Methods and systems for characterizing populations of non-metallic inclusions

Abstract

A method for characterizing a population of non-metallic inclusions (24) in a finished or semi-finished part (10) comprising a metal alloy, such as steel. The method comprises the steps of cutting and polishing the finished or semi-finished part (10) to provide an inspection plane (20) and scanning at least one control area (22) constituting at least a portion of the inspection plane (20). The method further comprises the steps of automatically counting the non-metallic inclusions within the at least one control region (22) using an automatic optical microscope and image analysis software and measuring or determining at least one morphological parameter, such as area and / or length and / or width and / or aspect ratio, and the position of the non-metallic inclusions within the at least one control region (22), therefore, measurement data is generated. The method further comprises the step of processing the measurement data using an extreme value analysis (EVA) or generalized Pareto distribution (GPD) analysis technique to characterize the population of non-metallic inclusions (24) in the finished or semi-finished part (10).

Description

Technical Field

[0001] The present invention relates to a method and system for characterizing a population of non-metallic inclusions in a finished or semi-finished component comprising a metal alloy such as steel. The present invention also relates to a method for modeling the fatigue life of a component such as a bearing component, and a method for controlling the quality of a finished or semi-finished component such as a bearing component. Background Art

[0002] Non-metallic inclusions are compounds and non-metals present in metal alloys such as steel. They are the product of chemical reactions, physical effects and contamination, and may be formed during the production of metal alloys and during the manufacture of parts from metal alloys.

[0003] Non-metallic inclusions present in components such as bearing components can act as stress risers and can initiate crack formation, which propagate under stress reversal until fatigue pits or spalling are formed on the surface of the bearing component, which can lead to ultimate fatigue failure. Non-metallic inclusions can shorten the service life of the component.

[0004] Given the importance of non-metallic inclusion content on component fatigue performance, the characterization of inclusion populations in metal alloys and components made from metal alloys is of vital importance. Therefore, several evaluation methods for non-metallic inclusion analysis have been developed.

[0005] Ultrasonic testing (UST) is a nondestructive testing technique based on the propagation of ultrasonic waves in the component or material being tested. In the most common UST applications, extremely short ultrasonic pulse waves with a center frequency range of 0.1 to 15 MHz are emitted into the component or material to detect internal defects or characterize the non-metallic inclusion population in the material. However, UST is only used to quantify the number and size of macroscopic inclusions (i.e. non-metallic inclusions greater than 200 μm in length), not microscopic inclusions (i.e. non-metallic inclusions with a length of 1 μm to 200 μm formed during the manufacturing process of metal alloys).

[0006] Micro-computed tomography (micro-CT) is another non-destructive testing (NDT) method used to detect material defects. Micro-CT is a 3D imaging technology that uses X-rays to observe the interior of a component or material slice by slice. A micro-CT scanner captures a series of 2D planar X-ray images and reconstructs the data into 2D cross-sectional slices. These slices can be further processed into 3D models and printed into 3D objects for analysis. 3D micro-CT systems can be used to reveal the internal features of a component or material and provide volumetric information about its microstructure. However, this technology can only scan very small volumes of test material, i.e. about 1mm 3 This volume is too small to characterize the non-metallic inclusion population in the stress volume of the bearing component.

[0007] Manual non-metallic inclusion counting and measurement is used to characterize the non-metallic inclusion population in raw material in the form of bars, billets or plates for quality control purposes. In manual non-metallic inclusion counting and measurement, the test sample is taken from an area that represents the average mass of the bulk material.

[0008] The disadvantage of manual counting and measuring is that it does not provide useful information about the final products that are produced from the raw materials being analyzed. Final products produced from raw materials will contain a different population of metal inclusions than the raw materials used to produce them. Therefore, this raw material quality control by manual counting and measuring is important and useful to the raw material manufacturer, but not to the user of the parts manufactured from that raw material.

[0009] Another disadvantage of manual counting and measuring is that it is time consuming and relies heavily on the operator’s knowledge and experience, making it prone to human error.

[0010] Due to improvements in raw material production methods, such as improvements in steelmaking technology, metal alloys with the highest purity, such as steel, contain only a few large micro-inclusions and many unavoidable fine inclusions. However, in a practically small reference volume or area, large micro-inclusions are too rare to be measured directly, so the statistical distribution of large inclusions is often used to predict extreme sizes that are not observed in large volumes (such as tons of steel or thousands of bearings) by extrapolation based on small sample measurements.

[0011] Extreme value theory, which deals with the statistical behavior of extreme values ​​in a single process, has been used to estimate the size of large inclusions in clean steel.

[0012] ASTM E2283, entitled "Standard Practice for Extreme Value Analysis (EVA) of Non-metallic Inclusions in Steel and Other Microstructural Features," describes a method for statistically characterizing the distribution of maximum inherent non-metallic inclusions in steel samples based on quantitative metallographic measurements. This practice enables experimenters to estimate the extreme value distribution of inclusions in steel.

[0013] The EVA technique is based on the Gumbel distribution of statistical extreme value theory. The extreme values ​​in this method come from the largest block (BM) sampling, in which multiple independent, equally sized samples are scanned and only the largest inclusion present in each sample is taken for analysis. The disadvantage of this technique is that only the largest inclusions are considered in this method, while other large inclusions are ignored, which may also cause failure if they are present in the stressed volume of the component.

[0014] The disadvantage of the practice described in ASTM E2283 is that nonmetallic inclusions must be counted and measured manually, which is tedious and time consuming and rarely surveys the large areas required to obtain representative statistics for heterogeneous steels. Summary of the invention

[0015] It is an object of the present invention to provide an improved method for characterizing a population of non-metallic inclusions in a component made of or comprising a metal alloy such as steel.

[0016] The method comprises the steps of cutting and polishing a finished or semi-finished component to provide an inspection plane. Thus, the method is not performed on raw material but on a finished or partially finished component. The method comprises the steps of scanning at least one control area constituting at least a portion of the inspection plane and automatically counting non-metallic inclusions in at least one control area using an automated optical microscope and image analysis software, and measuring or determining at least one morphological parameter, such as area and / or length and / or width and / or aspect ratio, and the location, i.e., location coordinates, of some or all non-metallic inclusions in at least one control area to generate measurement data. The method also comprises the step of processing the measurement data using extreme value analysis (EVA) or generalized Pareto distribution (GPD) analysis techniques to characterize the population of non-metallic inclusions in the finished or semi-finished component.

[0017] The Generalized Pareto Distribution (GPD) analysis technique is an EVA technique that is based on a Generalized Pareto Distribution (GPD) of values ​​greater than a certain threshold. The GPD analysis considers not only the largest non-metallic inclusion in the control area (as in the EVA analysis), but also all non-metallic inclusions greater than the threshold in the entire inspection area. Therefore, using the Generalized Pareto Distribution (GPD) analysis technique provides the metallurgist with a better representation of the non-metallic inclusion population than given by the EVA analysis technique, because it is not always the largest non-metallic inclusion that is the fatigue initiator in the component. If the component is subjected to sufficient stress during service, the second or third largest non-metallic inclusion, etc., can also be a fatigue initiator.

[0018] The EVA or GPD statistical parameters generated by the method according to the present invention can be used as input data for fatigue models to predict the durability or fatigue strength of components under specific operating conditions.

[0019] The method according to the invention provides an efficient and reliable way of measuring and characterizing inclusions, i.e. macroscopic and / or microscopic inclusions. The method makes it possible to characterize a large number of non-metallic inclusions, e.g. several hundred thousand non-metallic inclusions, and / or analyze a large inspection area, e.g. an area of ​​several thousand square millimeters, in a relatively short time. Because a large number of measurement data is taken into account, the accuracy of the characterization of the non-metallic inclusion population is increased compared to known evaluation methods. Furthermore, the errors usually caused by manual measurement are eliminated.

[0020] Another advantage of the method is that the characterization of the non-metallic inclusion population is more relevant and useful for component manufacturers and suppliers, since it is the region of the finished or semi-finished component that is analyzed, rather than the region of the raw material that represents the average mass of the bulk material. Non-metallic inclusions generated during the process used to manufacture the finished component, i.e. the process used to partially manufacture the semi-finished component, will be analyzed. Thus, the EVA or GPD statistical parameters generated by the method according to the invention can be provided to the component manufacturer or supplier, so that improvements can be made to the component manufacturing process if necessary.

[0021] As used herein, the term "cutting and polishing" is intended to refer to any method of preparing a flat, defect-free surface suitable for inspection using an automated optical microscope. "Cutting and polishing" may include at least one of the following metallurgical techniques: cutting, grinding, polishing, lapping, electropolishing, leveling, honing. During sample preparation, artifacts such as corrosion pits, dust particles, and scratches must be eliminated.

[0022] As used herein, the term "optical microscope" is intended to refer to a microscope that uses visible light and a lens system to produce a magnified image of non-metallic inclusions. The term "motorized" is intended to refer to the optical microscope being configured to be movable and / or the microscope stage on which the finished or semi-finished component to be analyzed is placed being configured to be movable so that at least one control area can be scanned.

[0023] As used herein, the term "length of a non-metallic inclusion" is intended to refer to the measurement or range of the non-metallic inclusions seen from one end to the other of a control area.

[0024] As used herein, the term "width of a non-metallic inclusion" is intended to refer to a measurement of a non-metallic inclusion seen in a control zone, taken at right angles to the length of the non-metallic inclusion.

[0025] The term "measure or determine" is intended to refer to directly measuring or determining one or more morphological parameters from analysis of at least one control area. For example, the area of ​​a non-metallic inclusion may be determined using the equivalent circular diameter (ECD).

[0026] According to an embodiment of the present invention, the area of ​​the non-metallic inclusions is measured or determined by the method according to the present invention, because the area is the most accurate morphological parameter that can be measured or determined using image analysis software.

[0027] According to an embodiment of the present invention, at least one control area has a width of at least 150 mm. 2 , or at least 200mm 2 、300mm 2 , 400mm 2 , 500mm 2 Or up to at least 1000mm 2 If EVA analysis techniques are used, multiple control areas of this size can be used.

[0028] According to an embodiment of the invention, the method comprises the step of providing an inspection plane such that the inspection plane at least partially extends through a volume of a finished or semi-finished component that is stressed during use of the component. This will characterize the population of non-metallic inclusions in the stress volume of the finished or semi-finished component, which is very important for future applications. As used herein, the term "stress volume" is intended to refer to a volume slightly below a surface, such as a raceway surface in a bearing component, which is subjected to Hertzian contact stresses or alternating Hertzian contact stresses when the component is in use. When the component is in use, the maximum shear stresses are then generated slightly below this surface, and contact fatigue is most likely to occur within this stress volume. If stress raisers such as non-metallic inclusions are present in such high stress areas, delamination may occur.

[0029] It should be noted that the method according to the invention can be used to analyze a finished component before its use (to predict its useful service life in a future application), during its use (to determine the remaining service life) and / or after its use (to investigate the reasons why the component may have failed). Thus, a "stress volume" may be the volume of the finished component that is subjected to Hertzian contact stress or alternating Hertzian contact stress during its use.

[0030] According to an embodiment of the invention, the finished or semi-finished component comprises a raceway surface and the method comprises the step of providing the inspection plane such that the inspection plane extends at least partially parallel to the raceway surface and / or within 1 mm, 800 μm or 500 μm of the raceway surface.

[0031] The at least one control region preferably extends at least partially parallel to the raceway surface and / or within 1 mm, within 800 μm or within 500 μm of the raceway surface.Such an inspection plane orientation captures non-metallic inclusions associated with fatigue crack initiation and growth.

[0032] It should be noted that the expression "parallel" used herein is intended to mean parallel or substantially parallel.

[0033] According to an embodiment of the invention, the method comprises the step of characterizing a microscopic inclusion population, i.e. a population of non-metallic inclusions having a length of 1 μm to 10 μm, or 1 μm to 20 μm, or 1 μm to 50 μm, 1 μm to 100 μm, or 1 μm to 200 μm and including 200 μm. The method according to the invention can be used as a supplement to the high frequency ultrasonic technique for characterizing macroscopic inclusions.

[0034] According to an embodiment of the present invention, the finished or semi-finished component is one of the following components: a bearing component, such as an inner bearing ring or an outer bearing ring, a bearing raceway, a roller bearing, a needle roller bearing, a tapered roller bearing, a spherical roller bearing, a toroidal roller bearing, a ball thrust bearing, a roller thrust bearing, a tapered roller thrust bearing, a wheel bearing, a wheel hub bearing unit, a compact spherical roller bearing (CARB TM ), angular contact ball bearings (ACBB), deep groove ball bearings (DGBB), angular contact ball bearings, spherical roller bearings used in continuous casting lines, backing bearings, slewing bearings or ball screws, finished or semi-finished transmission components such as sprockets, gears, bushings, hubs, couplings, bolts, screws, shafts such as spindles, rollers or roller cages, seals, tools, metal wheels, wire, or any other finished or semi-finished component used in applications subjected to Hertzian contact stress or alternating Hertzian contact stress, or any finished or semi-finished structural component.

[0035] As used herein, the term "structural member" is intended to refer to any member that supports both non-variable forces or weights (dead loads) and variable forces or weights (live loads), such as columns, longitudinal beams, cross beams, supporting structures.

[0036] The finished or semi-finished component analyzed in the method according to the invention may comprise or be made of steel, such as bearing steel, hardened steel, carbon steel, stainless steel or any other metal alloy, such as nickel-based superalloys, titanium alloys, brass or bronze.

[0037] According to an embodiment of the invention, the method comprises the step of recording the measurement data. The measurement data can thus be processed at a later time and / or used for other purposes at a later time. For example, the measurement data can be used to determine the aspect ratio of the non-metallic inclusions to classify the non-metallic inclusions, as there is a correlation between the morphology and the chemical composition of the non-metallic inclusions.

[0038] The present invention also relates to a system for automatically characterizing a population of non-metallic inclusions in a finished or semi-finished component comprising a metal alloy, such as steel, bearing steel, hardened steel, carbon steel, stainless steel or any other metal alloy, such as nickel-based superalloys, titanium alloys, brass or bronze.

[0039] The system includes an automated optical microscope, such as a motorized optical microscope, and image analysis software configured to automatically count non-metallic inclusions in at least one control region and measure or determine at least one morphological parameter, such as area and / or length and / or width and / or aspect ratio, and the location of the non-metallic inclusions in at least one control region to generate measurement data. The system also includes a processor, which includes a computer program, the computer program including program code means executable by the processor to process the measurement data using extreme value analysis (EVA) or generalized Pareto distribution (GPD) analysis techniques to characterize the non-metallic inclusion population in the finished or semi-finished component.

[0040] Optionally, the system may include a memory to store measurement data and / or any other data used by and / or generated by the system.

[0041] A system according to any of the embodiments described herein may be used to perform a method according to any of the embodiments described herein.

[0042] The present invention also relates to a method for modeling the fatigue life of a component, such as a bearing component. The method comprises the step of predicting the fatigue strength of a bearing component under specific operating conditions using EVA or GPD statistical parameters, the statistical parameters being obtained using a method according to any embodiment described herein and / or a system according to any embodiment described herein. Therefore, the method and system according to the present invention can be used to help manufacturers find ways to extend the service life of components, such as bearing components.

[0043] Since the method according to the invention provides a large amount of measurement data, the accuracy of the fatigue life model using such measurement data will be improved.

[0044] The present invention also relates to a method for controlling the quality of finished or semi-finished components, such as bearing components. The method comprises the steps of using EVA or GPD statistical parameters obtained using a method according to any embodiment described herein and / or a system according to any embodiment described herein. Manufacturers and suppliers can use this quality control method to check the quality of finished or semi-finished components. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In the following, the invention will be further explained by way of non-limiting examples with reference to the accompanying drawings, in which:

[0046] Figure 1 An example of a component including a population of non-metallic inclusions that can be characterized using a method according to an embodiment of the invention is shown,

[0047] Figure 2 and 3shows the extension of the inspection plane and / or control area through the finished or semi-finished component,

[0048] Figure 4 This is an image of a group of non-metallic inclusions observed by an automated optical microscope.

[0049] Figure 5 shows block maximum sampling using extreme value analysis (EVA) analysis techniques,

[0050] Figure 6 shows the peak over threshold (POT) sampling using the generalized Pareto distribution (GPD),

[0051] Figure 7 shows the probability of component failure determined using the method according to the invention,

[0052] Figure 8 is a flowchart showing method steps according to an embodiment of the present invention, and

[0053] Fig. 9 A system according to an embodiment of the present invention is schematically shown.

[0054] It should be noted that the drawings are not necessarily drawn to scale and that the dimensions of certain features may be exaggerated for the sake of clarity.

[0055] It should also be noted that in any part of this document, including the accompanying drawings, any feature described with reference to a specific embodiment of a method or system according to the invention can be used in any other embodiment of a method or system according to the invention, unless the description explicitly excludes such a combination. DETAILED DESCRIPTION

[0056] Figure 1 An example of a component 10 comprising a population of non-metallic inclusions that can be characterized using a method according to an embodiment of the invention is schematically shown, namely a rolling element bearing comprising an inner ring 12, an outer ring 14, a set of rolling elements 16, a cage (not shown) and a raceway surface 18.

[0057] At least a portion of the component 10 or the entire component can be made of steel. For example, the bearing rings 12, 14 can be made of 100Cr6, a steel containing about 1% carbon and 1.5% chromium. Typically, the rolling elements 16 use the same material as the bearing rings 12, 14. However, when necessary, the rolling elements 16 can be made of a ceramic material, whereby the component 10 is a hybrid bearing. The bearing surface 18 of the component can include a pure metal, such as iron, nickel, titanium, copper, aluminum, tin or zinc, or a metal alloy, such as steel, carbon steel, stainless steel, nickel-based superalloys, titanium alloys, brass or bronze.

[0058] As described herein, the component 10 can have a width or diameter of up to several meters and a load-bearing capacity of up to several thousand tons. The component 10 can be of any size and have any load-bearing capacity. The component 10 can be used in industries such as metals, mining, mineral processing, cement, automotive, railroads, renewable or traditional energy, pulp or paper, or marine.

[0059] The method according to the invention comprises the steps of cutting and polishing the finished or semi-finished component 10 to provide an inspection plane 20 .

[0060] The method comprises the steps of scanning at least one control area 22 constituting at least a portion of an inspection plane 20 and automatically counting non-metallic inclusions within the at least one control area 22 using an automated optical microscope and image analysis software and measuring or determining at least one morphological parameter, such as area and / or length and / or width and / or aspect ratio, and the location of the non-metallic inclusions within the at least one control area 22.

[0061] The method according to the present invention comprises the steps of analyzing a population 24 of non-metallic inclusions present in at least one control region 22 using one of two extreme value analysis techniques, namely extreme value analysis (EVA) or generalized Pareto distribution (GPD) analysis techniques, to characterize the population of non-metallic inclusions in the finished or semi-finished component 10.

[0062] In the method according to the invention, the chemical composition of the non-metallic inclusions is not determined. The non-metallic inclusions are classified according to their morphological similarity rather than their chemical properties.

[0063] Figure 2 A component 10 is shown including a raceway surface 18. The inspection plane 20 is provided by cutting the component 10 along a plane indicated by a dashed line 20 below the raceway surface 18 and polishing the cross-sectional plane using a plurality of polishing steps, thereby removing 200 μm of material in each polishing step to produce an inspection plane 20 extending at least partially within 1 mm of the raceway surface 18, and / or such that at least a portion of the inspection plane 20 is parallel to the raceway surface 18.

[0064] An automated optical microscope, such as a motorized optical microscope, is used to scan one or more control areas 22 (in Figure 2The control area 22 constitutes at least a portion of the inspection plane 20 (indicated by a solid line along the inspection plane 20). The image analysis software is used to automatically count the non-metallic inclusions in the scanned control area 22 and measure or determine at least one morphological parameter, such as area and / or length and / or width and / or aspect ratio, and the position of the non-metallic inclusions in the scanned control area 22. The entire inspection plane 20 can be scanned by the automated optical microscope 28, whereby the entire inspection plane 20 constitutes a single control area 22 if the GPD analysis technique is used. Alternatively, the inspection plane 20 can be divided into a plurality of control areas 22 if the EVA analysis technique is used.

[0065] Figure 3 Also shown is the component 10 comprising a surface 18 which is subject to stress during use of the component 10 . Figure 3 It is shown which portion of the component 10 will constitute the at least one control area 22 once the component 10 has been cut and polished along a line parallel to the surface 18 .

[0066] Figure 4 is an image representing a population of non-metallic inclusions 24 observed by an automated optical microscope. Image analysis is used to automatically count and measure at least one morphological parameter of the non-metallic inclusions in at least one control region 22.

[0067] According to an embodiment of the present invention, at least 100,000 or at least 200,000 or at least 500,000 non-metallic inclusions are analyzed.

[0068] Figure 4 The inspection plane 20 is shown divided into a plurality of control areas 22, namely 24 control areas. 24 control areas are the default number of control areas because the EVA method described in the international standard ASTM E2283 requires 24 150 mm 2 The control areas 22 are ideally obtained from at least six different parts to ensure sampling variability). Each control area 22 is analyzed using the EVA analysis technique. However, in the method according to the present invention, the inspection plane 20 can be divided into any number of control areas for EVA analysis of any suitable size. The largest non-metallic inclusion in each control area 22 is determined.

[0069] Figure 5 and Figure 6 The same set of non-metallic inclusions was used to illustrate the differences in sampling and curve fitting for EVA and GPD.

[0070] Figure 5 Block maximum sampling of the EVA analysis technique using 24 maximum values ​​is shown. Figure 5 That is, a graph of the maximum non-metallic inclusions measured in each of the plurality of control regions 22 .

[0071] Figure 6 Peak over threshold (POT) sampling using generalized Pareto distribution (GPD) is shown.

[0072] Figure 5 and Figure 6 The comparison shows that using much larger non-metallic inclusion sizes in the GPD technique outperforms the standard 24 maximum used in the EVA technique.

[0073] The EVA plot is a simplified variation of the Gumbel cumulative distribution function (CDF):

[0074]

[0075] Among them, X (N) max is the maximum non-metallic inclusion size in each control volume, λ and δ are the location parameter and scale parameter of the Gumbel distribution, respectively.

[0076] The control volume V0 is estimated by V0 = A0 × d, where A0 is the control area size (the default is 150mm 2 ), d is the average of the measured sizes of n largest non-metallic inclusions.

[0077] The characteristic size is the non-metallic inclusion size corresponding to the fitted CDF:

[0078]

[0079] This is followed by the ±95% CI, which is equal to ±2·SE(x).

[0080] SE(x) is the standard error.

[0081] The CDF of GPD, F(x), is written as:

[0082]

[0083] Where x is the portion of non-metallic inclusions whose size exceeds the threshold θ, and σ and k are the scale and shape parameters of the GPD.

[0084] By V0 = A ref ×d estimates the control volume V0, where A ref is the total scanned control area size, and d is the average of the measured sizes of non-metallic inclusions larger than the threshold θ.

[0085] The characteristic size is the non-metallic inclusion size corresponding to the fitted CDF:

[0086]

[0087] Among them, A0 is the control area size (the default is 150mm 2 ), N is the total number of non-metallic inclusions larger than θ.

[0088] The EVA or GPD analysis techniques thus produce statistical parameters that can be used to characterize the non-metallic inclusion population 24 .

[0089] The values ​​given in the table below are examples of EVA and GPD statistical parameters that have been determined for different types of non-metallic inclusions depending on their morphology.

[0090]

[0091] λ and δ are the location and scale parameters of the Gumbel distribution, respectively.

[0092] V0 is the control volume.

[0093]

[0094]

[0095] σ and k are the scale and shape parameters of GPD, and θ is the threshold.

[0096] V0 is the control volume.

[0097] like Figure 7 As shown, the EVA or GPD statistical parameters generated by the method according to the present invention can be used as input to a fatigue model to predict the durability or fatigue strength of a component under specific operating conditions. Figure 7 That is, the probability of component failure determined using the method according to the invention is shown. It provides a prediction of the performance of a randomly selected type of component subjected to the method according to the invention.

[0098] Figure 8 is a flow chart illustrating the steps of a method according to an embodiment of the present invention.All automated microscopy, image analysis and data processing steps can be fully automated.

[0099] Fig. 9 A system 26 for automatically characterizing a population of non-metallic inclusions in a finished or semi-finished component 10 comprising a metal alloy according to an embodiment of the present invention is schematically shown. The system 26 includes an automated optical microscope 28, such as a motorized optical microscope 28, and image analysis software configured to automatically measure or determine at least one morphological parameter, such as area and / or length and / or width and / or aspect ratio, and the location of the non-metallic inclusions within a cut and polished inspection plane 22 of the finished or semi-finished component 10.

[0100] The automated optical microscope 28 can be a digital camera and / or a motorized stage. The control unit can be used to fully automatically control the automated optical microscope 28. The system can be configured to save and reuse the settings of the automated optical microscope to automatically focus and acquire images, and optionally record the results.

[0101] By properly adjusting brightness, contrast, and shading correction, non-metallic inclusions can be automatically detected because they appear as darker spots in the brighter metal alloy matrix, e.g. Figure 4 Automated non-metallic inclusion counting and morphological parameter measurements can be performed at a fixed magnification, such as ×100 or ×200. The measured morphological parameters can be stored together with the non-metallic inclusion position coordinates.

[0102] The system 26 may include a control unit (not shown) to control one of: movement of the optical microscope 28 and / or a microscope stage on which the finished or semi-finished component 10 is placed, illumination of the finished or semi-finished component 10, and magnification of the optical microscope 28. The non-metallic inclusions being observed may be magnified at any suitable magnification, such as 400x, 500x, 600x, 700x, 800x, 900x, 1000x, or more.

[0103] The system 26 also includes a processor 28 including a computer program including program code means executable by the processor to process the measurement data using extreme value analysis (EVA) or generalized Pareto distribution (GPD) analysis techniques to characterize the population of non-metallic inclusions in the finished or semi-finished component. The system 26 may also include a memory 32 for storing the area and / or length and / or width and / or shape and / or location of non-metallic inclusions within the cut and polished inspection plane of the finished or semi-finished component, or any other data used or generated by the system 26 and / or a display device 34 for displaying data used or generated by the system 26.

[0104] Further modifications of the invention within the scope of the claims will be apparent to the skilled person.

Claims

1. A method for characterizing a population (24) of non-metallic inclusions in a finished or semi-finished component (10) comprising a metal alloy such as steel, characterized in that The method comprises the following steps: - cutting and polishing the finished or semi-finished component (10) to provide an inspection plane (20), - scanning at least one control area (22) constituting at least a portion of the inspection plane (20) and automatically counting non-metallic inclusions within the at least one control area (22) using an automated optical microscope and image analysis software and measuring or determining at least one morphological parameter, such as area and / or length and / or width and / or aspect ratio, and the position of the non-metallic inclusions within the at least one control area (22) to generate measurement data, and - Processing the measured data using extreme value analysis (EVA) or generalized Pareto distribution (GPD) analysis techniques to characterize the population (24) of non-metallic inclusions in the finished or semi-finished component (10).

2. The method according to claim 1, characterized in that The control area (22) has a width of at least 150 mm. 2 area.

3. The method according to claim 1 or 2, characterized in that: The method comprises the step of providing the inspection plane (20) such that the inspection plane (20) extends at least partially through a stress volume of the finished or semi-finished component (10).

4. The method according to any one of the preceding claims, characterized in that The method is used to characterize a microinclusion population (24).

5. The method according to any one of the preceding claims, characterized in that The finished or semi-finished component (10) is one of the following: a bearing component, such as an inner bearing ring or an outer bearing ring, a bearing raceway, a roller bearing, a needle bearing, a tapered roller bearing, a spherical roller bearing, a toroidal roller bearing, a ball thrust bearing, a roller thrust bearing, a tapered roller thrust bearing, a wheel bearing, a wheel hub bearing unit, a compact spherical roller bearing (CARB TM ), angular contact ball bearings (ACBB), deep groove ball bearings (DGBB), angular contact ball bearings, spherical roller bearings used in continuous casting lines, backing bearings, slewing bearings or ball screws, finished or semi-finished transmission components such as sprockets, gears, bushings, hubs, couplings, bolts, screws, shafts such as spindles, rollers or roller cages, seals, tools, metal wheels, wire, or any other finished or semi-finished component used in applications subjected to Hertzian contact stress or alternating Hertzian contact stress, or any finished or semi-finished structural component.

6. The method according to any one of the preceding claims, characterized in that The finished or semi-finished component (10) comprises a raceway surface (18), and the method comprises the step of providing the inspection plane (20) such that the inspection plane (20) extends at least partially in a direction parallel to the raceway surface (18) and / or such that the inspection plane (20) extends at least partially within 1 mm of the raceway surface (18).

7. The method according to any one of the preceding claims, characterized in that The method comprises the step of recording the measurement data.

8. A system (26) for automatically characterizing a population (24) of non-metallic inclusions in a finished or semi-finished component (10) comprising a metal alloy such as steel, characterized in that The system includes: an automated optical microscope (28), such as a motorized optical microscope, and image analysis software, configured to automatically count non-metallic inclusions (24) within at least one control area (22) and measure or determine at least one morphological parameter, such as area and / or length and / or width and / or aspect ratio, and the position of the non-metallic inclusions (24) within the at least one control area (22) to generate measurement data, a processor (30) comprising a computer program comprising program code means executable by the processor (30) to process said measurement data using extreme value analysis (EVA) or generalized Pareto distribution (GPD) analysis techniques to characterize the population (24) of non-metallic inclusions in said finished or semi-finished component (10), and - Optionally, a memory (32) for storing said measurement data and / or any other data used by and / or generated by the system (26).

9. A method for modeling the fatigue life of a component, such as a bearing component, characterized in that The method comprises the step of predicting the fatigue strength of a bearing component under specific operating conditions using EVA or GPD statistical parameters obtained using a method according to any one of claims 1 to 7 and / or a system (26) according to claim 8.

10. A method for controlling the quality of a finished or semi-finished component (10), such as a finished or semi-finished bearing component, characterized in that The method comprises the step of using EVA or GPD statistical parameters obtained using the method according to any one of claims 1 to 7 and / or the system according to claim 8.

Images