A method for processing and characterizing high-throughput mechanical test data of micro / nano indentation and scratch.
By automatically matching instrument and indenter parameters through a calibration parameter library, identifying contact points and dividing test stages, and batch calculating the mechanical performance parameters of micro-nano indentations and scratches, the problems of low efficiency and inconsistent results in existing technologies are solved, and unified processing and accurate spatial distribution characterization of high-throughput testing are realized.
Patent Information
- Application Number
- CN202611420490.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-14
- Publication Date
- 2026-11-17
AI Technical Summary
Existing micro-nano mechanical testing data processing suffers from low efficiency, insufficient repeatability and comparability of results, susceptibility to subjective influence in manual processing, difficulty in matching instrument parameters, fragmented processing of indentation and scratch data, and a lack of a unified anomaly identification mechanism, thus failing to meet the needs of high-throughput testing.
The calibration parameter library enables automatic matching of instrument and indenter parameters, automatic identification of contact points and division of test stages, batch calculation of indentation and scratch mechanical performance parameters, and evaluation of test points and anomaly classification by combining multi-dimensional quality indicators, ultimately generating two-dimensional or three-dimensional spatial distribution results of mechanical performance.
It achieves consistency and repeatability of calculation results for the same batch of measurement points, reduces the impact of instrument deformation, environmental drift and channel crosstalk on the results, identifies and marks abnormal measurement points, adapts to different instruments and test modes, and outputs integrated processing results of multiple mechanical performance parameters.
Smart Images

Figure CN123111867A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of micro-nano mechanical testing technology, specifically a method for processing and characterizing high-throughput mechanical testing data of micro-nano indentations and scratches. Background Technology
[0002] Micro-nano indentation testing is a data testing technique that uses an indenter to press into the surface of a sample while simultaneously recording changes in normal load and indentation depth. By analyzing the load-depth curve, mechanical property parameters such as indentation hardness, contact stiffness, reduced elastic modulus, sample elastic modulus, indentation work, and creep rate can be obtained. Micro-nano scratch testing involves simultaneously applying a normal force to the indenter while allowing relative lateral movement between the indenter and the sample, and simultaneously recording changes in normal force, lateral force, scratch position, and depth. Based on this, parameters such as coefficient of friction, scratch depth, residual depth, elastic recovery rate, and critical load can be obtained.
[0003] With the development of thin films, coatings, micro-nano structures, gradient materials, and heterogeneous materials, it is no longer sufficient to reflect the local performance differences of materials by using only a small number of discrete measurement points to obtain average mechanical properties. In order to obtain the changes of material mechanical properties with spatial position, it is necessary to set up a regular array or irregular set of measurement points on the sample surface, continuously complete the indentation or scratch test of dozens, hundreds, or even more measurement points, and correlate the mechanical property parameters of each measurement point with spatial coordinates, thereby forming high-throughput mechanical property characterization results.
[0004] Current mainstream micro- and nano-mechanical testing data processing mostly adopts a single-point manual processing mode. Operators need to import data file by file, manually determine contact points, select fitting intervals, input instrument and indenter parameters, and export results. When the number of measurement points is large, the processing cycle increases linearly with the number of measurement points, which is difficult to adapt to the efficiency requirements of high-throughput testing. At the same time, in the manual processing mode, the selection of contact points, setting of fitting intervals, and judgment of abnormal data are easily affected by the subjective judgment of operators. The processing rules for the same batch of measurement points are difficult to maintain completely consistent, resulting in insufficient repeatability and comparability of results. In addition, the instrument calibration parameters and indenter parameters rely on manual search and input, which is prone to parameter mismatch and version confusion. Furthermore, indentation data and scratch data are usually processed by independent processes, and the data correction, performance calculation, and quality evaluation are scattered. There is no unified automatic identification mechanism for abnormal measurement points. Finally, there is no complete correlation and traceability link between the mechanical results and spatial coordinates, raw data, and calibration parameters, which cannot fully support the spatial distribution characterization of material properties and data traceability analysis under high-throughput testing. Summary of the Invention
[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a high-throughput mechanical testing data processing and characterization method for micro-nano indentation and scratch. This method achieves automatic matching and calling of instrument and indenter parameters through a calibration parameter library, completes the standardization and multi-dimensional error correction of multi-channel raw data, automatically identifies contact points and divides the testing stages, calculates mechanical performance parameters for both indentation and scratch tests in batches, and combines multi-dimensional quality indicators to achieve graded evaluation and anomaly classification of test points. Finally, it correlates mechanical performance parameters with spatial coordinates to generate two-dimensional or three-dimensional distribution results.
[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a method for processing and characterizing high-throughput mechanical testing data of micro / nano indentations and scratches, the method comprising the following specific steps:
[0007] S1: The computer establishes a test task and reads the corresponding set of calibration parameters from the calibration parameter library according to the instrument identification, pressure head identification and test mode;
[0008] S2: Read the original data of each measuring point according to the measuring point number, save a copy of the original data, perform dimensional conversion on the original sensor values output by the instrument to obtain corrected data, and associate each piece of corrected data with the measuring point number, spatial coordinates, test mode and sampling time;
[0009] S3: Perform error correction on the data after dimension conversion, obtain the corrected data, and save the corrected data separately from the original data, recording the calibration parameter version and algorithm parameter version;
[0010] S4: Identify the contact point between the indenter and the sample surface based on the statistical characteristics of the normal load before contact, and automatically divide the data after contact into the corresponding test stages based on the rate of change of load and / or displacement over time.
[0011] S5: For the measuring points in the indentation test mode, the contact stiffness is obtained by fitting the unloading curve to the corrected indentation test data, and the mechanical performance parameters of the indentation are calculated.
[0012] S6: For test points in scratch test mode, the computer calculates the scratch mechanical performance parameters based on the corrected scratch test data;
[0013] S7: Calculate the comprehensive quality score for each measuring point, and classify the measuring points into valid measuring points, measuring points to be reviewed, and invalid measuring points according to the comprehensive quality score and the mandatory invalidation conditions, and record the reasons for the abnormality;
[0014] S8: Associate the mechanical performance parameters, quality scores, and quality grades of each measuring point with spatial coordinates to generate two-dimensional or three-dimensional spatial distribution results of mechanical performance parameters and output a high-throughput test report.
[0015] Furthermore, in S1, the computer establishes a test task, assigns a unique task identifier to the test task, and receives sample information, instrument identifier, indenter identifier, test mode, measurement point arrangement, and measurement point spatial coordinates.
[0016] The computer reads the corresponding calibration parameter set from the calibration parameter library based on the instrument identification, pressure head identification, and test mode. The calibration parameter set includes: normal channel zero point. and normal channel sensitivity Displacement channel zero point and displacement channel sensitivity ;Horizontal channel zero point and lateral channel sensitivity Instrument flexibility Thermal drift rate ;Indenter area function coefficient Contact depth coefficient Indenter geometric correction factor Indenter elastic modulus Poisson's ratio with pressure head Crosstalk correction matrix between normal and transverse channels Calibration date, valid range, and version number for each calibration parameter.
[0017] Furthermore, in S2, the dimensional transformation is expressed as: ,in, For the first Normal load at each sampling point The displacement measured by the displacement sensor. For the first Lateral force at each sampling point , and These are the original values for the normal, displacement, and lateral channels, respectively. , and These are the zero points for the normal, displacement, and lateral channels, respectively; , and These are the sensitivities for the normal, displacement, and lateral channels, respectively.
[0018] Furthermore, in S2, the computer standardizes the units of time, load, displacement, force, and spatial coordinates, and standardizes the signs of load direction, displacement direction, and scratch direction. Each standardized data point is associated with the measurement point number, spatial coordinates, test mode, and sampling time.
[0019] Furthermore, in S3, the error correction includes: instrument compliance correction, thermal drift correction, and crosstalk correction between normal force and lateral force;
[0020] For instrument compliance correction: When the displacement measured by the displacement sensor includes instrument frame deformation, the computer calculates and corrects the indentation depth. ,in: This is to deduct the insertion depth after the instrument frame is deformed; The displacement measured by the displacement sensor; For instrument flexibility; For normal loads;
[0021] For thermal drift correction: The computer performs thermal drift correction: ,in, This is the depth after thermal drift correction; This refers to the thermal drift rate; This is the current sampling time; Reference time for thermal drift correction;
[0022] For crosstalk correction between normal and lateral forces: When crosstalk exists between the normal and lateral channels, the computer performs correction: ,in, The corrected normal force; The corrected lateral force; This is the second-order crosstalk correction matrix obtained through normal and lateral loading calibration.
[0023] Furthermore, in step S3, the computer saves the corrected data and the original data separately, and records the calibration parameter version, algorithm parameter version and processing time used in this processing, ensuring that any mechanical performance result can be traced back to the corresponding original curve and calibration parameters.
[0024] Furthermore, in S4, the computer calculates the average value of the normal load baseline from the data segment before the indenter contacts the specimen. and baseline standard deviation And calculate the contact determination threshold according to the following formula: ,in, The contact detection threshold; This represents the average pre-contact load. The standard deviation of the pre-contact load; This is the contact sensitivity coefficient;
[0025] The computer searches from front to back for the first sampling point that meets the following conditions and identifies it as a candidate contact point:
[0026] Normal load is greater than or equal to ;
[0027] This condition must be maintained continuously for no less than One sampling point;
[0028] The direction of the corresponding displacement change is consistent with the direction of the indentation, wherein, The number of consecutive judgment points.
[0029] Furthermore, in S4, for the indentation test data, the computer calculates the normal load based on the rate of change over time. and the rate of change of indentation depth over time The data after contact is divided into one or more of the following stages: loading stage, maximum load holding stage, unloading stage, and low load holding stage.
[0030] For scratch test data, the computer divides the data into one or more of the following stages: pre-scanning stage, scratch loading stage, and residual morphology scanning stage, based on the changes in scratch location, normal force, and lateral force.
[0031] When the original file already contains a stage identifier, the computer reads the stage identifier and performs a consistency check based on the load change trend; when the original file does not contain a stage identifier, the computer automatically generates a stage identifier according to the above rules.
[0032] Furthermore, in step S5, the computer selects fitting data from the upper interval of the unloading phase and fits the unloading curve: ,in, For the unloading phase normal load; This refers to the push depth during the unloading phase. These are the unloading fitting coefficients; The residual depth of complete unloading obtained by fitting; To unload the exponent;
[0033] The computer changes the start and end positions of the fitting within the preset fitting range, calculates the fitting residuals respectively, and determines the fitting interval that satisfies the effective range of parameters and has the smallest RMSE as the target fitting interval.
[0034] Calculate the contact stiffness at the maximum depth corresponding to the maximum load: ,in, For contact stiffness; This represents the maximum indentation depth under maximum load.
[0035] Furthermore, in S5, the indentation mechanical performance parameters include: contact depth, projected contact area, indentation hardness, reduced elastic modulus, sample elastic modulus, indentation work, and indentation creep rate.
[0036] Furthermore, in S6, the mechanical performance parameters of the scratch include: coefficient of friction, scratch depth, residual depth, elastic recovery rate, and critical load.
[0037] Furthermore, in S7, the overall quality score is: ,in, For the first The overall quality score of each measuring point; The number of quality indicators used in the evaluation; For the first The first measuring point One normalized quality indicator; For the first The weights of each quality indicator satisfy... and ;
[0038] For mechanical performance parameters that require spatial neighborhood evaluation Computer-calculated robust spatial anomaly indicators: ,in: For the first Spatial anomaly indicators at each measuring point; For the first The mechanical performance parameters to be evaluated at each measuring point; For the first The set of spatial neighborhood measurement points of a measurement point; median The median of the corresponding parameters at neighboring measurement points; MAD This represents the median absolute deviation of the corresponding parameters at neighboring measurement points; A positive number set to prevent the denominator from being zero;
[0039] The computer divides the measurement points into the following categories based on the overall quality score and the mandatory invalidation condition:
[0040] Valid measurement points: calibration parameters are complete, calculation process is successful, and quality score reaches the valid threshold;
[0041] Measurement points to be verified: The calculation process is completed, but the fitting error, spatial anomaly index, or drift index exceeds the verification threshold;
[0042] Invalid measurement points: incomplete raw data, signal saturation, unidentifiable contact points, missing key calibration parameters, calculation results that do not meet physical conditions, or quality scores below the invalid threshold.
[0043] Furthermore, in S8, the computer associates the mechanical performance parameters, quality score, quality grade, original data index, and calibration parameter version of each valid measuring point and the measuring point to be verified with spatial coordinates according to the measuring point number;
[0044] For a regular rectangular array of measurement points, the computer forms a parameter matrix according to row and column numbers; for an irregular set of measurement points, the computer forms a scattered data set based on the actual coordinates of the measurement points, and performs triangulation, nearest neighbor interpolation, or other spatial interpolation according to the user's selection. Invalid measurement points are displayed as empty or special symbols in the distribution results and are not included in the default statistical calculations.
[0045] The computer outputs the following for the specified mechanical performance parameters:
[0046] Two-dimensional spatial distribution map;
[0047] Three-dimensional spatial distribution map;
[0048] Maximum value, minimum value, mean, median, standard deviation, and number of valid measurement points;
[0049] The original curve, corrected curve, fitted curve, and calculation process for a single measurement point;
[0050] List of abnormal measurement points and reasons for abnormality;
[0051] Instrument identification, pressure head identification, calibration parameter version, algorithm parameter version, and generation time.
[0052] Compared with existing technologies, this method for processing and characterizing high-throughput mechanical test data of micro- and nano-indentation and scratch has the following advantages:
[0053] This invention eliminates subjective differences caused by manual operation by unifying calibration parameter matching, contact point identification, stage division, and performance calculation rules, ensuring the consistency and repeatability of calculation results for all measuring points in the same batch. Through systematic error correction of multi-channel data and precise indenter parameter calling, it can reduce the impact of instrument deformation, environmental drift, channel crosstalk, and indenter shape deviation on calculation results. By automatically completing the quality evaluation and anomaly classification of measuring points, it identifies and marks various abnormal measuring points, avoiding invalid data from interfering with the overall characterization results. Furthermore, it only requires updating the calibration parameter set to adapt to different instruments, indenters, and test modes. It can also achieve integrated processing of indentation and scratch test data, outputting multiple types of mechanical performance parameters in the same process. Finally, the mechanical performance results of each measuring point can be accurately correlated with spatial coordinates to generate two-dimensional or three-dimensional spatial distribution characterization results. At the same time, the original data, correction data, calibration parameter version, and calculation parameters are all associated and retained throughout the process, ensuring that every result is traceable and recalculated.
[0054] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description
[0055] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.
[0056] Figure 1 A flowchart of a high-throughput mechanical testing data processing and characterization method for micro / nano indentation and scratch;
[0057] Figure 2 This is a flowchart of step S3 of a high-throughput mechanical testing data processing and characterization method for micro / nano indentation and scratch. Detailed Implementation
[0058] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.
[0059] This invention proposes a high-throughput mechanical testing data processing and characterization method for micro-nano indentation and scratch. It achieves automatic matching and calling of instrument and indenter parameters through a calibration parameter library, completes the standardization and multi-dimensional error correction of multi-channel raw data, automatically identifies contact points and divides the test stages, calculates mechanical performance parameters of indentation and scratch tests in batches, and realizes graded evaluation and anomaly classification of test points by combining multi-dimensional quality indicators. Finally, it correlates mechanical performance parameters with spatial coordinates to generate two-dimensional or three-dimensional distribution results.
[0060] like Figure 1 As shown, the method in this embodiment specifically includes:
[0061] S1: The computer establishes a test task and reads the corresponding set of calibration parameters from the calibration parameter library according to the instrument identification, pressure head identification and test mode;
[0062] S2: Read the original data of each measuring point according to the measuring point number, save a copy of the original data, perform dimensional conversion on the original sensor values output by the instrument to obtain corrected data, and associate each piece of corrected data with the measuring point number, spatial coordinates, test mode and sampling time;
[0063] S3: Perform error correction on the data after dimension conversion, obtain the corrected data, and save the corrected data separately from the original data, recording the calibration parameter version and algorithm parameter version;
[0064] S4: Identify the contact point between the indenter and the sample surface based on the statistical characteristics of the normal load before contact, and automatically divide the data after contact into the corresponding test stages based on the rate of change of load and / or displacement over time.
[0065] S5: For the measuring points in the indentation test mode, the contact stiffness is obtained by fitting the unloading curve to the corrected indentation test data, and the mechanical performance parameters of the indentation are calculated.
[0066] S6: For test points in scratch test mode, the computer calculates the scratch mechanical performance parameters based on the corrected scratch test data;
[0067] S7: Calculate the comprehensive quality score for each measuring point, and classify the measuring points into valid measuring points, measuring points to be reviewed, and invalid measuring points according to the comprehensive quality score and the mandatory invalidation conditions, and record the reasons for the abnormality;
[0068] S8: Associate the mechanical performance parameters, quality scores, and quality grades of each measuring point with spatial coordinates to generate two-dimensional or three-dimensional spatial distribution results of mechanical performance parameters and output a high-throughput test report.
[0069] In step S1, the computer establishes a test task and assigns a unique task identifier to the test task. This task identifier is used to associate all measurement point data, calibration parameter versions, and calculation results in subsequent processing to ensure that there is no confusion between batches of test data. The computer receives the sample information, instrument identifier, indenter identifier, test mode, measurement point arrangement, and spatial coordinates of each measurement point input by the user. The sample information includes the material name, sample number, and sample size. The measurement point arrangement includes a regular rectangular array, a ring array, or a user-defined irregular scatter distribution. Different arrangement methods affect the generation strategy of subsequent spatial distribution results.
[0070] The computer retrieves and reads the corresponding calibration parameter set from the calibration parameter library based on the input instrument identifier, indenter identifier, and test mode. The calibration parameter library establishes a multi-level index structure according to instrument model, indenter model, and test mode to ensure the uniqueness and accuracy of parameter matching. The calibration parameter set includes: normal channel zero point, normal channel sensitivity, displacement channel zero point, displacement channel sensitivity, lateral channel zero point, lateral channel sensitivity, instrument compliance, thermal drift rate, indenter area function coefficient, contact depth coefficient, indenter geometric correction coefficient, indenter elastic modulus, indenter Poisson's ratio, crosstalk correction matrix between the normal and lateral channels, and the calibration date, valid range, and version number of each calibration parameter.
[0071] The aforementioned calibration parameters are pre-obtained through sensor calibration, standard sample testing, indenter certificates, indenter morphology measurement, or manual input, and are stored in the calibration parameter library. The computer is only responsible for reading, verifying, and retrieving the corresponding calibration parameters. When a calibration parameter that is crucial to the target result is missing, the computer does not mark the result as a valid result, but instead outputs a status indicating the absence of calibration parameters or a status indicating the calculated result is displayed, thus avoiding misinterpreting results generated by nominal or default parameters as official test results. Furthermore, after reading the calibration parameters, the computer also verifies the validity period of the parameters. If it finds that a calibration parameter has exceeded its validity period, it outputs a warning message indicating that the calibration parameter has expired.
[0072] In step S2, the computer reads the original data files of each measuring point sequentially or in parallel according to the measuring point number. The original data files can be comma-separated value files exported by the instrument, text files, or other structured data files, or data streams acquired in real time through the communication interface. The original data includes the measuring point number, measuring point spatial coordinates, test mode, sampling time, original value of the normal channel, original value of the displacement channel, instrument identification, and indenter identification. When performing scratch data processing, the input data also includes the original value of the transverse channel and the scratch position. For all measuring points on the same sample, the original data files of each measuring point have a unified data format and column definition, which facilitates batch reading and parsing by the computer. The computer saves a copy of the original data, which remains unchanged during subsequent correction and calculation processes to facilitate result traceability and recalculation.
[0073] If the instrument outputs sensor voltage or digital sampled value, the computer uses the normal channel zero point, normal channel sensitivity, displacement channel zero point, displacement channel sensitivity, lateral channel zero point, and lateral channel sensitivity to perform dimensional conversion on the raw data to obtain the normal load, displacement measured by the displacement sensor, and lateral force. If the instrument has already output load, displacement, and lateral force with physical units, the computer reads the physical quantities and unifies the units, without repeating the voltage conversion. The computer unifies the units of time, load, displacement, force, and spatial coordinates, and unifies the signs of the load direction, displacement direction, and scratch direction. The load direction is positive with the pressing direction, the displacement direction is positive with the pressing direction, and the scratch direction is positive with the predetermined scratch travel direction. Unifying the directions is a prerequisite for ensuring the comparability of calculation results at each measuring point.
[0074] The dimension transformation is expressed as follows: ,in, For the first Normal load at each sampling point The displacement measured by the displacement sensor. For the first Lateral force at each sampling point , and These are the original values for the normal, displacement, and lateral channels, respectively. , and These are the zero points for the normal, displacement, and lateral channels, respectively; , and These are the sensitivities for the normal, displacement, and lateral channels, respectively.
[0075] Each standardized data point is associated with the measurement point number, spatial coordinates, test mode, and sampling time, forming a structured data record with complete contextual information. The computer organizes the standardized data of all measurement points into a data matrix with measurement points as rows and sampling points as columns, and establishes a mapping table from measurement point numbers to data matrix row indices, providing an efficient data access interface for subsequent batch calculations.
[0076] In S3, such as Figure 2 As shown, in instrument compliance correction, the computer processes the displacement measured by the displacement sensor using instrument compliance to obtain the indentation depth after deducting the deformation of the instrument frame. The instrument frame undergoes elastic deformation under load, and this deformation is added to the displacement measured by the displacement sensor, making the measured value greater than the true indentation depth. The purpose of instrument compliance correction is to subtract this portion of the frame deformation from the measured displacement to obtain the true indentation depth of the indenter relative to the sample surface. This calculation process is expressed as follows: ,in: This is to deduct the insertion depth after the instrument frame is deformed; The displacement measured by the displacement sensor; For instrument flexibility; This is the normal load.
[0077] In thermal drift correction, the computer processes the indentation depth after deducting instrument frame deformation using the thermal drift rate to obtain the thermally drift-corrected depth. Changes in ambient temperature cause thermal expansion and contraction of the instrument structure, resulting in a slow drift of the displacement sensor's measurement reference. The purpose of thermal drift correction is to subtract the drift component caused by temperature changes from the measured depth. This calculation process can be expressed as: ,in, This is the depth after thermal drift correction; The thermal drift rate is usually obtained by fitting the displacement-time curve during the low load holding phase, or by obtaining it through a separate thermal drift measurement test before the test. This is the current sampling time; Reference time for thermal drift correction.
[0078] In the crosstalk correction of normal and lateral forces, the computer uses a crosstalk correction matrix to process the original values of the normal and lateral channels to obtain the corrected normal and lateral forces. In micro- and nano-scratch testing, there is a coupling effect between the normal force sensor and the lateral force sensor; that is, the reading of the normal channel is affected by the lateral load, and vice versa. The purpose of crosstalk correction is to eliminate this inter-channel interference through decoupling operations to obtain the true normal and lateral forces. This calculation process can be expressed as: ,in, The corrected normal force; The corrected lateral force; The second-order crosstalk correction matrix is obtained through normal and lateral loading calibration. This is used when it has been confirmed that there is no crosstalk between the two force channels that needs correction. A diagonal matrix containing only channel sensitivity can be used. The off-diagonal elements in the crosstalk correction matrix reflect the cross-coupling coefficient between force channels, and their values are obtained through multi-level loading calibration tests.
[0079] The computer saves the corrected data and the original data separately, and records the calibration parameter version, algorithm parameter version and processing time used in this process, ensuring that any mechanical performance result can be traced back to the corresponding original curve and calibration parameters.
[0080] In step S4, during contact point identification, the computer calculates the mean and standard deviation of the normal load baseline from the pre-contact data segment. It then uses the contact sensitivity coefficient, the mean baseline, and the standard deviation to calculate the contact determination threshold. Before the indenter contacts the sample surface, the normal load signal exhibits small random fluctuations around a certain mean. Its statistical characteristics can be used to distinguish noise signals from genuine contact signals. This calculation process can be expressed as follows:
[0081] ,in, The contact detection threshold; This represents the average pre-contact load. The standard deviation of the pre-contact load; This is the contact sensitivity coefficient. The value of the contact sensitivity coefficient determines the sensitivity of the contact determination. A value that is too small may cause noise to be misjudged as contact, while a value that is too large may cause a lag in contact point determination. The computer searches from front to back for the first point that satisfies the condition that the normal load is greater than or equal to... And this condition should be maintained continuously for no less than A sampling point was selected, and it was confirmed that the displacement change direction at that point was consistent with the pressing direction. This point was then identified as a candidate contact point. The number of consecutive judgment points.
[0082] In the segmented indentation stage, the computer analyzes the indentation test data based on the rate of change of the normal load over time. and the rate of change of indentation depth over time The data after contact is divided into a loading stage, a maximum load holding stage, an unloading stage, and a low load holding stage. In the loading stage, the normal load increases overall; in the maximum load holding stage, the normal load is maintained within the allowable deviation range of the maximum load; in the unloading stage, the normal load decreases overall; and in the low load holding stage, the thermal drift rate is calculated. The boundary points of each stage are automatically determined by detecting changes in the sign and amplitude of the load change rate, without the need for manual intervention.
[0083] In the scratch testing phase segmentation, the computer divides the scratch test data into a pre-scanning phase, a scratch loading phase, and a residual morphology scanning phase based on the changes in scratch position, normal force, and lateral force. The pre-scanning phase satisfies the following conditions: normal force is close to zero, lateral force is zero, and scratch position changes. The scratch loading phase satisfies the following conditions: normal force is constant, and lateral force changes with scratch position. The residual morphology scanning phase satisfies the following conditions: normal force is close to zero, and lateral force is zero.
[0084] The division of the above stages ensures that the subsequent calculations of friction coefficient, depth of penetration, and elastic recovery rate can be performed within the correct time window. When the original file already contains stage identifiers, the computer reads the stage identifiers and performs consistency verification through the load change trend. If the stage identifiers are found to be inconsistent with the load change trend, a warning message is output. When the original file does not contain stage identifiers, the computer automatically generates stage identifiers according to the above rules.
[0085] In step S5, the batch calculation of indentation mechanical properties includes unloading curve fitting and contact stiffness calculation, contact depth calculation, projected contact area calculation, indentation hardness calculation, reduced elastic modulus and sample elastic modulus calculation, indentation work and energy ratio calculation, and indentation creep rate calculation.
[0086] Specifically, in the unloading curve fitting and contact stiffness calculation, the computer selects fitting data from the upper interval of the unloading stage and uses a power function to fit the relationship between the normal load and the indentation depth during the unloading stage, obtaining the unloading fitting coefficient, the residual depth after complete unloading, and the unloading power exponent. This fitting function can be expressed as: ,in, For the unloading phase normal load; This refers to the push depth during the unloading phase. These are the unloading fitting coefficients; The residual depth of complete unloading obtained by fitting; To unload the exponentiation.
[0087] The computer changes the start and end positions of the fitting within a preset fitting range, calculates the root mean square error of the fitting residuals, and determines the fitting interval that satisfies the effective range of parameters and has the smallest root mean square error as the target fitting interval. The contact stiffness is calculated using the unloading fitting coefficient, the unloading power exponent, the maximum indentation depth under maximum load, and the residual depth after complete unloading. This calculation process can be expressed as follows:
[0088] ,in, For contact stiffness; This represents the maximum indentation depth under maximum load.
[0089] In the contact depth calculation, the computer uses the maximum load, the contact stiffness, and the contact depth coefficient to calculate the contact depth. This calculation process can be expressed as: ,in, For contact depth; Maximum load; For contact stiffness; The contact depth coefficient is read from the calibration parameter library.
[0090] In calculating the projected contact area, the computer substitutes the contact depth into the indenter area function, which is a weighted sum of the square term, the first-order term, the half-power term, the quarter-power term, and higher-order terms of the contact depth. This calculation process can be expressed as: ,in, This represents the projected area of the indenter in contact with the sample. The coefficients for the indenter area function are read from the calibration parameter library; This refers to the contact depth.
[0091] In indentation hardness calculation, the computer uses the maximum load and projected contact area to calculate the indentation hardness. This calculation process can be expressed as: In the formula, To determine the indentation hardness; Maximum load; This represents the projected contact area.
[0092] In the calculation of the reduced elastic modulus and the elastic modulus of the specimen, the computer uses the indenter geometric correction factor, contact stiffness, and projected contact area to calculate the reduced elastic modulus. This calculation process can be expressed as: In the formula, To convert the elastic modulus; This is the indenter geometry correction factor, read from the calibration parameter library; For contact stiffness; This represents the projected contact area.
[0093] The computer calculates the elastic modulus of the sample using the reduced elastic modulus, the sample Poisson's ratio, the indenter elastic modulus, and the indenter Poisson's ratio. The specific calculation process can be expressed as follows: The result after sorting is: In the formula, The elastic modulus of the sample; The Poisson's ratio of the sample is input by the user or read from the material parameter library; The indenter's elastic modulus is read from the calibration parameter library; The pressure head Poisson's ratio is read from the calibration parameter library.
[0094] In the calculation of the forced work and energy ratio, the computer integrates the loading curve to obtain the total forced work: The unloading work is obtained by integrating the unloading curve: Calculate the proportions of plastic work and unloading work to the total pressure work using the total pressure work and unloading work: , ,in, To compress the total work; For unloading function; Plastic work; The proportion of unloading power to total input power; For the load-depth relationship during the loading phase; The load-depth relationship during the unloading phase; To completely unload the residual depth, for discrete data, the computer uses trapezoidal integrals or other numerical integration methods to calculate the area of the curve mentioned above.
[0095] In the calculation of indentation creep rate, during the maximum load holding phase, the computer records the thermal drift correction depth at the moment the constant load is reached as the first depth, and the thermal drift correction depth at the moment the constant load holding ends as the second depth. The indentation creep rate is calculated using the first and second depths. This calculation process is expressed as follows: In the formula, The creep rate is the indentation rate. To achieve constant load time Thermal drift correction depth; Hold the end time for constant load Thermal drift correction depth.
[0096] In step S6, during the friction coefficient calculation, the computer uses the corrected lateral force and the corrected normal force to calculate the friction coefficient at each sampling location. Furthermore, the friction coefficient is calculated only when the corrected normal force is greater than the preset minimum effective normal force, thus avoiding numerical anomalies when the normal force is close to zero. This calculation process can be expressed as: ,in, For the first The friction coefficient at each sampling location; The corrected lateral force; This is the corrected normal force.
[0097] In the calculation of scratch depth, residual depth, and elastic recovery rate, the computer uses the depth during the scratch loading stage, the surface baseline height during the pre-scanning stage, and the height during the residual morphology scanning stage to calculate the scratch depth, residual depth, and elastic recovery rate, respectively. This calculation process can be expressed as: , , ,in, To determine the depth of the area to be enclosed; Residual depth; The elastic recovery rate of the scratch; The depth for loading the scratch stage; This refers to the surface baseline height at the same scratch location during the pre-scanning phase. The height at the same scratch location during the residual morphology scanning stage.
[0098] In determining the critical load, the computer performs baseline normalization on the changes in friction coefficient, depth of penetration, lateral force, and optional acoustic emission signal. It then calculates a comprehensive mutation index using each normalized value and its corresponding non-negative weights, where the sum of the weights involved in the calculation is 1. This calculation process can be expressed as: In the formula, This represents the comprehensive mutation index at the j-th sampling location; This is the standardized value of the change in the coefficient of friction; The standardized value for the depth variation is included. This is the standardized value of the change in lateral force. This is the standardized value of the change in the acoustic emission signal; , , and The weights are non-negative, and the sum of the weights involved in the calculation is 1.
[0099] When acoustic emission data is not available, let =0, and normalize the remaining weights. The computer determines the first sampling position that satisfies the comprehensive mutation index being greater than or equal to the mutation judgment threshold and continuously maintaining no less than the preset number of continuous judgment points as the candidate failure position. The computer reads the corrected normal force corresponding to the candidate failure position to obtain the critical load: ,in, This is the critical load; This is the corrected normal force at the candidate failure location. When in-situ or post-test microscopic images are available, the candidate failure location can also be verified and confirmed based on the location where cracking, peeling, or interface detachment first appears in the image.
[0100] In step S7, the computer calculates data integrity index, curve fitting index, calibration parameter validity index, drift index, and spatial neighborhood consistency index for each measurement point. Each index is normalized and denoted as follows: The value ranges from 0 to 1. The computer calculates the comprehensive quality score using each normalized quality indicator and its corresponding weight. The weights of each quality indicator are non-negative and the sum of the weights is 1. The calculation process is expressed as follows: ,in, For the first The overall quality score of each measuring point; The number of quality indicators used in the evaluation; For the first The first measuring point One normalized quality indicator; For the first The weights of each quality indicator satisfy... and .
[0101] The computer calculates spatial anomaly indices using the mechanical performance parameters to be evaluated, the median and absolute deviation of the corresponding parameters at other measuring points within the spatial neighborhood of the measuring point. This calculation process is expressed as follows: ,in: For the first Spatial anomaly indicators at each measuring point; For the first The mechanical performance parameters to be evaluated at each measuring point; For the first The set of spatial neighborhood measurement points of a measurement point; median The median of the corresponding parameters at neighboring measurement points; MAD This represents the median absolute deviation of the corresponding parameters at neighboring measurement points; A positive number set to prevent the denominator from being zero.
[0102] The computer categorizes measurement points into valid, verification-needed, and invalid points based on a comprehensive quality score and mandatory invalidation conditions. Valid measurement points are those with complete calibration parameters, successful calculation, and a quality score reaching the valid threshold. Verification-needed measurement points are those where the calculation was completed but the fitting error, spatial anomaly index, or drift index exceeds the verification threshold. Invalid measurement points are those with incomplete original data, signal saturation, unidentifiable contact points, missing key calibration parameters, calculation results that do not meet physical conditions, or a quality score below the invalidation threshold. The computer retains the spatial location of abnormal measurement points, does not directly overwrite the original results of abnormal measurement points with results from adjacent measurement points, and records the reasons for the anomalies.
[0103] In step S8, the computer associates the mechanical performance parameters, quality score, quality grade, original data index, and calibration parameter version of each valid and unverified measuring point with spatial coordinates according to the measuring point number. For a regular rectangular measuring point array, the computer forms a parameter matrix according to the row and column numbers; for an irregular set of measuring points, the computer forms a scattered data set based on the actual coordinates of the measuring points and performs triangulation, nearest neighbor interpolation, or other spatial interpolation according to the user's selection. Invalid measuring points are displayed as empty or special symbols in the distribution results and do not participate in the default statistical calculations.
[0104] The computer outputs two-dimensional and three-dimensional spatial distribution maps for specified mechanical performance parameters, along with the maximum, minimum, average, median, standard deviation, number of valid measurement points, original and corrected curves and fitted curves for each measurement point, calculation process, list of abnormal measurement points and their causes, instrument identification, indenter identification, calibration parameter version, algorithm parameter version, and generation time. This yields high-throughput mechanical performance characterization results.
[0105] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.
Claims
1. A method for processing and characterizing high-throughput mechanical testing data of micro / nano indentations and scratches, characterized in that, The method includes: S1: The computer establishes a test task and reads the corresponding set of calibration parameters from the calibration parameter library according to the instrument identification, pressure head identification and test mode; S2: Read the original data of each measuring point according to the measuring point number, save a copy of the original data, perform dimension conversion on the original sensor values output by the instrument, obtain the dimension-converted data, and associate each dimension-converted data with the measuring point number, spatial coordinates, test mode and sampling time; S3: Perform error correction on the data after dimension conversion, obtain the corrected data, and save the corrected data separately from the original data, recording the calibration parameter version and algorithm parameter version; S4: Identify the contact point between the indenter and the sample surface based on the statistical characteristics of the normal load before contact, and automatically divide the data after contact into the corresponding test stages based on the rate of change of load and / or displacement over time. S5: For the measuring points in the indentation test mode, the contact stiffness is obtained by fitting the unloading curve to the corrected indentation test data, and the mechanical performance parameters of the indentation are calculated. S6: For test points in scratch test mode, the computer calculates the scratch mechanical performance parameters based on the corrected scratch test data; S7: Calculate the comprehensive quality score for each measuring point, and classify the measuring points into valid measuring points, measuring points to be reviewed, and invalid measuring points according to the comprehensive quality score and the mandatory invalidation conditions, and record the reasons for the abnormality; S8: Associate the mechanical performance parameters, quality scores, and quality grades of each measuring point with spatial coordinates to generate two-dimensional or three-dimensional spatial distribution results of mechanical performance parameters and output a high-throughput test report.
2. The method for processing and characterizing high-throughput mechanical testing data of micro / nano indentation and scratch according to claim 1, characterized in that, In S1, the set of calibration parameters includes: Normal channel zero point and normal channel sensitivity; Displacement channel zero point and displacement channel sensitivity; Lateral channel zero point and lateral channel sensitivity; Instrument flexibility; Thermal drift rate; Indenter area function coefficient; Contact depth coefficient; Indenter geometric correction factor; Indenter elastic modulus and indenter Poisson's ratio; Crosstalk correction matrix between normal channel and transverse channel; The calibration date, valid range, and version number of each calibration parameter.
3. The method for processing and characterizing high-throughput mechanical testing data of micro / nano indentations and scratches according to claim 1, characterized in that, In S2, the dimensional transformation is expressed as follows: ,in, For the first Normal load at each sampling point The displacement measured by the displacement sensor. For the first Lateral force at each sampling point , and These are the original values for the normal, displacement, and lateral channels, respectively. , and These are the zero points for the normal, displacement, and lateral channels, respectively; , and These are the sensitivities for the normal, displacement, and lateral channels, respectively.
4. The method for processing and characterizing high-throughput mechanical testing data of micro / nano indentations and scratches according to claim 1, characterized in that, In S3, the error correction includes: instrument compliance correction, thermal drift correction, and crosstalk between normal force and lateral force correction. For instrument compliance correction: When the displacement measured by the displacement sensor includes instrument frame deformation, the computer calculates and corrects the indentation depth. ,in: This is to deduct the insertion depth after the instrument frame is deformed; The displacement measured by the displacement sensor; For instrument flexibility; For normal loads; For thermal drift correction: The computer performs thermal drift correction: ,in, This is the depth after thermal drift correction; This refers to the thermal drift rate; This is the current sampling time; Reference time for thermal drift correction; For crosstalk correction between normal and lateral forces: When crosstalk exists between the normal and lateral channels, the computer performs correction: ,in, The corrected normal force; The corrected lateral force; The second-order crosstalk correction matrix is obtained through normal and lateral loading calibration. This is used when it has been confirmed that there is no crosstalk between the two force channels that needs correction. A diagonal matrix containing only channel sensitivity can be used.
5. The method for processing and characterizing high-throughput mechanical testing data of micro / nano indentation and scratch according to claim 1, characterized in that, In step S4, the computer calculates the mean value of the normal load baseline from the data segment before the indenter contacts the specimen. and baseline standard deviation And calculate the contact determination threshold according to the following formula: ,in, The contact detection threshold; This represents the average pre-contact load. The standard deviation of the pre-contact load; This is the contact sensitivity coefficient; The computer searches from front to back for the first sampling point that meets the following conditions and identifies it as a candidate contact point: Normal load is greater than or equal to ; This condition must be maintained continuously for no less than One sampling point; The direction of the corresponding displacement change is consistent with the direction of the indentation, wherein, The number of consecutive judgment points.
6. The method for processing and characterizing high-throughput mechanical testing data of micro / nano indentations and scratches according to claim 1, characterized in that, In step S4, for the indentation test data, the computer calculates the normal load based on the rate of change over time. and the rate of change of indentation depth over time The data after contact is divided into one or more of the following stages: loading stage, maximum load holding stage, unloading stage, and low load holding stage. For scratch test data, the computer divides the data into one or more of the following stages: pre-scanning stage, scratch loading stage, and residual morphology scanning stage, based on the changes in scratch location, normal force, and lateral force. When the original file already contains a stage identifier, the computer reads the stage identifier and performs a consistency check based on the load change trend; when the original file does not contain a stage identifier, the computer automatically generates a stage identifier according to the above rules.
7. The method for processing and characterizing high-throughput mechanical testing data of micro / nano indentations and scratches according to claim 1, characterized in that, In step S5, the computer selects fitting data from the upper interval of the unloading phase and fits the unloading curve: ,in, For the unloading phase normal load; This refers to the push depth during the unloading phase. These are the unloading fitting coefficients; The residual depth of complete unloading obtained by fitting; To unload the exponent; The computer changes the start and end positions of the fitting within the preset fitting range, calculates the fitting residuals respectively, and determines the fitting interval that satisfies the effective range of parameters and has the smallest RMSE as the target fitting interval. Calculate the contact stiffness at the maximum depth corresponding to the maximum load: ,in, For contact stiffness; This represents the maximum indentation depth under maximum load.
8. The method for processing and characterizing high-throughput mechanical testing data of micro / nano indentations and scratches according to claim 1, characterized in that, In S5, the indentation mechanical property parameters include: contact depth, projected contact area, indentation hardness, reduced elastic modulus, sample elastic modulus, indentation work, and indentation creep rate.
9. The method for processing and characterizing high-throughput mechanical testing data of micro / nano indentations and scratches according to claim 1, characterized in that, In S6, the mechanical performance parameters of the scratch include: friction coefficient, scratch depth, residual depth, elastic recovery rate, and critical load.
10. The method for processing and characterizing high-throughput mechanical testing data of micro / nano indentations and scratches according to claim 1, characterized in that, In S7, the overall quality score is as follows: ,in, For the first The overall quality score of each measuring point; The number of quality indicators used in the evaluation; For the first The first measuring point One normalized quality indicator; For the first The weights of each quality indicator satisfy... and ; The computer divides the measurement points into the following categories based on the overall quality score and the mandatory invalidation condition: Valid measurement points: calibration parameters are complete, calculation process is successful, and quality score reaches the valid threshold; Measurement points to be verified: The calculation process is completed, but the fitting error, spatial anomaly index, or drift index exceeds the verification threshold; Invalid measurement points: incomplete raw data, signal saturation, unidentifiable contact points, missing key calibration parameters, calculation results that do not meet physical conditions, or quality scores below the invalid threshold.
