of a physical parameter (for example, optical, thermal or sound field) related to object 11, allowing the formation of an image 13 of object 11.
    [0097] In step E22, the data referring to the physical parameter are sent to the treatment means 7.
    [0098] In step E23, the treatment means 7 are configured to process the data according to the organization chart of figure 2.
    [0099] In particular, in steps E4 and E12 (figure 2), it is verified whether the first current differentials are greater than a first predetermined limit. Likewise, in steps E7 and E15 (figure 2), it is verified whether the current second differentials are greater than a second predetermined limit. It will be noted that the values of the first and second limits can be modified depending on the size of the subdivision and, therefore, the row of the iteration. For example, in the first iteration, the detection is chosen quite sensitive (that is, small limits) to allow the identification of abnormal subdivisions. Indeed, if the subdivision is large, it will eventually become normal to normal and abnormal areas within the subdivision.
    [0100] At the end of step E23, if no defects are found, then in step E24 the database 17 for the latest detections will be reset to zero.
    [0101] On the contrary, if at the end of step E23, one or more abnormal subdivisions are detected, then in step E25, information relating to or the last abnormal subdivisions will be recorded in database 17, before going to step E26.
    [0102] In step E26, the treatment means 7 are configured to compare the differentials related to the abnormal subdivisions belonging to the last image with differentials related to the same abnormal subdivisions belonging to each of the previous images of the object 11.
    [0103] If it is found that image 13 presents an anomaly for the first time, then no alerts will be generated (step E27).
    [0104] On the contrary, if it is found that the differentials have increased during the last images, then an alert of high importance will be generated (step E28). You can add an additional alert level to follow the trend of the differentials. For example, an alert of very high importance will be generated, if the differentials increase and the extrapolation, if the case is determined (for example, 10 flights) an exceeding a predetermined limit.
    [0105] Finally, if it is found that the differentials remain constant throughout the last images, then an alert of medium importance will be generated (step E29).
    [0106] On the other hand, different levels of limits corresponding to different levels of alerts can be assigned.
    [0107] Figure 6 represents a detection algorithm, according to the invention, comprising a confirmation phase according to a second embodiment.
    [0108] The steps of the algorithm in figure 6 are identical to those in figure 5, except for steps E32 and E33.
    [0109] As previously, step E31 refers to the measurement or acquisition of a physical parameter (for example, optical, thermal or sound field) related to object 11, allowing the formation of an image 13 of the object. If the data correspond to a first image, then step E32 and otherwise step E33.
    [0110] Stage E32 is a learning phase during which a learning database is built, comparing the differentials of the subdivisions of the first image of a healthy object. This can be done according to the steps of comparisons between neighboring subdivisions of the organization chart in figure 2.
    [0111] Thus, a learning database is built in step E32, which records healthy differences between different subdivisions of the original healthy image, knowing that it is not necessarily uniform because of the intrinsic, but normal differences of the object.
    [0112] If the image of the object is not a first image, then it will proceed to step E33 where the treatment means 7 are configured to process the data, according to the organization chart of figure 2.
    [0113] However, in steps E4 and E12 (figure 2), the differences between the first current differentials of the neighboring subdivisions and the corresponding sound differentials are calculated to check if they are higher than a predetermined level. In steps E7 and E15 (figure 2), it is verified whether the current second differentials between remote subdivisions and corresponding healthy differentials are greater than a second predetermined limit.
    [0114] The detection process of the present invention is economical in calculation and can therefore be easily used by means of treatment embedded in an aircraft to, for example, detect defects or failure premises of a rotating element of an aircraft engine .
    [0115] In effect, figure 7 schematically illustrates a defect detection system on a rotating element of an aircraft engine, according to a first embodiment of the system in figure 1.
    [0116] The rotating element 111 is visible from the outside and corresponds, for example, to a blade of a wheel 112 or to a rotating cover of the engine 114. The wheel 112 can belong to a compressor of the engine 114 and can, for example, match an aircraft engine blown or unblown blower.
    [0117] According to the embodiment of figure 7, the signals relating to the rotating element are infrared signals from the element and, therefore, the excitation means are heating means 113 on board and the acquisition means are thermographic means 115 on board .
    [0118] The heating means 113 are intended to heat the rotating element 111 of the engine 114 by a thermal demand 119. Naturally, heat penetrates the material of the rotating element 3. Thus, local temperatures will vary from one region to another, because the heat will penetrate more or less depending on the presence or absence of defects 121. As an example, the heating means 113 may consist of one or more thermal emitters fixed on the engine 114 or the aircraft 116 in front of the rotating element 111. Thus, each thermal emitter 113 remains permanently on the aircraft 116 and can be adjusted to heat the rotating element 121 in a periodic or pulsational manner.
    [0119] On the other hand, the thermographic means 115 are intended to acquire at least one infrared image 113 of the rotating element 111, translating, in sequence to the thermal request 119 of the heating, a thermal field in transitory phase. It will be noted that the thermographic means 115 may consist of one or more thermal cameras fixed on the engine 114 or the aircraft 116 in front of the rotating element 111.
    [0120] Advantageously, the rotating elements 111 are filmed, when rotating at a very low speed (that is, at the beginning of the actuation phase, at the end of the stop phase or when operating in a windmill on the ground) ). This allows to have a complete view of the rotating elements 111, without disturbing the acquisition of the images 113. The interest in filming in low regime is the use of a single camera to detect defects on all the blades one after the other. You can naturally shoot at the stop, but in this case, you need a plurality of cameras to detect defects on all blades.
    [0121] It will be noted that the heating and acquisition of the images have the advantage of being able to be done without contact, which allows, on the one hand, not to damage the tested material and, on the other, not to have pickups to place on the blades 111 or very close to the blades, which could disturb their aerodynamics. In addition, the fact that the heating means 113 and thermographic 115 are shipped allows images 113 to be obtained with each flight, automatically, and without long and expensive human intervention.
    [0122] The treatment means 107 are configured to obtain the infrared image 113, from the thermographic means 115 and to use steps of the detection process, according to the organization charts of figures 2, 5 or 6.
    [0123] Differentials can be calculated for a component (for example, amplitude or phase) of the thermal field between different subdivisions of the infrared image 113. The detection of variations in the thermal field component is indicative of defects or rupture premises of the rotating element 111. Thus, the rotating elements 111 of the engine 114 can be monitored automatically and each time to detect the first signs of fatigue. , before a shovel loss occurs. In particular, the detection system 101 is well adapted to monitor the rotating elements 111 in composite materials that can undergo fatigue, generating defects 121 not visible on their surfaces.
    [0124] It will be noted that the fact of performing the analysis of the data, according to differential measures, on transient thermal phases, allows to get rid of the context such as the external temperature or the illumination by the ground. Indeed, external conditions act in the same way on a current subdivision in two successive moments.
    [0125] Advantageously, it is possible to explore the treatment means 107 of a calculator 118 embedded in the aircraft 116 or in a calculator 118 integrated in the engine 114 of type MEU (Engine Monitoring Unit) to exploit the detection system 101, of according to the invention. In particular, calculator 118 can be used to run a computer program registered to storage means 117 of calculator 118 and comprising code instructions for applying the detection process according to the invention.
    [0126] It will be noted that the acquired data can be directly processed during the flight of the aircraft. As a variant, data processing can be performed after the aircraft has landed, in order not to overload the calculator 118 during the flight. According to yet another variant, the acquired data can be transmitted to the ground to be processed by a calculation station.
    [0127] According to a first variant of the embodiment of figure 7, the heated means 113 are intended to heat the rotating element 111 by thermal pulsations or transient thermal phases.
    [0128] The heated means 113 correspond to a thermal emitter (for example, a heating lamp) fixed directly on the engine or the aircraft, in front of the rotating element 111 to heat it, in a pulsational manner. The rotating element is then heated in a sufficiently short time (a few milliseconds) so that the material of the rotating element does not reach a constant temperature. The thermal emitter is fixed at a predetermined distance from the rotating element which can vary from a few millimeters to a few meters.
    [0129] The thermographic media 115 corresponding, for example, to a thermal camera installed in the vicinity of the rotating element, for example, between a few centimeters and a few meters and acquire the images when heating.
    [0130] In this case, the treatment means 107 is configured to calculate differentials between an amplitude of the thermal field (that is, the temperature) of a current subdivision and the amplitude (that is, temperatures) thermal fields neighboring subdivisions. Thus, if the material of the rotating element has a defect in the surface or in depth, the temperature on the surface as a result of the pulsational thermal stress will be different. The comparison of temperatures between the different subdivisions then allows the detection of defects.
    [0131] According to a second variant of the embodiment of figure 7, the heating means 113 are intended to heat the rotating element 111 by periodic thermal waves for a determined time, for example, in the order of a few seconds. In this case, a thermal emitter 113 is fixed on the engine (or the aircraft) in front of the rotating element 111 at a predetermined distance that can vary from a few millimeters to a few meters. The thermal emitter 113 corresponds, for example, to a flash-type heating lamp, sending a periodic thermal wave of a predetermined frequency to periodically heat the rotating element.
    [0132] A thermal camera 115 is installed in the vicinity of the rotating element 111, for example, between a few centimeters and a few meters, and acquires the images when heating.
    [0133] The heat emitter 113 and thermal camera 115 can be placed directly on the aircraft's fuselage or canopy.
    [0134] In this second variant, the treatment means 107 are configured to perform, for example, a Fourier analysis to determine the phase variation between the different subdivisions of the rotating element's infrared image. If the material is uniform, the thermal energy is distributed identically and there is no lag between the different zones. On the contrary, if the material of the rotating element has a defect, the thermal energy will not propagate in an identical way and the thermal wave will be accelerated or reduced in the defect, which will translate into a lag. Thus, the treatment means 107 calculates the lags between the thermal field of a common subdivision and the thermal fields of the neighboring subdivisions, in order to detect the defects.
    [0135] It will be noted that this second variant has the advantage of being little influenced by the distance from the heat source or the sunlight, as the temperature is not measured, but the lag. In order to increase the accuracy of the measurements, it is preferable that the heat emitter is not too far from the rotating element.
    [0136] According to a third variant of the embodiment of figure 7, the heating means 113 are made up of at least one pre-heated heating element already existing in the engine.
    [0137] Indeed, if the rotating elements 111 already have heating means to prevent frost, the detection system of the present invention can cunningly use that heat source and, therefore, the installation of supplementary heating means can be omitted and, therefore, reduce the shipment mass.
    [0138] In this case, the anti-heat element is regulated, when self-tests on the drive, for example, to provide heat for predetermined durations.
    [0139] If the heating element is not integrated in the blade, but fixed on the outside, then the detection process is strictly identical to that of the first and the second variant. On the contrary, if the heating element is sufficiently powerful and is integrated into the blade, a relatively short heating time of a few seconds accompanied by a cooling time of a few seconds can be used.
    [0140] More particularly, if the heating element consists, for example, of heating wires distributed over the surface of the blade, the heating element is fed with a constant current intensity, during a determined heating time, then stops feed it in order to lower the temperature. After a determined waiting time (always identical from flight to flight), one is then in a transient phase of the thermal field and the treatment means 107 activate the camera to take an infrared photo. In the event of an anomaly in the material of the rotating element, the cooling will be different and you can then compare each subdivision to its near and far neighbors, and this from flight to flight. On the contrary, in this case, defects cannot be detected under the wires, as their temperature will make the thermal response of the material at that location false.
    [0141] On the contrary, if the heating wires are not on the surface, but integrated into the material of the blade, the situation will be more favorable than before, because the wires do not hide any surface of the blade and the answer is directly accessed in thickness, and you can then detect internal defects and over the entire surface of the blade. The treatment of the data is the same as that previously detailed.
    [0142] In addition, the treatment means 107 are advantageously configured to check the proper functioning of the anti-heated heating element, monitoring the amplitude differential of the rotating elements. Thus, if the amplitude response is less and less, even null or increasingly greater from flight to flight, even considering the aging effect of the blades on the thermal responses, the treatment means 107 can incriminate the heating element .
    [0143] Figure 8 is a block diagram that illustrates the steps of detecting defects on a wheel with an engine blade, according to the system in figure 7.
    [0144] According to this example, the rotating element 111 corresponds to each of the paddles of the wheel with paddle 112.
    [0145] In block B1, treatment means 107 receive data, from engine 114 (represented by block B2), regarding the rotation speed of the wheel with paddle 112 to be monitored. The treatment means 107 triggers the detection process, when the paddle wheel 112 starts to rotate at very low speed.
    [0146] In block B3, the heating means 113 heats the paddles 111 of the wheel with paddle 112 of the engine 114 (block B2) by a thermal request 119 generating a thermal field that evolves according to the heating and cooling phases. It will be noted that the thermal demand (a thermal pulse or a periodic thermal wave) penetrates the material of the blade 111, so that if the material presents a defect 121 (in the surface or in depth), the amplitude and / or the phase of the thermal field at the surface will be different.
    [0147] Then, while the thermal field is in its transient heating or cooling phase, the thermal camera (s) 115 films the paddles 111 of the wheel with paddle 112 of the engine ( block B2) to acquire at least one infrared image 113 of the blades 111.
    [0148] Thus, in block B4, at least one infrared image 113 is generated. It will be noted that each thermal camera 115 can be configured to acquire an image per paddle or a single image for all paddles of the paddle wheel 112.
    [0149] In block B5, means of identification 122 of the paddles 111 are used to distinguish the different paddles from the wheel with paddle 112. This allows for monitoring over time of the different paddles and to identify the one (s) that presents (s) defects.
    [0150] These means are, for example, optical means of shape recognition. One can, for example, use the thermal camera 115 itself coupled to a shape recognition algorithm to identify the blades.
    [0151] As a variant, the means of identification are means of individualization by means of a marking 122 or labeling on one or two paddles 111 of the wheel with paddle 112. Paddles 111 can be individualized, numbering them by painting or any other material inserted in the shovel or arranged on its surface.
    [0152] In block B6, the treatment means 107 carry out the steps of the organization chart of figures 2, 5 or 6. In particular, the treatment means 107 carry out, for example, a Fourier analysis to calculate a component (the amplitude or the phase) of the thermal field of each subdivision of the infrared image 113 for each of the blades 111 and compare the different subdivisions between them.
    [0153] More particularly, when a thermal pulse is used to heat the paddles 111 of the wheel with paddle 112, then the component of the thermal field corresponds to the amplitude of the thermal field (i.e., the temperature). In contrast, when a periodic thermal wave is used to heat the blades 111, then the component of the thermal field corresponds to the phase of the thermal field.
    [0154] If the blade material is uniform, its entire surface will respond in the same way to thermal demand 119 and, therefore, the thermal field component is constant over all zones. Conversely, if the material has a surface or depth roughness, the thermal field component on the surface following thermal stress 119 will be different. Thus, by comparing the components over different relatively close zones, defects can be detected.
    [0155] When a defect is detected in block B7, then, before sending the alert, treatment means 107 compares the results on several flights to block B7, to see if the anomaly is always detected (see also steps E26- E29 and E36-E39 of figures 5 and 6). In effect, the treatment means 107 are configured to record the differences related to the thermal fields of the different subdivisions for each flight, in order to analyze the evolution of these differences in flight in flight. Thus, one can quantify the evolution of defects for each blade 111, comparing the data from the current flight with data from previous flights stored in a database 117a (block B8).
    [0156] Advantageously, the detection system 101 comprises an anomaly library or a database 117b (block B8) of signatures of degradations representing different forms of degradation and their progress states. This allows the treatment facilities 107 to compare the differentials related to the thermal fields of the zones that present premises of failure in the characteristic signatures of degradations and thus to determine the type of degradation and its progress.
    [0157] Figure 9 schematically illustrates a defect detection system on a rotating element of an aircraft engine, according to a second embodiment of the system in figure 1.
    [0158] According to this embodiment, the signals relating to the rotating element are ultrasonic signals coming from the rotating element and, therefore, the excitation means are means of emitting sound waves 213 and the acquisition means are means of receiving 215 of ultrasonic waves reflected by element 211.
    [0159] The issuing means 213 and the receiving means 215 can form a single device. More particularly, the device may comprise an electrocapacitive or piezometric type ultrasound source coupled to a receiver of the same type (i.e., electrocapacitor, if the source is electrocapacitor).
    [0160] The issuing means 213 and receiving 215 are loaded and installed on the engine 214, by means of the rotating and / or rotating means 232. Thus, the issuing means 213 and the receiving means 215 can rotate and / or rotate to scan several zones of space, scanning, for example, each rotating element 211 of a wheel with paddle 212. Thus, it is not necessary to install an ultrasonic emitter / receiver device on each paddle. It will be noted that the rotating elements 211 are scanned when they are stopped.
    [0161] Advantageously, the ultrasound source 213 produces ultrasound in the low range (for example, [50 kHz-1 MHz]) to avoid a very strong attenuation that is a function of the square of the frequency. The frequency can be adapted according to the desired resolution (that is, the size of the defects sought). The higher the frequency and the more important the resolution, the more attenuated the signal will be and, therefore, must be found depending on the application.
    [0162] The treatment means 207 are configured to form an image 213 representative of the rotating element 211 from the ultrasonic waves 219 reflected by the element 211 and captured by the receiving means 215 and to use the steps of the detection process, according to the organization charts of figures 2, 5 or 6.
    [0163] Differentials can be calculated with respect to the intensity or direction of reflected 219 ultrasound waves. It will be noted that the ultrasonic waves penetrate very little into the material and will therefore reflect more than 99.9% (due to the difference in acoustic impedance between air and metals or composite materials). Thus, in the event of an anomaly or surface defect, the reflection will be different in intensity or in direction and it is then possible to detect small structural defects as a result, for example, of an impact from a foreign body.
    [0164] Thus, the rotating elements 211 of the engine 214 can be monitored automatically and each time to detect the first signs of fatigue, before a blade loss occurs.
    [0165] It will be noted that the fact of performing the data analysis, according to differential measures on the intensities or directions of the reflected waves, allows to get rid of the context, just as the fact that the monitored element cannot always be at the same distance from the means of receipt.
    [0166] The present invention thus allows to monitor the rotating elements in metal or in composite materials of an aircraft engine, in order to detect the first signs of fatigue with the aid of the means fixed on the engine or the aircraft, each flight, automatically and individually. It is advantageously applied to the accompaniment of the blades of a turbofan blower, on the propellers of a turboprop or open rotor, as well as its rotating caps.
    权利要求:
    Claims (12)
    [0001]
    1. Defect detection process on an object, characterized by the fact that it comprises the following steps: - forming an image (13) representative of that object (11), from signs (9) related to the object; - build subdivisions (15) of this image according to self-adaptive resolutions, comprising comparing a given subdivision with its close neighbors and then with distant neighbors, these resolutions adapting, in an iterative way to the extent of the defect; and - iteratively calculate the differentials between different subdivisions to detect an abnormal subdivision indicative of the breakdown premises.
    [0002]
    2. Process, according to claim 1, characterized by the fact that it comprises a confirmation phase that comprises a comparison of the differentials related to an abnormal subdivision belonging to a last image (13) with differentials related to the same abnormal subdivision belonging to each one of a predetermined number of previous images of that object.
    [0003]
    3. Process, according to claim 2, characterized by the fact that it includes: - a generation of alerts of high importance or very high if it is found that the differentials increased during the last images; and - a generation of alerts of medium importance, if it is found that the differentials remain constant during the last images.
    [0004]
    4. Process according to any one of claims 1 to 3, characterized by the fact that the stages of building the subdivisions and calculating the differentials comprise the following steps: - (a1) grid this image in a plurality of current subdivisions; - (a2) calculate first current differentials between each current subdivision and neighboring current subdivisions; - (a3) check if there is a current subdivision for which first current differentials with at least a first determined number of neighboring subdivisions are indicative of anomaly; - (a4) calculate, in the case where the previous step (a3) is confirmed, second current differentials between that current subdivision and distant current subdivisions; - (a5) verify that this current subdivision, present with at least a second number of current subdivisions apart, second current differentials indicating anomaly; - (a6) declare this current subdivision to be invalid in the event that the previous verification step (a5) is confirmed; - (a7) requadricular an area that covers this invalid current subdivision to form new subdivisions that overlap the previous invalid subdivision, the new subdivisions being considered as the current current subdivisions; - (a8) repeat steps (a2) - (a6) for each of the new current subdivisions of that covering area; - (a9) make a mask, according to a logical ET operation, in this covering area between the previous invalid subdivisions and the new subdivisions thus forming reduced size subdivisions, these reduced size subdivisions being considered as the current current subdivisions; - (a10) check if the current subdivision size is larger than a predetermined resolution; and - (a11) reiterate, in the event that the previous verification step (a10) is confirmed, the previous steps (a2) - (a10) for each current current subdivision, if not, declare the current current subdivision (s) as abnormal subdivisions .
    [0005]
    5. Process, according to claim 4, characterized by the fact that step (a3) is verified if the first current differentials are higher than a first predetermined limit, and verified in step (a5), if the second current differentials are greater than a second predetermined limit.
    [0006]
    6. Process according to any one of claims 1 to 4, characterized by the fact that it comprises a construction of a learning database, which records healthy differences between different subdivisions of the image and by the fact that in step (a3) they are calculated the differences between the first current differentials and corresponding sound differentials to check if they were higher than a predetermined level and the fact that in step (a5) the differences between the second current differentials and corresponding sound differentials are calculated to check if they were greater than one second predetermined level.
    [0007]
    7. Process according to any one of the preceding claims, characterized in that this object is a rotating element (111) of an aircraft engine (114).
    [0008]
    8. Process, according to any of the preceding claims, characterized by the fact that the signals related to that object are infrared signals coming from the object, so that this representative image of that object is an infrared image that translates a thermal field in a transient phase , after heating the object by a thermal stress.
    [0009]
    9. Process, according to any of the preceding claims, characterized by the fact that the signals related to that object are ultrasound signals coming from the object, so that this representative image of that object is an image that translates ultrasonic waves reflected by the object .
    [0010]
    10. Defect detection system on at least one rotating element (111) of an aircraft engine (114), characterized by the fact that it comprises: - means of on-board excitations to cause the emission of signals by this rotating element; - embedded acquisition means to acquire the signals sent by that rotating element; and - treatment means configured to carry out the process steps as defined in any of the preceding claims.
    [0011]
    11. System according to claim 10, characterized by the fact that the excitation means are heating means (113) to heat this rotating element of the motor by a thermal request, and the fact that the acquisition means are thermographic means (115) ) to acquire an infrared image, translating a thermal field in transitory phase of this rotating element.
    [0012]
    12. System, according to claim 10, characterized by the fact that the excitation means are means of emission (213) of ultrasonic waves, and the fact that the acquisition means are means of receiving (215) of ultrasonic waves reflected by the object.
    类似技术:
    公开号 | 公开日 | 专利标题
    BR112015024826B1|2020-10-06|PROCESS AND DEFECT DETECTION SYSTEM ON AN OBJECT
    US9562870B2|2017-02-07|Method and device for non-destructive testing of wind turbine blades
    RU2610973C2|2017-02-17|Endoscopic examination system and method of gas turbines examination
    BR112015029618B1|2020-12-01|method of inspection of an object by ultrasound transmission, use of inspection method and system
    EP3282252A1|2018-02-14|Method and system for inspection of additive manufactured parts
    US7954372B2|2011-06-07|Diagnosis of pitch and load defects
    BR112017008942B1|2021-03-02|method and system for structural health monitoring with frequency synchronization
    EP3014255B1|2022-01-05|Modular device for structural diagnostics of various materials and structures, using thermographic techniques based on multiple excitations
    BR102012032958B1|2021-01-05|system and method for remote and automatic assessment of structural damage and repair
    US20120074932A1|2012-03-29|Nondestructive inspection of a structure in an aircraft
    US8910511B2|2014-12-16|Tire inner liner anomaly probe grounding apparatus and method
    US20170336156A1|2017-11-23|Method and system for thermographic analysis
    EP2605213B1|2020-02-05|Method and system for processing images for inspection of an object
    BR102015026105B1|2021-07-06|method for monitoring a surge condition, and turbomachine surge monitoring system
    Hwang et al.2020|Remote inspection of internal delamination in wind turbine blades using continuous line laser scanning thermography
    Avdelidis et al.2013|Non-destructive evaluation | of Composites: Infrared | thermography of wind turbine blades
    JP5187695B2|2013-04-24|Nondestructive inspection equipment for parts by analysis of radiation dissipation
    WO2018215103A1|2018-11-29|Method of performing structural inspection of a wind turbine rotor blade
    Błachnio2014|Capabilities to assess health/maintenance status of gas turbine blades with non-destructive methods
    US20200409322A1|2020-12-31|Control of power generation system by visually monitoring gauge during operation
    US20200408105A1|2020-12-31|Control of power generation system by visually monitoring component during operation
    US20200409321A1|2020-12-31|Control of power generation system by visually monitoring valve during operation
    Heyns et al.2019|Towards the use of hybrid models for diagnosis and prognosis in turbomachinery health management
    Sheppard et al.2011|Blade inspection guidance for offshore wind turbine facilities
    BR102012029569A2|2015-04-07|Integrity problem prediction method in an electrical appliance
    同族专利:
    公开号 | 公开日
    RU2015146547A|2017-05-05|
    BR112015024826A2|2017-07-18|
    CA2907946A1|2014-10-02|
    WO2014155011A1|2014-10-02|
    CN105122046A|2015-12-02|
    US9976967B2|2018-05-22|
    JP2016515704A|2016-05-30|
    JP6349383B2|2018-06-27|
    US20160054233A1|2016-02-25|
    EP2979082A1|2016-02-03|
    CA2907946C|2021-10-19|
    RU2645443C2|2018-02-21|
    CN105122046B|2018-01-23|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    GB2161602B|1984-07-12|1987-04-23|Atomic Energy Authority Uk|Examination over a surface|
    JPH0739999B2|1991-01-24|1995-05-01|肇産業株式会社|Defect detection method|
    JP3693508B2|1998-10-28|2005-09-07|株式会社東京精密|Pattern comparison method and visual inspection apparatus|
    FR2815123B1|2000-10-10|2003-02-07|Snecma Moteurs|ACOUSTIC CONTROL OF MONOBLOCK BLADES|
    FR2823344B1|2001-04-06|2003-06-13|Univ Toulouse|METHOD FOR OBTAINING A SIMULATED NUMERICAL REPRESENTATION OF THE RADIATIVE STATE OF A HETEROGENEOUS THREE-DIMENSIONAL SCENE|
    CN1140796C|2001-06-08|2004-03-03|清华大学|Taxonomic real-time reconstruction checking method of large-size workpiece, defect|
    EP1532433A2|2002-08-28|2005-05-25|Wayne State University|System forinfrared imaging by inducing acoustic chaos|
    US7064330B2|2003-04-30|2006-06-20|United Technologies Corporation|Infrared defect detection via broad-band acoustics|
    US7489811B2|2004-10-08|2009-02-10|Siemens Energy, Inc.|Method of visually inspecting turbine blades and optical inspection system therefor|
    US7653235B2|2005-10-27|2010-01-26|Honeywell International Inc.|Surface anomaly detection system and method|
    US8818683B2|2006-04-21|2014-08-26|General Electric Company|Method and apparatus for operating a gas turbine engine|
    US7617730B2|2006-06-28|2009-11-17|The Boeing Company|Ultrasonic inspection and repair mode selection|
    US8985848B2|2006-06-30|2015-03-24|Bdc Capital Inc.|Thermal inspection system|
    US7553070B2|2006-11-06|2009-06-30|The Boeing Company|Infrared NDI for detecting shallow irregularities|
    US20090175530A1|2007-11-12|2009-07-09|Fredrik Sjostrom|Methods and apparatuses for detecting pattern errors|
    US7855368B2|2008-07-23|2010-12-21|The United States Of America As Represented By The Administrator Of The National Aeronautics And Space Administration|Air-coupled acoustic thermography for in-situ evaluation|
    FR2940449A1|2008-12-24|2010-06-25|Snecma|METHOD FOR NON-DESTRUCTIVE CONTROL OF A MECHANICAL PART|
    US8345949B2|2010-03-09|2013-01-01|General Electric Company|Sequential approach for automatic defect recognition|
    DE102010032241A1|2010-07-26|2012-01-26|Smartvision Gmbh|Method for detecting surface defects e.g. cracks at plastic surface of corrugated pipe, involves testing differential image between two partial region images to determine whether coherent regions are provided with grey values|FR3011936B1|2013-10-11|2021-09-17|Snecma|PROCESS, SYSTEM AND COMPUTER PROGRAM FOR ACOUSTIC ANALYSIS OF A MACHINE|
    FR3027667B1|2014-10-22|2020-10-09|Snecma|METHOD AND DEVICE FOR MONITORING A BLADED WHEEL OF AN AIRCRAFT ENGINE BY MEASURING THE BALANCE POSITION|
    GB2538700B|2015-04-27|2021-06-09|Thermoteknix Systems Ltd|Conveyor belt monitoring system and method|
    CN104964886A|2015-05-27|2015-10-07|东南大学|Welded member fatigue stress and strain real-time non-contact type monitoring method|
    CN105118044B|2015-06-16|2017-11-07|华南理工大学|A kind of wheel shape cast article defect automatic testing method|
    EP3327198B1|2015-07-21|2021-04-21|Kabushiki Kaisha Toshiba|Crack analyzer, crack analysis method, and crack analysis program|
    CN106959292A|2016-01-11|2017-07-18|宝山钢铁股份有限公司|The quick system for determining precisely imaging scheme is detected for strip defects|
    CN106198731A|2016-07-19|2016-12-07|中国人民解放军装甲兵工程学院|Matrix fatigue crack recognition methods under sprayed coating|
    CN106197913B|2016-08-30|2019-06-07|大唐东北电力试验研究院有限公司|Wind electricity blade detection system|
    US10197517B2|2017-03-24|2019-02-05|Rosemount Aerospace, Inc.|Probe heater remaining useful life determination|
    US10914777B2|2017-03-24|2021-02-09|Rosemount Aerospace Inc.|Probe heater remaining useful life determination|
    US11060992B2|2017-03-24|2021-07-13|Rosemount Aerospace Inc.|Probe heater remaining useful life determination|
    US10895592B2|2017-03-24|2021-01-19|Rosemount Aerospace Inc.|Probe heater remaining useful life determination|
    US10564203B2|2017-03-24|2020-02-18|Rosemount Aerospace Inc.|Probe heater remaining useful life determination|
    US10473603B2|2017-04-18|2019-11-12|Saudi Arabian Oil Company|Apparatus, system and method for inspecting composite structures using quantitative infra-red thermography|
    US10726540B2|2017-10-17|2020-07-28|International Business Machines Corporation|Self-similarity analysis for defect detection on patterned industrial objects|
    US10800553B2|2018-11-30|2020-10-13|The Boeing Company|Solar radiography for non-destructive inspection|
    US10962580B2|2018-12-14|2021-03-30|Rosemount Aerospace Inc.|Electric arc detection for probe heater PHM and prediction of remaining useful life|
    US11061080B2|2018-12-14|2021-07-13|Rosemount Aerospace Inc.|Real time operational leakage current measurement for probe heater PHM and prediction of remaining useful life|
    KR20200095758A|2019-02-01|2020-08-11|엘지전자 주식회사|Non-identical camera based image processing device|
    法律状态:
    2018-11-13| B06F| Objections, documents and/or translations needed after an examination request according [chapter 6.6 patent gazette]|
    2020-02-11| B06U| Preliminary requirement: requests with searches performed by other patent offices: procedure suspended [chapter 6.21 patent gazette]|
    2020-07-21| B09A| Decision: intention to grant [chapter 9.1 patent gazette]|
    2020-10-06| B16A| Patent or certificate of addition of invention granted [chapter 16.1 patent gazette]|Free format text: PRAZO DE VALIDADE: 20 (VINTE) ANOS CONTADOS A PARTIR DE 27/03/2014, OBSERVADAS AS CONDICOES LEGAIS. |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1352860A|FR3003952B1|2013-03-29|2013-03-29|SYSTEM FOR DETECTING DEFECTS ON AN OBJECT|
    FR1352860|2013-03-29|
    FR1352859A|FR3003951B1|2013-03-29|2013-03-29|DEFECT DETECTION SYSTEM IN AN AIRCRAFT ENGINE ROTATING ELEMENT|
    FR1352859|2013-03-29|
    PCT/FR2014/050730|WO2014155011A1|2013-03-29|2014-03-27|System for detecting defects on an object|
    [返回顶部]