Spectrometer-based method for predicting mechanical properties of connecting rod materials

Abstract

The application discloses a connecting rod material mechanical property prediction method based on a spectrometer, relates to the technical fields of spectral detection and connecting rod material mechanical property prediction, and comprises the following steps: continuously exciting at the same detection position of a key stress area of a connecting rod, and acquiring a spectrum sequence arranged according to the excitation order; extracting carbon relative response and excitation state response from the spectrum sequence, and determining a surface interference attenuation boundary; correcting carbon relative response before the boundary based on carbon relative response after the boundary, forming carbon response depth distribution, and determining a base material stable spectrum segment; determining base material reference mechanical property according to the base material stable spectrum segment, and determining a decarburization influence quantity according to the carbon response depth distribution; and correcting the base material reference mechanical property by using the decarburization influence quantity, so as to obtain a local mechanical property prediction result of the detection position.

Description

Technical Field

[0001] This invention relates to the field of spectroscopic detection and prediction of mechanical properties of connecting rod materials, and more specifically, to a method for predicting the mechanical properties of connecting rod materials based on a spectrometer. Background Technology

[0002] Connecting rods are critical force-transmitting components in reciprocating machinery such as engines and compressors. They are subjected to tensile, compressive, inertial forces, and alternating loads over long periods. Stress concentrations easily occur in the fillets of the large and small ends, the transition zone of the rod body, and the area adjacent to the fracture surface. The surface material condition of these critical stress areas has a significant impact on fatigue crack initiation, local hardness, load-bearing capacity, and service life. During processes such as heat treatment, forging, shot peening, cleaning, or rust prevention, an oxide layer, deposits, or a decarburized layer may form on the connecting rod surface. Surface decarburization reduces the surface carbon content and strengthening capacity, causing the actual mechanical properties of the critical stress areas to be lower than those of the internal base material.

[0003] Existing spectrometer-based methods for predicting material composition or mechanical properties typically utilize continuous laser excitation to obtain multiple spectra. Spectra exhibiting significant fluctuations in the initial stages of continuous excitation are discarded as data indicating surface contamination, oxide layer influence, or unstable excitation states. Material property predictions are then established based on subsequent stable spectra. While this method reduces the impact of surface interference on the prediction of base material properties, it is insufficient for detecting decarburization on the connecting rod surface. This is because the carbon response changes caused by decarburization are primarily concentrated in the initial stages of continuous excitation and overlap with interference from oxide layers, surface deposits, and changes in plasma state. Directly discarding the initial spectra can easily lead to the loss of true decarburization information, causing the prediction results to primarily reflect the internal base material properties and failing to accurately characterize the localized mechanical property weakening caused by decarburization and its depth distribution on the critical stress-bearing surfaces of the connecting rod.

[0004] The above-disclosed technical solutions have at least the following technical problems: Existing spectrometer-based methods for predicting the mechanical properties of connecting rod materials are difficult to distinguish between the real decarburization information in the spectrum before continuous excitation and the interference information caused by the oxide layer, surface deposits and changes in the excitation state. They cannot restore the distribution characteristics of carbon response with depth while eliminating surface interference, resulting in prediction results that mainly reflect the properties of the internal parent material, rather than accurately characterizing the local mechanical property weakening caused by decarburization and its depth distribution on the key stress surface of the connecting rod. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a method for predicting the mechanical properties of connecting rod materials based on a spectrometer. By continuously exciting the same point in the critical stress area of ​​the connecting rod and distinguishing between carbon response and non-carbon excitation state response, the surface interference attenuation boundary is determined. The carbon response before the boundary is corrected and the carbon response depth distribution is constructed. Then, the method is corrected by combining the baseline mechanical properties of the parent material and the amount of decarburization influence within the stress-sensitive depth. This solves the problem that existing methods are unable to retain surface decarburization information while eliminating surface interference, resulting in the local mechanical property prediction results not reflecting the weakening of the surface properties under critical stress.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A spectrometer-based method for predicting the mechanical properties of connecting rod materials includes the following steps: continuously exciting the same detection location in the critical stress zone of the connecting rod to obtain a spectral sequence arranged in excitation order; extracting the relative carbon response and the excitation state response determined by non-carbon spectral characteristics from the spectral sequence, and determining the surface interference attenuation boundary based on the excitation state response; correcting the relative carbon response before the boundary based on the migration characteristics of the relative carbon response after the surface interference attenuation boundary, forming a carbon response depth distribution and determining the stable spectral band of the parent material; determining the baseline mechanical properties of the parent material based on the stable spectral band of the parent material, and determining the decarburization influence within the stress-sensitive depth based on the carbon response depth distribution; and correcting the baseline mechanical properties of the parent material using the decarburization influence to obtain the local mechanical property prediction results at the detection location.

[0008] In a preferred embodiment, the step of continuously exciting at the same detection position in the critical stress area of ​​the connecting rod to obtain a spectral sequence arranged in the excitation order includes: acquiring the structural information and target load condition of the target connecting rod; determining the stress distribution under the target load condition based on the structural information; determining the critical stress area of ​​the connecting rod based on the stress distribution; determining the detection position based on the critical stress area of ​​the connecting rod; establishing a detection position identifier and excitation positioning conditions based on the surface morphology of the detection position, wherein the excitation positioning conditions are used to limit the center of the excitation spot, the excitation direction, and the detection distance; performing multiple continuous excitations at the detection position according to the excitation positioning conditions and preset excitation parameters, and sequentially acquiring the original spectrum and excitation information corresponding to each excitation; associating each original spectrum with the corresponding excitation information and excitation order according to the detection position identifier, and arranging them in the excitation order to form a spectral sequence.

[0009] In a preferred embodiment, the step of extracting the carbon relative response and the excitation state response determined by non-carbon spectral features from the spectral sequence, and determining the surface interference attenuation boundary based on the excitation state response, includes: performing spectral correction on each original spectrum in the spectral sequence to obtain a corrected spectrum corresponding to each excitation order; extracting the carbon net spectral line response and the steel substrate reference response based on the corrected spectrum, and determining the carbon relative response corresponding to each excitation order based on the change of the carbon net spectral line response relative to the steel substrate reference response, forming a carbon relative response sequence; extracting the surface adhesion response, plasma state response, and substrate excitation response corresponding to each excitation order based on the non-carbon spectral features that avoid the carbon characteristic bands in the corrected spectrum, forming a non-carbon state feature sequence; determining the stable state reference of the current detection position based on the stable change characteristics of the latter part of the non-carbon state feature sequence; determining the excitation state response corresponding to each excitation order based on the deviation of the non-carbon state features corresponding to each excitation order from the stable state reference, forming an excitation state response sequence; identifying the excitation state stable interval based on the excitation state response sequence, and determining the surface interference attenuation boundary within the excitation state stable interval based on the response level and change trend of the surface adhesion response.

[0010] In a preferred embodiment, determining the relative carbon response for each excitation sequence based on the change in the net carbon spectral response relative to the steel matrix reference response includes: determining a carbon feature window and its local background window in the calibration spectrum corresponding to the i-th excitation; determining the background response of the carbon feature window based on the local background window; and subtracting the background response from the spectral response of the carbon feature window to obtain the net carbon spectral response corresponding to the i-th excitation; determining an iron feature window and its local background window corresponding to at least one iron reference spectral line in the same calibration spectrum; determining the steel matrix reference response corresponding to the i-th excitation based on the net response of the iron feature window; determining the relative carbon response corresponding to the i-th excitation based on the ratio between the net carbon spectral response and the steel matrix reference response; and arranging the relative carbon responses corresponding to each excitation in the excitation sequence to form a relative carbon response sequence.

[0011] In a preferred embodiment, determining the stable state benchmark of the current detection position based on the stable change characteristics of the latter part of the non-carbon state feature sequence includes: selecting continuous effective spectra from the end of the spectral sequence to form a candidate spectral range for the latter part, and dividing the candidate spectral range for the latter part into multiple candidate state windows; determining the center value, fluctuation degree, and change trend of the non-carbon state features in each candidate state window; screening stable candidate windows that meet the benchmark window screening conditions from the multiple candidate state windows based on the center value, fluctuation degree, and change trend; merging continuous stable candidate windows from the stable candidate window corresponding to the end of the spectral sequence forward along the excitation order to obtain a state benchmark spectral segment; and establishing a stable state benchmark for the current detection position based on the benchmark center value and benchmark fluctuation value of each non-carbon state feature in the state benchmark spectral segment.

[0012] In a preferred embodiment, the step of correcting the carbon relative response before the boundary based on the migration characteristics of the carbon relative response after the surface interference attenuation boundary to form a carbon response depth distribution and determine the stable spectrum of the parent material includes: determining the equivalent excitation depth according to the cumulative ablation depth corresponding to each excitation sequence; determining a reliable migration spectrum from the carbon relative response after the boundary based on the surface interference attenuation boundary and the equivalent excitation depth; establishing a carbon response migration relationship based on the reliable migration spectrum; determining a carbon response interference correction relationship based on the carbon response migration relationship and the directed state deviation component before the boundary; correcting the carbon relative response before the boundary according to the carbon response interference correction relationship and combining it with the carbon relative response after the boundary to obtain a corrected carbon response sequence; associating the corrected carbon response sequence with the equivalent excitation depth to form a carbon response depth distribution; and determining the stable spectrum of the parent material based on the carbon response depth distribution.

[0013] In a preferred embodiment, determining the carbon response interference correction relationship based on the carbon response migration relationship and the directed state deviation component before the boundary includes: selecting consecutive excitation sequences adjacent to the surface interference attenuation boundary before the boundary, forming a boundary adjacency segment, based on the carbon response migration relationship; determining the migration reference response corresponding to each excitation sequence within the boundary adjacency segment, based on the difference between the measured relative carbon response and the migration reference response corresponding to each excitation sequence within the boundary adjacency segment; associating the carbon response interference quantity corresponding to each excitation sequence with the directed state deviation component corresponding to the same excitation, forming an interference correction sample; and determining the correspondence between the directed state deviation component and the carbon response interference quantity based on the interference correction sample, thereby obtaining the carbon response interference correction relationship at the current detection position.

[0014] In a preferred embodiment, determining the baseline mechanical properties of the base material based on the stable spectrum of the base material includes: determining a corresponding mechanical property prediction model based on the material state of the target link and the target mechanical properties to be predicted; extracting the base material spectral features that match the mechanical property prediction model from the stable spectrum of the base material; inputting the base material spectral features into the mechanical property prediction model to obtain the predicted mechanical property value corresponding to the stable spectrum of the base material; and determining the baseline mechanical properties of the base material at the current detection location based on the predicted mechanical property value corresponding to the stable spectrum of the base material.

[0015] In a preferred embodiment, determining the decarburization impact within the stress-sensitive depth based on the carbon response depth distribution includes: determining the parent material carbon response benchmark at the current detection location based on the corrected carbon relative response in the parent material's stable spectrum; determining the stress-sensitive depth at the current detection location based on the stress distribution along the surface normal at the detection location under the target load condition; determining the corrected carbon relative response corresponding to each material layer within the stress-sensitive depth range based on the carbon response depth distribution and the stress-sensitive depth; determining the degree of carbon response depletion corresponding to each material layer based on the difference between the corrected carbon relative response corresponding to each material layer and the parent material carbon response benchmark; determining the stress-sensitive weight corresponding to each material layer based on the stress state of each material layer under the target load condition; and determining the decarburization impact within the stress-sensitive depth based on the degree of carbon response depletion and the stress-sensitive weight corresponding to each material layer.

[0016] In a preferred embodiment, the step of correcting the baseline mechanical properties of the base material using the decarburization effect to obtain the predicted local mechanical properties at the detection location includes: determining the corresponding decarburization correction relationship based on the material state and target mechanical properties of the target connecting rod; determining the mechanical property attenuation ratio corresponding to the detection location based on the decarburization correction relationship and the decarburization effect; correcting the baseline mechanical properties of the base material based on the mechanical property attenuation ratio to obtain the predicted local mechanical properties at the detection location; and associating the predicted local mechanical properties with the corresponding detection location, target load condition, and target mechanical property type to form the predicted local mechanical properties.

[0017] The technical effects and advantages of the spectrometer-based method for predicting the mechanical properties of connecting rod materials in this invention are as follows:

[0018] This invention retains and utilizes the spectral information of the continuous excitation stage to identify the surface interference attenuation process using non-carbon spectral characteristics, and corrects the carbon response affected by interference. This reduces the influence of oxide layer, surface deposits, and excitation state fluctuations while retaining the true carbon response changes corresponding to decarburization of the connecting rod surface. Furthermore, by combining the carbon response depth distribution and the stress-sensitive depth of the critical stress area of ​​the connecting rod, the decarburization influence is corrected for the baseline mechanical properties of the base material. This makes the prediction results not only reflect the internal base material properties but also more reasonably characterize the local mechanical property weakening caused by decarburization and its depth distribution on the critical stress surface, improving the pertinence and interpretability of the performance prediction of the critical parts of the connecting rod. Attached Figure Description

[0019] Figure 1 This is a schematic flowchart of the method for predicting the mechanical properties of connecting rod materials based on a spectrometer, as described in this invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0021] Example 1, Figure 1 The present invention provides a method for predicting the mechanical properties of connecting rod materials based on a spectrometer, comprising the following steps:

[0022] S1, continuously excite at the same detection position in the critical stress area of ​​the connecting rod to obtain a spectral sequence arranged in the excitation order;

[0023] In this embodiment, the step of continuously exciting at the same detection position in the critical stress area of ​​the connecting rod to obtain a spectral sequence arranged in excitation order includes:

[0024] Obtain the structural information and target load conditions of the target link, determine the stress distribution under the target load conditions based on the structural information, and determine the key stress areas of the link based on the stress distribution;

[0025] The detection position is determined based on the key stress area of ​​the connecting rod, and the detection position mark and excitation positioning conditions are established based on the surface morphology of the detection position. The excitation positioning conditions are used to limit the center of the excitation spot, the excitation direction and the detection distance.

[0026] According to the excitation positioning conditions and preset excitation parameters, multiple continuous excitations are performed at the detection position, and the original spectrum and excitation information corresponding to each excitation are collected in sequence.

[0027] Based on the detection location identifier, each original spectrum is associated with the corresponding excitation information and excitation order, and arranged according to the excitation order to form the spectral sequence.

[0028] In this embodiment, the target connecting rod can be an engine connecting rod, a compressor connecting rod, or other rod-shaped force transmission component subjected to cyclic tensile and compressive loads. The structural information includes at least one of the following: the location of the large end hole, the location of the small end hole, the rod body profile, the rounded transition region, the location of the fracture surface, the material thickness, and the surface curvature.

[0029] The target load condition refers to the typical stress state that the connecting rod experiences during actual use or design verification, including at least one of tensile load conditions, compressive load conditions, inertial load conditions, and alternating load conditions. Mechanical calculations or finite element analysis are performed based on the connecting rod structural parameters and the target load condition to obtain the stress distribution under the target load condition; historical failure locations are used to supplement the verification of the key stress areas determined by the stress distribution.

[0030] The critical stress zone of the connecting rod refers to the area where stress concentration is prone to occur under the target load condition, and where the surface material condition has a significant impact on the initiation of fatigue cracks or local load-bearing capacity. It includes at least one of the following: the rounded transition zone of the large end hole, the rounded transition zone of the small end hole, the transition zone between the rod body and the large end, the transition zone between the rod body and the small end, and the area adjacent to the fracture surface.

[0031] In one implementation, the region where the stress value under the target load condition is higher than a preset stress threshold is identified as the critical stress zone of the connecting rod. The preset stress threshold can be determined based on the allowable stress of the material or a preset proportion of the maximum equivalent stress of the connecting rod. For example, the region where the equivalent stress is not lower than 70% of the maximum equivalent stress is identified as a candidate critical stress zone, and then a location where the excitation spot can be stably arranged is selected from the candidate critical stress zones as the detection location.

[0032] The detection location refers to a local surface position selected within the critical stress area of ​​the connecting rod for continuous multiple spectral excitations. The detection location is preferably located in an area with high stress and relatively gentle surface curvature changes to reduce the impact of excitation spot shift, focal length changes, and spectral acquisition angle changes on the detection results.

[0033] After selecting the detection location, a detection identifier is established for that location. The detection identifier may include at least one of the following: link number, critical stress zone number, detection point number, detection coordinates, surface normal, and detection direction, to ensure that the spectral sequence corresponds to the specific structural location of the link.

[0034] When pre-treating the inspection location, easily detachable dust or loose particles can be removed, but grinding, polishing, or deep cleaning that would alter the material's surface condition should be avoided. This preserves the actual oxide layer, surface adhesion, and decarburized layer information of the connecting rod. If the object being inspected has anti-rust oil or a process coating, its surface condition can be recorded, and the corresponding surface interference can be identified in subsequent steps using the excitation state response.

[0035] Continuous excitation at the same detection position refers to maintaining the center position of the excitation spot unchanged during continuous acquisition, allowing the excitation areas corresponding to different excitation sequences to gradually extend into the connecting rod along the surface normal of the detection position. To reduce positional errors during continuous excitation, the connecting rod can be fixed using a clamp, and the detection position and focal length can be determined through visual positioning, contour positioning, or laser ranging.

[0036] The preset excitation parameters include at least one of the following: single excitation energy, excitation pulse width, time interval between adjacent excitations, spectral acquisition delay, spectral integration time, and number of consecutive excitations. During the continuous excitation process of the same spectral sequence, each excitation uses the same set excitation energy, pulse width, acquisition delay, and integration time. The deviation of the actual excitation energy from the set excitation energy is recorded. When the deviation exceeds the allowable range, the corresponding spectrum is set as an abnormal spectrum and is not used for subsequent carbon response migration relationship fitting and depth distribution construction.

[0037] After each excitation, the spectrometer collects the light radiation generated in the excitation region to obtain the corresponding raw spectrum. The raw spectrum includes at least wavelength information and spectral intensity information within a preset wavelength range, and retains the characteristic spectral lines of carbon, the reference spectral lines of iron, the spectral lines of non-carbon elements, and the continuous background response.

[0038] The excitation information includes at least one of the following: excitation order, excitation time, single excitation energy, cumulative excitation energy, acquisition delay, integration time, and device status. Each original spectrum is designated as the first spectrum, second spectrum, and so on up to the Nth spectrum according to the excitation order, and then associated with the corresponding excitation information to form a spectral sequence.

[0039] In this step, the original spectrum is not directly deleted due to spectral intensity fluctuations, abnormal carbon response, or surface deposit response during the initial stage of continuous excitation. For spectra exhibiting signal loss, detector saturation, or device trigger failure, an abnormal state marker is set, and in subsequent processing, the marker status is used to determine whether to participate in the establishment of a stable state benchmark, fitting of carbon response migration relationships, and correction of carbon response interference.

[0040] Through the above processing, the obtained spectral sequence retains the spectral information of the gradual change from the actual surface state to the internal parent material at the detection location, as well as the excitation order and excitation conditions corresponding to each spectrum, providing basic data for subsequent extraction of carbon relative response, determination of surface interference attenuation boundary and construction of carbon response depth distribution.

[0041] S2, extract the relative carbon response and the excitation state response determined by the non-carbon spectral characteristics from the spectral sequence, and determine the surface interference attenuation boundary based on the excitation state response;

[0042] In this embodiment, the relative carbon response refers to the relative characterization value of the net carbon spectral line response in the same excitation spectrum relative to the reference response of the steel matrix, used to reflect the change of the carbon response at the detection position with the excitation sequence during continuous excitation. The relative carbon response does not directly represent the absolute content of carbon.

[0043] The excitation state response refers to the degree of deviation of the current excitation state from the stable excitation state and the degree of state change between adjacent excitations, determined based on the non-carbon state characteristics in the same excitation spectrum. It is used to characterize the influence of surface deposits, oxide layer, plasma state and the degree of steel substrate participation on the current spectrum.

[0044] The extraction of the carbon relative response from the spectral sequence and the excitation state response determined by non-carbon spectral features, and the determination of the surface interference attenuation boundary based on the excitation state response, include:

[0045] Spectral correction is performed on each original spectrum in the spectral sequence to obtain the corrected spectrum corresponding to each excitation order;

[0046] Based on the corrected spectrum, the net carbon spectral response and the steel matrix reference response are extracted. According to the change of the net carbon spectral response relative to the steel matrix reference response, the relative carbon response corresponding to each excitation order is determined to form a carbon relative response sequence.

[0047] Based on the non-carbon spectral features that avoid the carbon characteristic bands in the corrected spectrum, the surface adhesion response, plasma state response and matrix excitation response corresponding to each excitation sequence are extracted to form a non-carbon state feature sequence.

[0048] Based on the stable change characteristics of the latter part of the non-carbon state characteristic sequence, the stable state benchmark of the current detection position is determined;

[0049] Based on the difference between the non-carbon state characteristics corresponding to each excitation sequence and the stable state benchmark, the directed state deviation components of each non-carbon state characteristic are determined, and the comprehensive state deviation amount and excitation state response corresponding to each excitation sequence are determined based on the directed state deviation components, forming an excitation state response sequence.

[0050] The excitation state stable interval is identified based on the excitation state response sequence, and the surface interference attenuation boundary is determined within the excitation state stable interval based on the response level and change trend of the surface adhesion response.

[0051] The carbon relative response sequence, each directional state deviation component, the excitation state response sequence, and the surface disturbance attenuation boundary are correlated and output, which are then used to divide and correct the carbon relative response before and after the surface disturbance attenuation boundary.

[0052] In this embodiment, the relative response of carbon is obtained specifically by the following method:

[0053] For the correction spectrum corresponding to the i-th excitation, a carbon feature window is set in the band where the carbon element characteristic spectral lines are located, and a local background window is set on at least one side of the carbon feature window; the local background response corresponding to the carbon feature window is determined according to the spectral response in the local background window, and the local background response is subtracted from the spectral response in the carbon feature window to obtain the net carbon spectral line response corresponding to the i-th excitation.

[0054] One or more iron reference spectral lines are selected from the same calibration spectrum. Iron feature windows and corresponding local background windows are set in the bands where each iron reference spectral line is located. The net response of each iron reference spectral line is obtained by subtracting the local background response. The steel matrix reference response corresponding to the i-th excitation is determined based on the net response of each iron reference spectral line.

[0055] The relative carbon response corresponding to the i-th excitation can be determined by the following formula:

[0056]

[0057] in, Let be the relative carbon response corresponding to the i-th excitation. Let be the net carbon spectral line response in the i-th spectrum. Let be the reference response of the steel matrix corresponding to the i-th spectrum.

[0058] When multiple iron reference spectral lines are selected, it is determined whether each iron reference spectral line has spectral line overlap, detector saturation, or signal-to-noise ratio lower than the preset requirements, and iron reference spectral lines that do not meet the usage conditions are excluded; the net response of the remaining iron reference spectral lines is scaled uniformly, and the steel matrix reference response is determined based on the median, truncated mean, or weighted mean of the net response after scale uniformity.

[0059] In this embodiment, the surface adhesion response is used to characterize the degree to which oxides or process residues on the connecting rod surface participate in the current excitation. Based on the forging, heat treatment, shot peening, cleaning, and rust prevention processes of the target connecting rod, at least one surface adhesion characteristic element or characteristic spectral band is predetermined, and the corresponding net spectral line response is extracted from each corrected spectrum.

[0060] The surface adhesion characteristic elements or characteristic spectral bands are determined by comparing the spectral differences of the same connecting rod material in the state of surface adhesion retention and the state of surface adhesion removal, including at least one of oxygen element characteristic spectral lines, shot peening medium residual element characteristic spectral lines, cleaning medium residual element characteristic spectral lines and rust prevention medium characteristic spectral bands.

[0061] The surface adhesion response corresponding to the i-th excitation is determined by the following formula:

[0062]

[0063] in, For the surface adhesion response corresponding to the i-th excitation, This represents the net response of the p-th term of the surface adhesion feature in the i-th corrected spectrum. Let i be the reference response of the steel matrix corresponding to the i-th spectrum. denoted by , where is the weight corresponding to the p-th surface adhesion feature, and r is the number of surface adhesion features used.

[0064] When only one surface adhesion feature is used, the surface adhesion response is the ratio of the net response of that surface adhesion feature to the reference response of the steel substrate.

[0065] The plasma state response is used to characterize the plasma state generated by the current excitation. Preferably, at least one iron atom spectral line and at least one iron ion spectral line are selected from the same calibration spectrum, and the plasma state response corresponding to the i-th excitation is determined according to the ratio of the net response of the iron ion spectral line to the net response of the iron atom spectral line.

[0066]

[0067] in, The plasma state response corresponding to the i-th excitation is... This represents the net response of the iron ion spectral line. The net response of the iron atom spectral lines. A positive number set to prevent the denominator from being zero.

[0068] The matrix excitation response is used to characterize the degree to which the steel matrix participates in plasma formation during the current excitation. The matrix excitation response corresponding to the i-th excitation is determined according to the following formula:

[0069]

[0070] in, This represents the matrix excitation response corresponding to the i-th excitation. For the steel matrix reference response, The effective integral response of the full spectrum after avoiding the carbon characteristic band and the surface adhesion characteristic band in the i-th correction spectrum.

[0071] The surface adhesion response, plasma state response, and matrix excitation response corresponding to the i-th excitation are combined to form the non-carbon state characteristics. .

[0072] The process of determining the stable state benchmark at the current detection position based on the non-carbon state characteristics of the latter part of the spectral sequence includes:

[0073] A predetermined number of consecutive valid spectra are selected from the end of the spectral sequence to form a subsequent candidate spectral range, and the subsequent candidate spectral range is divided into multiple candidate state windows.

[0074] The center value, fluctuation degree, and variation trend of non-carbon state characteristics with excitation order were determined for each candidate state window.

[0075] Based on the central value, fluctuation level and trend of change, stable candidate windows that meet the baseline window selection conditions are selected from multiple candidate state windows;

[0076] Starting from the stable candidate window corresponding to the end of the spectral sequence, consecutive stable candidate windows are merged forward along the excitation order to obtain the state reference spectral band;

[0077] Based on the reference center value and reference fluctuation value of each non-carbon state characteristic within the aforementioned state reference spectrum, a stable state reference for the current detection position is established.

[0078] The baseline window selection criteria include: the center value of the surface adhesion response within the candidate state window is not higher than the adhesion response limit, and the trend of the surface adhesion response is within the adhesion change limit; the fluctuation degree and trend of the plasma state response within the candidate state window are both within the corresponding baseline stability limit; and the fluctuation degree and trend of the matrix excitation response within the candidate state window are both within the corresponding baseline stability limit.

[0079] For the j-th non-carbon state characteristic within the state reference spectrum, the median or robust average value of this non-carbon state characteristic is determined as the reference center value. The degree of dispersion of this non-carbon state characteristic relative to the benchmark center value is determined as the benchmark fluctuation value. .

[0080] For the j-th non-carbon state characteristic corresponding to the i-th excitation, the directed state deviation component is determined based on the difference between this non-carbon state characteristic and the corresponding reference center value, as well as the corresponding reference fluctuation value:

[0081]

[0082] in, Let be the directed state deviation component of the j-th non-carbon state characteristic corresponding to the i-th excitation. Let j be the non-carbon state characteristic corresponding to the i-th excitation.

[0083] Based on the directed state deviation components corresponding to the same excitation sequence, determine the overall state deviation:

[0084] in, Let i be the overall state deviation corresponding to the i-th excitation. is the weight corresponding to the j-th non-carbon state feature, and m is the number of non-carbon state features.

[0085] The state transitions corresponding to the i-th excitation are determined according to the following formula:

[0086]

[0087] The overall state deviation corresponding to the i-th excitation and adjacent state changes Together they are determined to be the excitation state response corresponding to this excitation.

[0088] In this embodiment, the conditions for stabilizing the excitation state include: the overall state deviation corresponding to the current excitation sequence is not higher than a preset state deviation threshold; the adjacent state changes corresponding to the current excitation sequence are not higher than a preset state change threshold; the spectrum corresponding to the current excitation sequence does not have detector saturation, spectral loss, excitation failure, iron reference spectral signal-to-noise ratio lower than a preset requirement, or necessary non-carbon state features cannot be extracted; the above conditions are maintained continuously for a preset number of times starting from the same excitation sequence.

[0089] The preset number of excitations is at least three consecutive excitations, preferably 3 to 5. When multiple consecutive excitation sequences all satisfy the above-mentioned excitation state stability condition, the corresponding range of consecutive excitation sequences is determined as the excitation state stability interval.

[0090] The state deviation threshold and state change threshold are determined based on the normal fluctuation range of the comprehensive state deviation and adjacent state changes within the state reference spectrum, or obtained by repeatedly exciting and calibrating the same connecting rod material under the same excitation parameters.

[0091] Within the stable range of the excitation state, a predetermined number of consecutive surface adhesion responses are selected starting from each excitation sequence to form a surface adhesion judgment window. The center value of the surface adhesion response within the surface adhesion judgment window and the trend of the surface adhesion response with the excitation sequence are then determined.

[0092] When the center value of the surface adhesion response within the surface adhesion judgment window is not higher than the adhesion response limit, and the trend of the surface adhesion response change is within the adhesion change limit, it is determined that the corresponding surface adhesion response has entered the allowable range of the adhesion response and no longer continues to decay.

[0093] Judging from the excitation order backward from the starting position of the stable interval of the excitation state, the first excitation order that meets the above conditions is determined as the surface interference attenuation boundary.

[0094] The surface interference attenuation boundary only indicates that, starting from this excitation sequence, the influence of surface deposits, oxide layers, and changes in excitation state on the relative carbon response has been reduced to an acceptable range, and does not indicate that the relative carbon response has reached the level of the parent material.

[0095] S3, based on the migration characteristics of the carbon relative response after the surface interference attenuation boundary, corrects the carbon relative response before the boundary, forms the carbon response depth distribution and determines the stable spectrum of the parent material;

[0096] In this embodiment, the migration characteristics refer to the relationship between the relative carbon response and the equivalent excitation depth as the carbon relative response approaches the stable level of the parent material, under the condition that the surface disturbance has decayed and the excitation state remains stable. This includes the migration direction, the rate of change, and the continuous relationship between the relative carbon response and the equivalent excitation depth.

[0097] The carbon relative response based on the migration characteristic correction boundary before the boundary after the surface interference attenuation boundary forms the carbon response depth distribution and determines the stable spectral band of the parent material, including:

[0098] Based on the calibration relationship of the cumulative ablation depth formed by the calibration sample with corresponding material state, surface state and detection geometry to the target connecting rod under the same excitation parameters, the equivalent excitation depth corresponding to the ablated material layer of each excitation is determined.

[0099] From the carbon relative response after the surface interference attenuation boundary, exclude the carbon relative response with the corresponding spectrum set with abnormal state mark, identify the continuous spectrum segment where the excitation state remains stable, the carbon relative response changes continuously in the same direction, and the absolute value of the rate of change of the carbon relative response with the equivalent excitation depth is continuously higher than the stable rate of change limit of the parent material, and determine it as the reliable migration spectrum segment.

[0100] A carbon response migration relationship is established based on the relationship between the relative carbon response and the equivalent excitation depth within the reliable migration spectrum band, and the carbon response migration relationship is locally extended to the boundary adjacent segment before the boundary to obtain the corresponding migration reference response.

[0101] Based on the difference between the measured relative carbon response and the migration reference response within the boundary adjacent segment, and the directional state deviation component of the corresponding non-carbon state characteristics, the carbon response interference correction relationship at the current detection position is determined.

[0102] The carbon relative response before the boundary is corrected according to the carbon response interference correction relationship, and the corrected carbon relative response before the boundary and the carbon relative response after the boundary are combined according to the excitation order to obtain the corrected carbon response sequence.

[0103] The relative carbon responses in the corrected carbon response sequence are correlated with the corresponding equivalent excitation depths to form a carbon response depth distribution.

[0104] Identify continuous spectral segments in the latter part of the carbon response depth distribution that are continuously within the stable range of the corresponding parent material, maintain a stable excitation state, and reach the preset minimum number of effective spectra. These segments are then identified as stable spectral segments of the parent material.

[0105] In this embodiment, the cumulative ablation depth refers to the normal distance between the bottom of the ablation pit and the surface of the connecting rod before excitation, after a preset number of excitations have been completed.

[0106] When calibrating the cumulative ablation depth, multiple independent calibration locations are selected on the calibration sample, and continuous excitation is performed at different numbers of times. For example, continuous excitation is performed once at the first calibration location, and twice at the second calibration location, and so on, increasing the number of excitations. The depth of the ablation pit formed at each calibration location is measured to obtain the correspondence between the number of excitations and the cumulative ablation depth.

[0107] When the number of excitations to be determined is not directly set to a corresponding calibration position, the cumulative ablation depth corresponding to the number of excitations is determined by interpolation based on the adjacent calibration numbers and their cumulative ablation depths.

[0108] The i-th excitation continues to remove material into the connecting rod based on the ablation pit formed by the (i-1)-th excitation. Therefore, the spectrum generated by the i-th excitation mainly reflects the comprehensive spectral response of the material layer between the cumulative ablation depth after the (i-1)-th excitation and the cumulative ablation depth after the i-th excitation, rather than only reflecting the material state at the bottom of the pit after the i-th excitation.

[0109] Based on this, the cumulative ablation depth after the (i-1)th excitation and the cumulative ablation depth after the ith excitation are defined as the ablation material layer corresponding to the ith excitation, and the middle depth of this ablation material layer along the surface normal at the detection position is defined as the equivalent excitation depth corresponding to the ith excitation.

[0110] The limit for the stable change rate of the parent material is determined based on the normal range of change of the relative carbon response with the equivalent excitation depth when the same connecting rod material without surface carbon anomalies is continuously excited under the same excitation parameters. If the absolute value of the change rate of the relative carbon response is higher than the limit for the stable change rate of the parent material, it indicates that the current relative carbon response is still in the process of migrating from the surface anomaly region to the parent material level, and has not yet entered the stable state of the parent material.

[0111] Further, determining the carbon response interference correction relationship at the current detection position based on the difference between the measured relative carbon response and the migration reference response within the boundary adjacent segment, and the directional state deviation component of the corresponding non-carbon state characteristics, includes:

[0112] Based on the surface interference attenuation boundary, a predetermined number of consecutive excitation sequences with corresponding spectra without abnormal state markers are selected forward along the excitation sequence to form a boundary adjacent segment.

[0113] Based on the carbon response migration relationship, the migration reference response corresponding to each excitation sequence within the boundary adjacent segment is determined, and the difference between the measured carbon relative response and the migration reference response corresponding to each excitation sequence is calculated to obtain the carbon response interference difference.

[0114] The carbon response interference difference corresponding to each excitation order within the boundary adjacent segment is paired with the directional state deviation components of each non-carbon state characteristic corresponding to the same excitation order to form interference correction samples.

[0115] Based on the interference correction samples, the correspondence between the carbon response interference difference and each directional state deviation component is fitted to determine the interference correction coefficients corresponding to each non-carbon state feature.

[0116] Based on the interference correction coefficient, establish the carbon response interference correction relationship at the current detection location.

[0117] In this embodiment, the boundary adjacency segment consists of multiple excitation sequences that are consecutively adjacent to the surface interference attenuation boundary, preceding it. The number of excitation sequences within the boundary adjacency segment is not less than the number of non-carbon state features used to establish the carbon response interference correction relationship, ensuring that the interference correction coefficient can be determined based on the interference correction sample.

[0118] For the i-th excitation within the boundary adjacent segment, its carbon response interference difference is determined according to the following formula:

[0119]

[0120] in, Let be the carbon response interference difference corresponding to the i-th excitation. Let i be the measured relative carbon response corresponding to the i-th excitation. The reference response for the i-th excitation is obtained by local extension of the carbon response migration relationship.

[0121] The migration reference response represents the expected relative carbon response for the i-th excitation, assuming the carbon response migration trend remains intact after the boundary and without considering the influence of surface disturbances before the boundary. Therefore, the carbon response disturbance difference is used to characterize the deviation of the measured relative carbon response from the local carbon response migration trend.

[0122] Let the directed state deviation component of the j-th non-carbon state characteristic corresponding to the i-th excitation be . The correspondence between the carbon response disturbance difference and each directional state deviation component is expressed as:

[0123]

[0124] in, The number of non-carbon state characteristics. Here is the interference correction coefficient corresponding to the j-th non-carbon state characteristic. The residual error not explained by the directed state deviation component. Let be the directed state deviation component of the j-th non-carbon state characteristic corresponding to the i-th excitation.

[0125] Based on the interference correction samples within the boundary adjacent segments, the interference correction coefficients are determined using least squares fitting or least squares fitting with regularization constraints. Preferably, they are determined according to the following formula:

[0126]

[0127] in, This is the set of excitation orders corresponding to the boundary adjacent segments. This is a regularization coefficient used to limit the excessive increase of the interference correction coefficient. When the number of samples within the boundary adjacent segment is sufficient and there is no significant correlation between the directed state deviation components, the regularization coefficient can be zero.

[0128] The carbon response interference correction relationship at the current detection location is composed of various interference correction coefficients:

[0129]

[0130] in, The carbon response disturbance amount is determined based on the non-carbon state characteristics of the i-th excitation.

[0131] The interference correction coefficient is determined by the boundary adjacent segment data of the current detection position, instead of directly using the fixed correction coefficient of other detection positions, thereby adapting to the differences in oxide layer state, surface adhesion state and excitation conditions of different connecting rod detection positions.

[0132] For the i-th excitation before the surface disturbance attenuation boundary, the carbon response disturbance amount is determined based on the corresponding directional state deviation component and the disturbance correction coefficient, and the carbon response disturbance amount is subtracted from the measured carbon relative response to obtain the corrected carbon relative response:

[0133]

[0134] in, Let be the corrected carbon relative response corresponding to the i-th excitation.

[0135] When the number of excitation sequences without abnormal state markers within the boundary adjacent segment is insufficient, or when the fitting residual of the established carbon response interference correction relationship is higher than the preset residual limit, the carbon response interference correction relationship is not used to correct the relative carbon response before the boundary, and the number of consecutive excitations is increased, the boundary adjacent segment is expanded, or the current detection position is re-detected.

[0136] The stable spectral range of the parent material refers to a continuous spectral range in the latter part of the carbon response depth distribution where the rate of change and fluctuation of the corrected relative carbon response both enter the stable range of the parent material, and the corresponding excitation state remains stable. The stable spectral range of the parent material is used to characterize the stable spectral response state of the parent material within the current detection location.

[0137] S4. Determine the baseline mechanical properties of the parent material based on the stable spectrum of the parent material, and determine the decarburization influence within the stress-sensitive depth based on the carbon response depth distribution.

[0138] In this embodiment, the reference mechanical properties of the parent material refer to the target mechanical properties of the parent material at the current detection location, which are predicted based on the spectral characteristics in the stable spectrum of the parent material without considering the influence of surface decarburization. These properties are used as a benchmark for subsequently determining the predicted results of local mechanical properties.

[0139] The target mechanical properties include at least one of hardness, yield strength, tensile strength, and fatigue strength. When multiple target mechanical properties need to be predicted, corresponding mechanical property prediction models are established for different target mechanical properties, and the corresponding base material reference mechanical properties are determined for each.

[0140] The decarburization impact refers to the degree of carbon deficit in the corrected relative carbon response of each material layer relative to the baseline carbon response of the parent material within the stress-sensitive depth range. It is a quantified value obtained by weighting and combining the stress sensitivity of each material layer under the target load condition. The decarburization impact is used to characterize the comprehensive influence of the degree, depth, and location of surface decarburization on local mechanical properties, and is not directly equivalent to a reduction in mechanical properties.

[0141] The determination of the reference mechanical properties of the base material based on the stable spectrum of the base material includes:

[0142] Call the mechanical property prediction model corresponding to the material grade, heat treatment state, and target mechanical properties of the target link;

[0143] Extract the spectral features required by the mechanical property prediction model from each spectrum in the stable spectrum of the parent material that has no abnormal state markers, and form the performance prediction features corresponding to each spectrum.

[0144] The performance prediction features corresponding to each spectrum are input into the mechanical performance prediction model to obtain the mechanical performance prediction values ​​corresponding to each spectrum in the stable spectrum range of the parent material.

[0145] Based on the center value of each predicted mechanical property within the stable spectrum of the base material, the reference mechanical properties of the base material corresponding to the current testing location are determined.

[0146] In this embodiment, the mechanical property prediction model is established using calibration specimens. The calibration specimens have the same material grade and similar heat treatment conditions as the target connecting rod, and cover the calibration range of the target mechanical properties.

[0147] Specifically, stable spectra corresponding to the parent material inside each calibration sample are collected, and the measured mechanical properties of each calibration sample are obtained through hardness test, tensile test or fatigue test; the relative response of carbon in each stable spectrum and the relative response of alloying elements related to the target mechanical properties are extracted, and the correspondence between the spectral features and the measured mechanical properties is established to form a mechanical property prediction model.

[0148] The relative response of alloying elements related to the target mechanical properties may include the ratio of the net spectral response of at least one element selected from manganese, chromium, molybdenum, silicon, nickel and vanadium to the reference response of the steel matrix. The specific spectral features used are consistent with those used when establishing the mechanical property prediction model.

[0149] In one implementation, partial least squares regression can be used to establish a mechanical performance prediction model; alternatively, calibrated and validated multiple linear regression, support vector regression, or tree models can be used to establish a mechanical performance prediction model.

[0150] Let the performance prediction feature corresponding to the i-th spectrum be... The prediction model corresponding to the target mechanical properties is Then the predicted mechanical properties corresponding to the i-th spectrum are:

[0151]

[0152] The median or robust average of the predicted mechanical properties corresponding to each spectrum within the stable spectral range of the base material is determined as the baseline mechanical properties of the base material.

[0153]

[0154] in, For the reference mechanical properties of the base material, This is the set of excitation sequences corresponding to the stable spectral bands of the parent material.

[0155] The determination of the decarburization influence within the stress-sensitive depth based on the carbon response depth distribution includes:

[0156] The carbon response benchmark of the parent material at the current detection location is determined based on the corrected carbon relative response within the stable spectrum of the parent material. The carbon response benchmark of the parent material is used as a comparison benchmark to determine whether the carbon response of each material layer at each depth is lower than the normal level of the parent material.

[0157] Based on the stress change along the surface normal at the detection location under the target load condition, determine the stress-sensitive depth corresponding to the current detection location;

[0158] Extract each corrected relative carbon response within the stress-sensitive depth range from the carbon response depth distribution, compare each corrected relative carbon response with the parent material carbon response benchmark, and determine the degree of carbon response loss corresponding to each material layer.

[0159] The stress sensitivity weight of each material layer is determined based on the stress level and thickness of each material layer under the target load condition.

[0160] The amount of decarbonization impact within the stress-sensitive depth is determined based on the degree of carbon response deficit and stress sensitivity weight of each material layer.

[0161] In this embodiment, the carbon response benchmark of the parent material is determined based on the median or robust average value of the relative responses of each corrected carbon within the stable spectrum of the parent material:

[0162]

[0163] in, As the carbon response benchmark of the parent material, Let be the corrected carbon relative response corresponding to the i-th excitation.

[0164] In this embodiment, the stress-sensitive depth refers to the depth range extending from the surface of the detection location along the surface normal into the inside of the connecting rod, where changes in the material state can significantly affect the local bearing capacity of the current detection location.

[0165] Based on the stress distribution under the target load condition, the equivalent stress change along the surface normal at the detection location is extracted. When the equivalent stress decreases to below a preset proportion of the surface equivalent stress or the maximum equivalent stress along the normal path, and remains below this proportion in subsequent depths, the corresponding depth is determined as the stress-sensitive depth.

[0166] The preset ratio can be determined based on the linkage structure, target load conditions, and target mechanical properties. In one embodiment, a continuous depth range where the equivalent stress is not less than 30% to 50% of the maximum equivalent stress along the normal path can be defined as the stress-sensitive depth.

[0167] For the i-th material layer within the stress-sensitive depth range, the degree of carbon response deficit is determined based on the difference between the corrected carbon relative response of this material layer and the baseline carbon response of the parent material.

[0168]

[0169] in, The degree of carbon response deficit corresponding to the i-th material layer.

[0170] When the corrected carbon relative response is not lower than the parent material carbon response benchmark, the corresponding carbon response deficit is determined to be zero, so that the decarbonization impact only characterizes the decarbonization impact when the carbon response is lower than the parent material level, without including local carbon increase or normal fluctuations in the decarbonization impact.

[0171] Based on the average equivalent stress of the i-th material layer under the target load condition and the thickness of the material layer, determine the unnormalized stress sensitivity quantity corresponding to the material layer.

[0172] The unnormalized stress sensitivity values ​​of all material layers within the stress-sensitive depth range are normalized to obtain the stress sensitivity weights corresponding to each material layer:

[0173]

[0174] in, This represents the stress-sensitive weight corresponding to the i-th material layer. Let be the average equivalent stress of the i-th material layer under the target load condition. Let be the equivalent thickness of the i-th material layer. It is a collection of material layers within the stress-sensitive depth range.

[0175] The decarbonization impact is determined based on the degree of carbon deficit in each material layer within the stress-sensitive depth range and the stress-sensitive weight:

[0176]

[0177] in, This represents the decarburization effect within the stress-sensitive depth.

[0178] The decarburization effect is a dimensionless quantity. The larger the value, the more severe the carbon response loss within the stress-sensitive depth, or the more concentrated the carbon response loss is in the shallow material layer with higher stress level and greater impact on local load-bearing capacity.

[0179] When the boundary of the stress-sensitive depth is located between two adjacent equivalent excitation depths, the effective thickness of the material layer participating in the calculation of the decarburization influence is determined based on the overlap thickness between the stress-sensitive depth and the corresponding material layer.

[0180] S5 uses the decarburization effect to correct the baseline mechanical properties of the base material, and obtains the predicted results of the local mechanical properties at the test location.

[0181] In this embodiment, the predicted local mechanical properties refer to the predicted actual effective mechanical properties of the detection location under the target load condition, obtained by further considering the degree of decarburization and its depth distribution within the stress-sensitive depth of the detection location, based on the baseline mechanical properties corresponding to the composition and heat treatment state of the base material itself.

[0182] The method of correcting the baseline mechanical properties of the base material using the decarburization effect to obtain the predicted results of the local mechanical properties at the detection location includes:

[0183] Call upon the decarburization correction relationship corresponding to the material grade, heat treatment state, and target mechanical properties of the target connecting rod;

[0184] Input the amount of decarburization influence within the stress-sensitive depth into the decarburization correction relationship to determine the mechanical property attenuation ratio corresponding to the current detection location;

[0185] The baseline mechanical properties of the base material are corrected downward based on the mechanical property attenuation ratio to obtain the predicted local mechanical properties at the detection location.

[0186] The predicted local mechanical properties are associated with the link number, critical stress zone number, detection location, target load condition, and target mechanical property type to form the predicted local mechanical properties.

[0187] In this embodiment, the decarburization correction relationship is used to characterize the correspondence between the amount of decarburization influence and the reduction ratio of the target mechanical property relative to the baseline mechanical property of the parent material. For different target mechanical properties such as hardness, yield strength, tensile strength, and fatigue strength, corresponding decarburization correction relationships are established respectively.

[0188] Let the decarburization effect within the stress-sensitive depth be... The mechanical property degradation ratio is determined based on the corresponding decarburization correction relationship:

[0189]

[0190] in, The percentage decrease in mechanical properties corresponding to the target mechanical properties. The decarburization correction relationship is the one corresponding to the target mechanical properties.

[0191] The mechanical property attenuation ratio is not less than zero and not greater than one, and remains unchanged as the decarburization effect increases. When the decarburization effect is zero, the mechanical property attenuation ratio is zero.

[0192] The predicted local mechanical properties at the detection location are determined according to the following formula:

[0193]

[0194] in, To predict the local mechanical properties at the detection location, The reference mechanical properties are those of the base material.

[0195] Therefore, when there is no carbon response deficit within the stress-sensitive depth of the detection location, the decarburization effect is zero, and the predicted local mechanical properties are equal to the reference mechanical properties of the base material. As the degree of carbon response deficit increases or the carbon response deficit concentrates in the shallow material layer with higher stress levels, the decarburization effect increases, and the predicted local mechanical properties decrease accordingly.

[0196] The decarburization correction relationship is established through calibration of samples, specifically including:

[0197] Multiple calibration samples with the same material grade and heat treatment state as the target connecting rod, but with different surface decarburization states, were selected.

[0198] Each calibration sample was tested separately to determine the baseline mechanical properties of the base material and the amount of decarburization effect corresponding to each calibration sample;

[0199] Using mechanical testing methods corresponding to the target mechanical properties, the mechanical properties of the stress-sensitive areas of each calibrated specimen are measured to obtain the corresponding measured local mechanical properties.

[0200] Based on the difference between the reference mechanical properties of the base material and the measured local mechanical properties of each calibration sample, the attenuation ratio of the measured mechanical properties of each calibration sample is determined.

[0201] Based on the decarburization effect and the measured mechanical property attenuation ratio of each calibration sample, a decarburization correction relationship is established.

[0202] For the s-th calibration sample, the attenuation ratio of its measured mechanical properties is determined according to the following formula:

[0203]

[0204] in, The reference mechanical properties of the base material for the s-th calibration sample are given. The measured local mechanical properties of the s-th calibration sample are given. This represents the corresponding measured mechanical property attenuation ratio.

[0205] The decarburization effect of each calibration sample and the measured mechanical property decay ratio are used to form calibration data pairs, which are then arranged in order of increasing decarburization effect.

[0206] In one embodiment, the calibration data pair is monotonically fitted to establish a decarbonization correction relationship in which the mechanical property attenuation ratio remains unchanged when the decarbonization influence increases.

[0207] In another embodiment, piecewise linear interpolation is performed based on adjacent calibration data points. When the decarburization effect of the connecting rod under test is located between two adjacent calibration decarburization effects, the corresponding mechanical performance attenuation ratio is determined based on its position between the two calibration points.

[0208] When using piecewise linear interpolation, let the decarburization effect of the connecting rod under test be considered. Located at adjacent calibration points and Between these values, the corresponding mechanical property attenuation ratio is:

[0209]

[0210] in, , These represent the measured mechanical property attenuation ratios corresponding to two adjacent calibration points.

[0211] When the decarburization effect of the connecting rod under test exceeds the maximum calibration range used when establishing the decarburization correction relationship, instead of directly performing unconstrained extrapolation, the predicted result of the local mechanical properties is marked as exceeding the calibration range, and a calibration sample with the corresponding decarburization state is added or the test result is verified by actual mechanical test.

[0212] When the baseline mechanical properties of the base material cannot be effectively determined, the decarburization effect is marked as low confidence, or the decarburization correction relationship does not correspond to the material grade, heat treatment state, or target mechanical properties of the target connecting rod, the predicted value of local mechanical properties will not be output, and the current detection position will be re-detected or the corresponding calibration relationship will be reselected.

[0213] Through the above steps, the early spectra in the spectral sequence are not directly discarded, but are corrected and utilized under the constraints of surface interference attenuation boundaries and non-carbon state characteristics. This allows us to obtain the reference mechanical properties corresponding to the internal parent material, quantify the decarburization effect within the stress-sensitive depth, and obtain the predicted results of the local mechanical properties at the detection location.

[0214] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0215] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0216] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0217] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0218] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0219] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for predicting the mechanical properties of connecting rod materials based on a spectrometer, characterized in that, Includes the following steps: Continuous excitation at the same detection location in the critical stress area of ​​the connecting rod to obtain a spectral sequence arranged in excitation order; The relative carbon response and the excitation state response determined by non-carbon spectral features are extracted from the spectral sequence, and the surface disturbance attenuation boundary is determined based on the excitation state response. Based on the migration characteristics of the carbon relative response after the surface interference attenuation boundary, the carbon relative response before the correction boundary is corrected to form the carbon response depth distribution and determine the stable spectrum of the parent material. The baseline mechanical properties of the parent material are determined based on the stable spectrum of the parent material, and the decarburization effect within the stress-sensitive depth is determined based on the carbon response depth distribution. By correcting the baseline mechanical properties of the base material using the decarburization effect, the predicted results of the local mechanical properties at the test location are obtained.

2. The method for predicting the mechanical properties of connecting rod materials based on a spectrometer according to claim 1, characterized in that, The step of continuously exciting at the same detection position in the critical stress area of ​​the connecting rod to obtain a spectral sequence arranged in excitation order includes: Obtain the structural information and target load conditions of the target link, determine the stress distribution under the target load conditions based on the structural information, and determine the key stress areas of the link based on the stress distribution; The detection position is determined based on the key stress area of ​​the connecting rod, and the detection position mark and excitation positioning conditions are established based on the surface morphology of the detection position. The excitation positioning conditions are used to limit the center of the excitation spot, the excitation direction and the detection distance. According to the excitation positioning conditions and preset excitation parameters, multiple continuous excitations are performed at the detection position, and the original spectrum and excitation information corresponding to each excitation are collected in sequence. Based on the detection location markers, each original spectrum is associated with its corresponding excitation information and excitation order, and then arranged according to the excitation order to form a spectral sequence.

3. The method for predicting the mechanical properties of connecting rod materials based on a spectrometer according to claim 2, characterized in that, The extraction of the carbon relative response from the spectral sequence and the excitation state response determined by non-carbon spectral features, and the determination of the surface interference attenuation boundary based on the excitation state response, include: Spectral correction is performed on each original spectrum in the spectral sequence to obtain the corrected spectrum corresponding to each excitation order; Based on the correction spectrum, the carbon net spectral line response and the steel matrix reference response are extracted. Then, according to the change of the carbon net spectral line response relative to the steel matrix reference response, the carbon relative response corresponding to each excitation order is determined, forming a carbon relative response sequence. Based on the non-carbon spectral features that avoid the carbon characteristic bands in the corrected spectrum, the surface adhesion response, plasma state response and matrix excitation response corresponding to each excitation sequence are extracted to form a non-carbon state feature sequence. Based on the stable change characteristics of the latter part of the non-carbon state characteristic sequence, the stable state benchmark of the current detection position is determined. Based on the deviation of the non-carbon state characteristics corresponding to each excitation sequence from the steady state baseline, the excitation state response corresponding to each excitation sequence is determined, forming an excitation state response sequence. The excitation state stable interval is identified based on the excitation state response sequence, and the surface interference attenuation boundary is determined within the excitation state stable interval based on the response level and trend of the surface adhesion response.

4. The method for predicting the mechanical properties of connecting rod materials based on a spectrometer according to claim 3, characterized in that, The determination of the relative carbon response for each excitation order based on the change in the net carbon spectral response relative to the reference response of the steel matrix includes: In the correction spectrum corresponding to the i-th excitation, the carbon feature window and its local background window are determined. The background response of the carbon feature window is determined based on the local background window. The background response is subtracted from the spectral response of the carbon feature window to obtain the net carbon spectral response corresponding to the i-th excitation. In the same calibration spectrum, at least one iron reference spectral line is identified, along with its local background window. The steel matrix reference response corresponding to the i-th excitation is determined based on the net response of the iron feature window. The relative carbon response corresponding to the i-th excitation is determined based on the ratio between the net carbon spectral response and the reference response of the steel matrix. Arrange the carbon relative responses corresponding to each excitation in the order of excitation to form a carbon relative response sequence.

5. The method for predicting the mechanical properties of connecting rod materials based on a spectrometer according to claim 4, characterized in that, The step of determining the stable state benchmark at the current detection position based on the stable change characteristics of the latter part of the non-carbon state characteristic sequence includes: Select continuous effective spectra from the end of the spectral sequence forward to form the candidate spectral range of the latter segment, and divide the candidate spectral range of the latter segment into multiple candidate state windows; The center value, fluctuation degree, and variation trend of non-carbon state characteristics with excitation order were determined for each candidate state window. Based on the central value, fluctuation level and trend of change, stable candidate windows that meet the baseline window selection conditions are selected from multiple candidate state windows; Starting from the stable candidate window corresponding to the end of the spectral sequence, consecutive stable candidate windows are merged forward along the excitation order to obtain the state reference spectral band; Based on the baseline center value and baseline fluctuation value of each non-carbon state characteristic within the state baseline spectrum, a stable state baseline for the current detection location is established.

6. The method for predicting the mechanical properties of connecting rod materials based on a spectrometer according to claim 5, characterized in that, The carbon relative response based on the migration characteristic correction boundary before the boundary after the surface interference attenuation boundary forms the carbon response depth distribution and determines the stable spectral band of the parent material, including: The equivalent excitation depth is determined based on the cumulative ablation depth corresponding to each excitation sequence. Based on the surface disturbance attenuation boundary and the equivalent excitation depth, a reliable migration spectrum is determined from the carbon relative response after the boundary. Establish carbon response migration relationships based on reliable migration spectrum bands; Based on the carbon response migration relationship and the boundary forward directed state deviation component, the carbon response disturbance correction relationship is determined. The carbon relative response before the correction boundary is corrected according to the carbon response interference correction relationship, and combined with the carbon relative response after the boundary to obtain the corrected carbon response sequence. The corrected carbon response sequence is correlated with the equivalent excitation depth to form a carbon response depth distribution; The stable spectral band of the parent material was determined based on the depth distribution of the carbon response.

7. The method for predicting the mechanical properties of connecting rod materials based on a spectrometer according to claim 6, characterized in that, The determination of the carbon response disturbance correction relationship based on the carbon response migration relationship and the directed state deviation component before the boundary includes: Based on the surface interference attenuation boundary, a continuous excitation sequence adjacent to the surface interference attenuation boundary is selected from before the boundary to form a boundary adjacent segment. Based on the carbon response migration relationship, determine the migration reference response corresponding to each excitation sequence within the boundary adjacent segment; Based on the difference between the measured relative carbon response and the migration reference response corresponding to each excitation sequence within the boundary adjacent segment, the carbon response interference amount corresponding to each excitation sequence is determined. The carbon response interference quantity corresponding to each excitation sequence is correlated with the directional state deviation component corresponding to the same excitation to form an interference correction sample. Based on the interference correction samples, the correspondence between the directional state deviation component and the carbon response interference amount is determined, and the carbon response interference correction relationship at the current detection position is obtained.

8. The method for predicting the mechanical properties of connecting rod materials based on a spectrometer according to claim 7, characterized in that, The determination of the reference mechanical properties of the base material based on the stable spectrum of the base material includes: Based on the material state of the target link and the mechanical properties of the target to be predicted, determine the corresponding mechanical property prediction model; Extract the spectral features of the parent material that match the mechanical property prediction model from the stable spectral range of the parent material; The spectral characteristics of the base material are input into the mechanical property prediction model to obtain the predicted mechanical properties corresponding to the stable spectral band of the base material; Based on the predicted mechanical properties corresponding to the stable spectrum of the base material, the reference mechanical properties of the base material at the current testing location are determined.

9. The method for predicting the mechanical properties of connecting rod materials based on a spectrometer according to claim 8, characterized in that, The determination of the decarburization influence within the stress-sensitive depth based on the carbon response depth distribution includes: Based on the corrected carbon relative response in the stable spectrum of the parent material, determine the parent material carbon response benchmark at the current detection location; Based on the stress distribution along the surface normal at the detection location under the target load condition, determine the stress-sensitive depth at the current detection location; Based on the carbon response depth distribution and the stress-sensitive depth, the corrected carbon relative response corresponding to each material layer within the stress-sensitive depth range is determined. The degree of carbon response deficit for each material layer is determined based on the difference between the corrected relative carbon response of each material layer and the carbon response benchmark of the parent material. Based on the stress state of each material layer under the target load condition, determine the stress sensitivity weight of each material layer. The amount of decarbonization impact within the stress-sensitive depth is determined based on the degree of carbon response deficit and stress sensitivity weight of each material layer.

10. The method for predicting the mechanical properties of connecting rod materials based on a spectrometer according to claim 9, characterized in that, The method of correcting the baseline mechanical properties of the base material using the decarburization effect to obtain the predicted results of the local mechanical properties at the detection location includes: Based on the material state and mechanical properties of the target connecting rod, determine the corresponding decarburization correction relationship; Based on the decarburization correction relationship and the amount of decarburization influence, determine the mechanical property attenuation ratio corresponding to the detection location; The baseline mechanical properties of the base material are corrected based on the rate of mechanical property decay to obtain the predicted local mechanical properties at the test location. The predicted local mechanical properties are associated with the corresponding detection location, target load condition, and target mechanical property type to form the predicted local mechanical properties.

Images