Multi-hardness composite detection method and portable hardness tester
Patent Information
- Application Number
- CN202611078278.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-20
- Publication Date
- 2026-10-30
Smart Images
Figure CN123008530A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the technical field of material hardness testing equipment and testing methods, specifically relating to a multi-hardness composite testing method and a portable hardness tester. Background Technology
[0002] Technical solution of existing technology 1 The existing integrated hardness testing equipment in patent CN118655028A mostly uses conventional hardness pens with cylinder drive and manual operation of electric control valves to control the testing. It can only achieve single scale testing, without data fusion, error correction and intelligent recognition functions. The structure is redundant and the testing accuracy depends on the finished hardness tester. It has not formed a true intelligent testing logic, is prone to operational errors, and has limited practicality and innovation.
[0003] Disadvantages of existing technology 1 The core innovation is insufficient, and most of them are just splicing together conventional mechanical structures and finished hardness testers without original detection logic; the detection accuracy depends on the finished hardness tester itself, and there is no systematic error correction mechanism, which makes it easy to make operational errors; the detection function is limited, only supporting single scale detection, and cannot realize multi-scale synchronous detection and data cross-verification, resulting in limited reliability and practicality of the detection results, and the level of intelligence is low, requiring manual intervention throughout the operation.
[0004] Technical solution of existing technology 2 Patent CN112067427A uses a benchtop multi-hardness testing device to achieve simultaneous testing of multiple scales. However, it employs a complex mechanical structure and fixed testing logic, using multiple sets of indenters and loading mechanisms working simultaneously to complete the testing of different hardness scales.
[0005] Disadvantages of existing technology 2 The device has built-in basic temperature compensation logic, which only performs simple linear correction on the detection results of a single scale, without dynamic calibration and position error correction functions. In addition, the device is large and heavy, making it impossible to achieve portable on-site testing.
[0006] Therefore, we propose a multi-hardness composite testing method and a portable hardness tester. Summary of the Invention
[0007] The present invention aims to solve at least one of the technical problems existing in the prior art, and to provide a multi-hardness composite testing method and a portable hardness tester.
[0008] One aspect of the present invention provides a multi-hardness composite testing method, comprising the following steps: Step S1: Acquiring raw data of the material to be tested and preprocessing the raw data to obtain preprocessed data; Step S2: Comparing the preprocessed data with a material property database to identify the type of the material to be tested and the estimated hardness range, and matching the optimal combination of detection parameters from the material property database based on the type of the material to be tested and the estimated hardness range, and generating a composite detection logic instruction; Step S3: Performing composite detection on the material to be tested based on the composite detection logic instruction, and acquiring displacement data and image data; Step S4: Calculating the displacement data and the image data to obtain multiple initial hardness values on different scales; and performing abnormal data identification and removal processing and multi-source data fusion processing on each of the different initial hardness values on different scales to obtain a preliminary hardness value; Step S5: Performing multi-level error adaptive correction on the preliminary hardness value and each of the different initial hardness values on different scales to obtain a comprehensive hardness value and a corrected hardness value on each different scale.
[0009] Furthermore, in step S2, the material property database is established based on national standards GB / T 1172-1999, GB / T 230.1-2018 and the Handbook of Metallic Materials, including the hardness range, elastic-plastic parameters, surface properties and suitable testing scales of various commonly used metallic materials; custom new material information can also be added to the material property database by using machine automatic learning mode or by manually inputting standard values.
[0010] Further, in step S3, based on the composite detection logic instruction, composite detection is performed on the material to be tested, and displacement data and image data are acquired. The steps specifically include: according to the composite detection logic instruction, the pressure head switching mechanism controls the composite detection probe to switch to the corresponding pressure head combination; a preset graded test force is applied to the surface of the material to be tested using a force loading mechanism, and the pressure is maintained for a preset holding time; at the same time, a time-division multiplexing strategy is used to synchronously acquire displacement data and image data.
[0011] Furthermore, in the time-division multiplexing strategy, the acquisition timing is configured as follows: during the first 10% of the pressure holding phase, the displacement peak is acquired; during the middle 80% of the pressure holding phase, image acquisition and processing are performed; and during the last 10% of the pressure holding phase, secondary displacement verification is performed.
[0012] Further, in step S4, the specific steps of the abnormal data identification and removal process include: using the Mahalanobis distance algorithm to calculate the Mahalanobis distance between the initial hardness values of each different scale and the historical mean vector of multiple scales; if the Mahalanobis distance is greater than the Mahalanobis distance judgment threshold, the initial hardness values of the different scales are determined to be abnormal data, and the abnormal data is removed and a re-detection is triggered.
[0013] Furthermore, in step S4, the specific steps of the multi-source data fusion processing include: using a random forest regression algorithm to fuse the initial hardness values of each different scale after removing the abnormal data, so as to obtain the preliminary hardness value.
[0014] Furthermore, in step S5, the multi-level adaptive error correction includes: vibration and jitter error correction, detection position error correction, and dynamic calibration correction; wherein, the detection position error correction includes: surface curvature correction and contact angle correction.
[0015] Furthermore, the multi-hardness composite testing method also includes: outputting the final test result, which includes: the corrected hardness values of each of the different scales and the comprehensive hardness value, and displaying the multi-scale comparison curve and the material hardness uniformity analysis result. After the test is completed, the equipment automatically resets and waits for the next test to achieve closed-loop testing.
[0016] Furthermore, during the testing process, if any abnormal conditions occur, an adaptive adjustment logic is automatically triggered. These abnormal conditions include: fluctuations in loading force, poor contact of the indenter, and abnormal data acquisition. If the loading force fluctuates, the testing is paused, the force loading mechanism is controlled to perform a recalibration operation, and a preset graded test force is reapplied to the surface of the material under test. If the indenter has poor contact, the composite detection probe is controlled to perform a fine-tuning of its position and re-contact the surface of the material under test, and this adjustment continues until a stable and good contact state is achieved. If the data acquisition is abnormal, a data completion mechanism is triggered to complete the missing data. If two consecutive completion attempts fail, a fault message is output, and the corresponding abnormal information is recorded.
[0017] Another aspect of the present invention provides a portable hardness tester for implementing the multi-hardness composite testing method according to any one of the preceding claims. The portable hardness tester includes: a main housing, a composite testing probe, a force loading mechanism, a testing module, a main control unit, a display module, a power module, an operation module, and a storage and transmission module. The main housing is a portable handheld structure, integrating the display module, the operation module, and the power interface. The outer surface of the main housing is provided with anti-slip rubber texture, and the interior is provided with a shock-absorbing and vibration-damping mechanism. The composite testing probe is threadedly connected to the main housing and integrates multiple indenters. The force loading mechanism is used to output graded test forces to adapt to the loading requirements of different hardness scales. The testing module integrates a displacement detection unit, an image detection unit, and a position sensing unit for realizing multi-source data. Synchronous data acquisition; the main control unit integrates a material property database, a hardness conversion database, a composite detection logic algorithm, an error correction algorithm, and a data fusion algorithm. The main control unit is electrically connected to the force loading mechanism, the detection module, the display module, and the storage and transmission module. The display module displays detection mode information, parameter information, multi-scale hardness detection results, and error correction process information in real time. The power module has a built-in rechargeable lithium battery and is equipped with a power interface for external power supply to the portable hardness tester. The operation module integrates button components and a touch screen component, and is equipped with a voice broadcast unit for human-computer interaction. The storage and transmission module integrates a Bluetooth or WiFi wireless transmission module for storing, tracing, and transmitting detection data.
[0018] The beneficial effects of this invention are as follows: Significantly improved testing efficiency: The single-load multi-scale synchronous testing logic eliminates the need for multiple loadings and manual conversions, greatly reducing testing time and balancing efficiency and accuracy; the material-testing logic is automatically matched, eliminating the need for manual parameter setting and reducing the professional requirements for operators. Compared with traditional portable hardness testers with manual switching modes, the overall on-site testing efficiency is significantly improved. High detection accuracy and reliability: The design incorporates multi-level error adaptive correction logic, comprehensively considering the influence of multiple factors such as temperature, vibration, curvature, and contact angle. Combined with multi-source data fusion and anomaly removal algorithms, it is far superior to existing portable hardness testers. Dynamic calibration correction ensures long-term detection accuracy of the equipment with no drift error. Highly adaptable and widely applicable: It supports automatic identification and adaptation of 20 common metal materials and allows users to expand the database. It can detect flat, curved, and irregularly shaped components and is suitable for various scenarios such as machinery manufacturing, metallurgy, aerospace, pipeline inspection, and field engineering. One machine can meet diverse on-site inspection needs. It combines portability and practicality: lightweight handheld structure, built-in rechargeable lithium battery, can continuously test ≥600 times on a single charge, waterproof and dustproof, shockproof and shake-proof, suitable for on-site testing in areas with no power supply, limited space and harsh environment; supports wireless data transmission, storage and voice broadcast, easy to operate and no need for manual recording. Minimal workpiece damage and low equipment failure rate: Multiple scale inspections are completed in one loading cycle, avoiding repeated damage to the workpiece surface caused by multiple loading cycles; The force loading mechanism with closed-loop feedback control and the pressure head switching mechanism driven by a stepper motor are stable in operation and have a low failure rate; The adaptive adjustment logic for abnormal working conditions ensures the continuity of the inspection process and reduces manual intervention. Controllable manufacturing costs and easy to scale up: The hardware structure uses mature micro-sensors and control components, without complex mechanical structures, resulting in low manufacturing costs; the detection logic is implemented through the built-in algorithm of the main control unit, without the need for additional hardware costs, which facilitates large-scale production and market promotion. Attached Figure Description
[0019] Figure 1 This is a flowchart illustrating the steps of the multi-hardness composite testing method according to an embodiment of the present invention. Figure 2 A schematic diagram of the overall structure of the portable hardness tester according to an embodiment of the present invention; Figure 3 Hardware structure connection diagram of the portable hardness tester according to an embodiment of the present invention.
[0020] The components include: 1. Main unit housing; 2. Composite detection probe; 3. Display module; 4. Operation module; 5. Pressure head; 6. Pressure head switching mechanism; and 7. Power interface. Detailed Implementation
[0021] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0022] Please refer to the following: Figure 1The multi-hardness composite testing method provided by the specific embodiments of the present invention includes the following steps: Step S1, collecting raw data of the material to be tested and preprocessing the raw data to obtain preprocessed data; Step S2, comparing the preprocessed data with a material property database to identify the type of the material to be tested and the estimated hardness range, and matching the optimal combination of detection parameters from the material property database based on the type of the material to be tested and the estimated hardness range, and generating a composite detection logic instruction; Step S3, performing composite detection on the material to be tested based on the composite detection logic instruction, and collecting displacement data and image data; Step S4, calculating the displacement data and the image data to obtain multiple initial hardness values of different scales; and performing abnormal data identification and removal processing and multi-source data fusion processing on each of the different initial hardness values of different scales to obtain a preliminary hardness value; Step S5, performing multi-level error adaptive correction on the preliminary hardness value and each of the different initial hardness values of different scales to obtain a comprehensive hardness value and a corrected hardness value of each different scale.
[0023] Specifically, in step S1, the raw data of the material to be tested includes at least: the surface roughness of the material to be tested, the radius of curvature at the detection position, and the probe contact angle. Preprocessing the raw data of the material to be tested specifically involves using a wavelet denoising algorithm to reduce noise in the raw data, thereby filtering out high-frequency noise interference in the original signal and improving the accuracy and reliability of subsequent hardness testing data.
[0024] In this embodiment, in step S2, the material property database is established based on national standards GB / T 1172-1999, GB / T 230.1-2018 and the Handbook of Metal Materials, including the hardness range, elastic-plastic parameters, surface properties and suitable testing scales of various commonly used metal materials; custom new material information is added to the material property database by means of automatic machine learning mode or manual input of standard values.
[0025] The commonly used metal materials include at least 20 kinds of metal materials such as mild steel, aluminum alloy, copper, stainless steel and quenched parts.
[0026] Specifically, the preprocessed data is compared and analyzed with material parameters in the aforementioned material property database to automatically identify the type of material to be tested and the estimated hardness range. Based on this identification result, the optimal combination of detection parameters is automatically matched, and a unique composite detection logic instruction is generated accordingly, thereby achieving adaptive configuration of detection parameters without manual intervention.
[0027] In a preferred embodiment, the specific matching rules include: When the material to be tested is identified as a soft material (such as pure copper or aluminum alloy), the HB and HL composite test scale is matched, and the test force is set to 500N~1000N and the holding time is set to 2s~3s. When the material to be tested is identified as a medium-hard material (e.g., mild steel or stainless steel), the HRC, HB and HL composite test scale is matched, and the test force is set to 1000N~1500N and the holding time is set to 3s~5s. When the material to be tested is identified as a high-hardness material (such as quenched parts or mold steel), the HRC, HV and HL composite test scale is matched, and the test force is set to 1500N~3000N and the holding time is set to 3s~5s.
[0028] Furthermore, the material property database supports expansion through automatic machine learning or manual input of standard values to add custom new material information. When the material to be tested is detected as not included in the material property database, an unknown material prompt mechanism is triggered, guiding the user to select an approximate material type or manually set relevant parameters. Simultaneously, the system defaults to using a composite detection scale of HRC, HB, and HL and a moderate test force as the initial detection strategy to ensure the continuity of the detection process and avoid equipment downtime due to abnormal material identification.
[0029] When the detection location is a curved surface, the estimated indentation diameter is calculated based on the estimated hardness range and the currently set test force. And by calculating the ratio of the estimated indentation diameter to the radius of curvature of the detection location ( This is used to assess whether curvature correction is needed, thus providing a basis for prediction in subsequent error correction stages.
[0030] In this embodiment, in step S3, based on the composite detection logic instruction, composite detection is performed on the material to be tested, and displacement data and image data are acquired. The steps specifically include: according to the composite detection logic instruction, the pressure head switching mechanism controls the composite detection probe to switch to the corresponding pressure head combination; the force loading mechanism applies a preset graded test force to the surface of the material to be tested and holds the pressure for a preset holding time, while simultaneously acquiring displacement data and image data using a time-division multiplexing strategy.
[0031] The time-division multiplexing strategy was used to simultaneously acquire displacement data and image data, which enabled the synchronous acquisition of multi-dimensional data during a single loading process, effectively avoiding surface damage and testing errors caused by repeated loading of the material under test.
[0032] Furthermore, in the time-division multiplexing strategy, the acquisition timing is configured as follows: during the first 10% of the pressure holding phase, the displacement peak is acquired; during the middle 80% of the pressure holding phase, image acquisition and processing are performed; and during the last 10% of the pressure holding phase, secondary displacement verification is performed.
[0033] Specifically, according to the composite detection logic instructions, the indenter switching mechanism precisely controls the internal mechanical structure of the composite detection probe, seamlessly switching the indenter currently in the working position to an indenter combination that matches the type of material to be tested and the estimated hardness range. Subsequently, the force loading mechanism smoothly applies test force to the surface of the material to be tested according to a preset graded loading curve, and after reaching the target force value, maintains it for the preset holding time to eliminate the influence of material creep on the test results.
[0034] During the pressure holding phase, to maximize data acquisition efficiency while ensuring test accuracy, the system employs a time-division multiplexing (TDM) strategy to synchronously and sequentially acquire displacement data and image data. The specific acquisition timing configuration is as follows: During the first 10% of the pressure holding phase, the system triggers a high-frequency sampling mode to rapidly acquire displacement peak values. This phase is primarily used to capture the elastic deformation of the material at the moment of force application and the initial indentation depth, in order to obtain high-precision displacement reference data. During the middle 80% of the pressure holding phase, the system switches to vision processing mode to continuously acquire and process image data in real time. This phase utilizes a stable pressure holding environment to acquire indentation images, and processes these images using an edge recognition algorithm to calculate the actual indentation diameter or diagonal length, providing geometric data for calculating Brinell hardness (HB) and Vickers hardness (HV). This avoids the impact of electromagnetic interference from high-frequency displacement acquisition on the vision sensor. During the last 10% of the pressure holding phase, the system performs a secondary displacement verification and acquisition. By comparing the displacement data from the first 10% of the pressure holding phase with the displacement data from the last 10% of the pressure holding phase, the system can assess the creep recovery of the material or the minute rebound of the indenter during the pressure holding period, providing reliable data support for subsequent error correction and accurate calculation of hardness values.
[0035] The aforementioned acquisition timing ensures that critical transient data is not lost, while avoiding crosstalk between multi-channel signals, thus ensuring the synchronization and integrity of data acquisition. Simultaneously, it acquires real-time data on the depth of the indenter's penetration into the material under test, providing depth parameters for calculating Rockwell hardness (HRC) and Leeb hardness (HL).
[0036] In this embodiment, the specific steps of abnormal data identification and removal in step S4 include: using the Mahalanobis distance algorithm to calculate the Mahalanobis distance between the initial hardness values of each different scale and the historical mean vector of multiple scales; if the Mahalanobis distance is greater than the Mahalanobis distance judgment threshold, the initial hardness values of the different scales are determined to be abnormal data, and the abnormal data is removed and a re-detection is triggered.
[0037] In this embodiment, in step S4, the specific steps of the multi-source data fusion processing include: using a random forest regression algorithm to fuse the initial hardness values of each different scale after removing the abnormal data, so as to obtain the preliminary hardness value.
[0038] The displacement data and the image data are calculated in parallel using dual algorithms to obtain initial hardness values on different scales, including initial hardness values on different scales such as Brinell, Rockwell, Vickers, and Leeb.
[0039] Specifically, the calculation formula for the Mahalanobis distance algorithm is as follows: ; in, For sample points Mahalanobis distance to the overall distribution This is the average vector of multi-scale hardness values from the most recent 10 tests. This is the covariance matrix of hardness values across multiple scales.
[0040] like > ,in, If the initial hardness value is determined to be an abnormal value caused by factors such as indenter slippage, surface oil contamination, or impact, it will be automatically rejected and a re-detection will be triggered. The maximum number of re-detections is two. The MAHD distance determination threshold is... The value is dynamically adjusted based on the material type and the estimated hardness range, ranging from 2.0 to 4.0. In a preferred embodiment, 2.5 is used for high-hardness materials, 3.5 for low-hardness materials, and 3.0 for medium-hardness materials. In other embodiments, methods that can effectively identify and eliminate abnormal data can also be used for calculation, such as the 3σ criterion.
[0041] For the initial hardness values of each different scale after removing outliers, a random forest regression algorithm is used for data fusion to generate the preliminary hardness values. In a preferred embodiment, a random forest model is used to intelligently optimize the detection parameters. The core hyperparameters of this model are specifically configured as follows: the number of decision trees (…). The maximum depth of a single tree is set to 100. The minimum number of samples required for node splitting is set to 10 to 15 layers. The number of features considered when searching for the optimal segmentation is set to 2. The system employs an adaptive mode (Auto). In other embodiments, methods that enable efficient fusion of multi-source data can be used for computation, such as machine learning algorithms like gradient boosting trees and support vector machines.
[0042] Furthermore, the calculation weights of each hardness testing scale are automatically matched and assigned based on the material type to generate the preliminary hardness value. In a specific application scenario, the weight allocation rules include: When the material being tested is identified as a quenched part, the weights of each scale are assigned as follows: HRC scale 0.45, HV scale 0.30, HL scale 0.15, HB scale 0.10; When the material being tested is identified as mild steel, the weights assigned to each scale are: HB scale 0.40, HRC scale 0.30, HL scale 0.20, and HV scale 0.10.
[0043] The abnormal data identification and removal process and the multi-source data fusion process can improve data reliability.
[0044] In this embodiment, in step S5, the multi-level adaptive error correction includes: vibration and jitter error correction, detection position error correction, and dynamic calibration correction; wherein, the detection position error correction includes: surface curvature correction and contact angle correction.
[0045] Specifically, the multi-level error adaptive correction process comprehensively considers the cross-coupling effect of multiple factors such as ambient temperature, detection position (e.g., surface curvature), and external vibration, thereby effectively eliminating system errors and ensuring the final detection accuracy.
[0046] In a preferred embodiment, within a preset test condition range, i.e., an ambient temperature of -10℃ to 50℃, and when the surface of the material to be tested is a plane or a curved surface with a radius of curvature R ≥ 30mm, after the multi-level error adaptive correction processing, the comprehensive detection error can be strictly controlled within ±1.5%.
[0047] The specific steps for each correction stage are as follows: Vibration and jitter error correction: To address the impact of vibration during handheld operation, a weighted moving average method is used to smooth the indentation depth or actual indentation diameter data. The sampling window length... For 5 to 10 sampling points, a weighting strategy based on time distance is adopted, that is, sampling points closer to the current time are assigned a larger weight coefficient, and sampling points farther from the current time are assigned a smaller weight coefficient. The specific calculation formula is as follows: ; in, The data has been smoothed. The sampling window length is... For the first The initial hardness value for each sampling period, This represents the weighting coefficient corresponding to the sampling point.
[0048] Detection position error correction includes surface curvature correction and contact angle correction, wherein: Surface curvature correction: based on actual indentation diameter With respect to the radius of curvature at the detection position To assess whether curvature correction is needed, if When the value is 0.1, curvature correction is required. The specific calculation formula is as follows: ; ; in, For curvature compensation factor, This is the curvature correction factor. The power exponent. This is the actual indentation diameter.
[0049] In other embodiments, the curvature correction coefficient It can be adjusted appropriately according to different types of materials to be tested or types of indenters.
[0050] Contact angle correction: For the probe contact angle, the allowable contact angle threshold is set to 15°. If the angle is 15°, the current probe contact posture is determined to be inconsistent with the test specifications. The current testing process is immediately locked, and corresponding prompts are displayed on the human-machine interface. The voice broadcast module is also triggered to output a voice prompt "Please adjust the angle" to guide the operator to realign. This effectively avoids the problem of uneven contact stress distribution caused by probe tilt, thereby eliminating testing errors exceeding 5% caused by this, and ensuring the reliability and consistency of hardness test results.
[0051] Dynamic calibration correction: Based on a preset dynamic calibration cycle, the system automatically retrieves the hardness standard value of the built-in standard test block and automatically tests the standard test block. By comparing and analyzing the measured value with the hardness standard value, the system calculates and updates the error correction parameters in real time, thereby effectively compensating for performance drift caused by long-term use of the equipment and ensuring long-term testing accuracy. The error tolerance threshold is set as follows: single test relative error ≤ ±1.5%, long-term drift ≤ ±3%. When the error approaches or exceeds the error tolerance threshold, the aforementioned dynamic calibration correction is automatically triggered for self-calibration.
[0052] In this embodiment, the multi-hardness composite testing method further includes: outputting the final test result, which includes: the corrected hardness values of each of the different scales and the comprehensive hardness value, and displaying the multi-scale comparison curve and the material hardness uniformity analysis result. After the test is completed, the device automatically resets and waits for the next test to achieve closed-loop testing.
[0053] Specifically, after completing the multi-level error adaptive correction, the final test result is output. The final test result includes: the corrected hardness values for each different scale and the comprehensive hardness value. Simultaneously, multi-scale comparison curves and material hardness uniformity analysis results can be displayed. Test data is automatically stored in the main control unit's storage module and supports Bluetooth 5.0 / WiFi wireless transmission to terminal devices such as computers and mobile phones via the storage and transmission module, facilitating data traceability and test report generation. After the test is completed, the device automatically resets, awaiting the next test, achieving closed-loop testing.
[0054] In this embodiment, if an abnormal condition occurs during the testing process, an adaptive adjustment logic is automatically triggered. These abnormal conditions include: fluctuations in loading force, poor contact of the indenter, and abnormal data acquisition. If the loading force fluctuates, the testing is paused, the force loading mechanism is controlled to perform a recalibration operation, and a preset graded test force is reapplied to the surface of the material under test. If the indenter has poor contact, the composite detection probe is controlled to fine-tune its position and re-contact the surface of the material under test, continuously adjusting until a stable and good contact state is achieved. If data acquisition is abnormal, a data completion mechanism is triggered to complete missing data. If completion fails twice consecutively, a fault message is output, and the corresponding abnormal information is recorded.
[0055] Specifically, by adaptively adjusting to abnormal operating conditions, the continuity of the testing process can be ensured, and the anti-interference ability and reliability of the equipment in complex environments such as the field and high altitude can be improved.
[0056] Please refer to the following as well. Figure 2 and Figure 3Another aspect of the present invention provides a portable hardness tester for implementing the multi-hardness composite testing method described in any of the above claims. The portable hardness tester includes: a main housing 1, a composite testing probe 2, a force loading mechanism, a testing module, a main control unit, a display module 3, a power module, an operation module 4, and a storage and transmission module. The main housing 1 is a portable handheld structure, integrating the display module 3, the operation module 4, and a power interface 7. The outer surface of the main housing 1 is provided with anti-slip rubber texture, and the interior is provided with a shock-absorbing and vibration-damping mechanism. The composite testing probe 2 is threadedly connected to the main housing 1 and integrates multiple indenters. The force loading mechanism is used to output graded test forces to adapt to the loading requirements of different hardness scales. The testing module integrates a displacement detection unit, an image detection unit, and a position sensing unit to achieve multi-source... The system includes synchronous data acquisition; the main control unit integrates a material property database, a hardness conversion database, a composite detection logic algorithm, an error correction algorithm, and a data fusion algorithm; the main control unit is electrically connected to the force loading mechanism, the detection module, the display module 3, and the storage and transmission module; the display module 3 displays detection mode information, parameter information, multi-scale hardness detection results, and error correction process information in real time; the power module has a built-in rechargeable lithium battery and is equipped with the power interface 7 for external power supply to the portable hardness tester; the operation module 4 integrates button components and a touch screen component, and is equipped with a voice broadcast unit for human-computer interaction; the storage and transmission module integrates a Bluetooth or WiFi wireless transmission module for storing, tracing, and transmitting detection data.
[0057] Specifically, the portable hardness tester has an overall weight of ≤1500g, an IP54 waterproof and dustproof rating, a built-in rechargeable lithium battery, and can perform ≥600 tests on a single charge.
[0058] The main unit housing 1 is made of ABS engineering plastic, and the anti-vibration and anti-shake mechanism inside the main unit housing 1 can effectively reduce the impact of vibration caused by hand operation on the detection.
[0059] The composite detection probe 2 integrates multiple indenters 5 at its end. The indenters 5 include: a diamond conical indenter, a Φ10mm cemented carbide ball indenter, and a diamond pyramidal indenter. They are detachable and replaceable. It is also equipped with an indenter switching mechanism 6 driven by a stepper motor, which is automatically switched by the main control unit. The switching accuracy is 0.005mm and the switching time is ≤3s.
[0060] The force loading mechanism adopts a micro servo loading mechanism, supports closed-loop feedback control, and can output graded test forces of 50N~3000N. The force control resolution is ≤±0.5N, the force control stability is ≤±0.5%FS, and the holding time can be automatically adjusted between 0.5s and 10s to adapt to the loading requirements of different hardness scales.
[0061] The detection module is the core of multi-source data acquisition, realizing the synchronous acquisition of indentation, depth, environment, and location data, wherein: The displacement detection unit is equipped with a high-precision grating displacement sensor with a measurement range of 0~10mm and a resolution of ≤0.001mm; The image detection unit integrates a 5-megapixel miniature high-definition camera, an LED fill light, and an image processing module. The image detection unit is used to acquire images of the indentation formed by the indenter on the workpiece surface, transmit them to the image processing module, and calculate the indentation diameter or diagonal length using an edge recognition algorithm, providing data for calculating HB and HV hardness values. The repeatability of the indentation measurement is ≤ ±0.002 mm. The position sensing unit is equipped with a micro-sensor network, which can detect the radius of curvature and probe contact angle at the detection position. The contact angle detection accuracy is ≤±1°. It is used to collect the depth data of the indenter pressing into the material to be tested in real time, and provide data for the calculation of HRC hardness value and HL hardness value.
[0062] The main control unit is the control core of the portable hardness tester, used to realize functions such as logic control, data processing, and algorithm calculation.
[0063] The display module is equipped with a 3.5-inch high-definition LCD touch screen with a resolution of 800×480. It can display information such as detection mode, parameters, multi-scale hardness results, and error correction process in real time, and supports touch operation.
[0064] The power module has a built-in 2500mAh rechargeable lithium battery, supports 5V / 2A fast charging, and is equipped with a USB-C power interface 7 for external power supply. In addition, the power module also has a low-power standby function.
[0065] The operation module 4 integrates four physical buttons: power button, mode switch button, confirmation button and calibration button. It is operated in conjunction with a touch screen and also has a voice broadcast function, which can broadcast test results, operation prompts and fault information.
[0066] The storage and transmission module integrates Bluetooth and WiFi wireless transmission modules, and works with the built-in storage module of the main control unit to realize the storage and traceability of detection data, and can wirelessly transmit the detection data to terminal devices such as computers and mobile phones.
[0067] To make the objectives, technical solutions, and advantages of the present invention clearer, the multi-hardness composite testing method and portable hardness tester will be described in detail below with reference to specific embodiments.
[0068] In this specific embodiment, taking the testing of quenched parts with an estimated hardness of HRC 40-60 as an example, the portable hardness tester described in this invention is used, and the multi-hardness composite testing method described in this invention is followed: In this embodiment, the ambient temperature T=25℃, the detection position is a curved surface with a radius of curvature R=50mm, and the probe contact angle θ=8°. All parameters meet the technical requirements of this invention. The surface roughness of the quenched part was Ra=1.6μm, the ambient temperature was 25℃, the vibration amplitude was 0.002mm, the radius of curvature of the detection position was 50mm, and the probe contact angle was 8°. The raw data was transmitted to the main control unit after wavelet denoising. The main control unit compares the material properties database, identifies the tested material as a quenched part, automatically matches the detection scale combination HRC+HV+HL, sets the test force to 1500N, and the holding time to 3s. Simultaneously, based on the estimated hardness and test force, it calculates the estimated indentation diameter. The ratio of the estimated indentation diameter to the radius of curvature at the detection location. It is predicted that no curvature correction is needed; The composite detection probe automatically switches to a diamond conical indenter + a diamond pyramidal indenter. The force loading mechanism applies a test force of 1500N and holds the pressure for 3 seconds. The detection module uses time-division multiplexing to optimize the timing. In the first 10% of the time period, the displacement peak is acquired at high speed. In the middle 80% of the time period, the indentation image is acquired and processed. In the last 10% of the time period, a secondary displacement verification is performed, and the diagonal length of the indentation and the indentation depth are acquired simultaneously, with no signal crosstalk. The main control unit performed parallel calculations to obtain initial hardness values HRC=52.3, HV=501, and HL=786; using the Mahalanobis distance algorithm, the Mahalanobis distance threshold for high-hardness materials was determined as follows: Calculate the deviation No abnormal data was found; preliminary hardness values were obtained by fusing data using the random forest algorithm with weights assigned as HRC: 0.45, HV: 0.30, and HL: 0.25. (equivalent to HRC); Multi-level adaptive error correction is performed to obtain the initial hardness value after fusion. For the object of correction, the ambient temperature is the reference temperature of 25℃, so no temperature correction is required; the vibration amplitude is relatively small, and after correction using the weighted moving average method, No significant change; contact angle 8°≤15°, no angle correction required; actual measured indentation diameter , No curvature correction is required; this was the 10th test conducted by the equipment, and dynamic calibration was not triggered. The final corrected value yielded the comprehensive hardness value. Meanwhile, the correction results for each scale are retained: HRC=52.1, HV=500, HL=785, with an error ≤±1%; The display module simultaneously displays the overall hardness value. The system simultaneously displays the test results of each scale: HRC=52.1, HV=500, and HL=785, and generates a comprehensive hardness evaluation report. The test data is automatically stored and transmitted to the terminal device via Bluetooth through the storage and transmission module. The device automatically resets and completes the test. The entire test process takes 6 seconds.
[0069] This embodiment verifies that the multi-hardness composite detection logic of the present invention can achieve high-precision and high-efficiency detection of quenched parts, with accurate and reliable detection results, and the operation requires no manual intervention, fully meeting the needs of on-site detection. Similarly, the present invention can be adapted to different materials such as mild steel, aluminum alloy, and copper, and can achieve accurate detection in different detection positions such as planes and curved surfaces.
[0070] In summary, the embodiments disclosed herein have at least the following technical effects: Significantly improved testing efficiency: The single-load multi-scale synchronous testing logic eliminates the need for multiple loadings and manual conversions, greatly reducing testing time and balancing efficiency and accuracy; the material-testing logic is automatically matched, eliminating the need for manual parameter setting and reducing the professional requirements for operators. Compared with traditional portable hardness testers with manual switching modes, the overall on-site testing efficiency is significantly improved. High detection accuracy and reliability: The design incorporates multi-level error adaptive correction logic, comprehensively considering the influence of multiple factors such as temperature, vibration, curvature, and contact angle. Combined with multi-source data fusion and anomaly removal algorithms, it is far superior to existing portable hardness testers. Dynamic calibration correction ensures long-term detection accuracy of the equipment with no drift error. Highly adaptable and widely applicable: It supports automatic identification and adaptation of 20 common metal materials and allows users to expand the database. It can detect flat, curved, and irregularly shaped components and is suitable for various scenarios such as machinery manufacturing, metallurgy, aerospace, pipeline inspection, and field engineering. One machine can meet diverse on-site inspection needs. It combines portability and practicality: lightweight handheld structure, built-in rechargeable lithium battery, can continuously test ≥600 times on a single charge, waterproof and dustproof, shockproof and shake-proof, suitable for on-site testing in areas with no power supply, limited space and harsh environment; supports wireless data transmission, storage and voice broadcast, easy to operate and no need for manual recording. Minimal workpiece damage and low equipment failure rate: Multiple scale inspections are completed in one loading cycle, avoiding repeated damage to the workpiece surface caused by multiple loading cycles; The force loading mechanism with closed-loop feedback control and the pressure head switching mechanism driven by a stepper motor are stable in operation and have a low failure rate; The adaptive adjustment logic for abnormal working conditions ensures the continuity of the inspection process and reduces manual intervention. Controllable manufacturing costs and easy to scale up: The hardware structure uses mature micro-sensors and control components, without complex mechanical structures, resulting in low manufacturing costs; the detection logic is implemented through the built-in algorithm of the main control unit, without the need for additional hardware costs, which facilitates large-scale production and market promotion.
[0071] It is understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present invention, and the present invention is not limited thereto. For those skilled in the art, various modifications and improvements can be made without departing from the spirit and essence of the present invention, and these modifications and improvements are also considered to be within the scope of protection of the present invention.
Claims
1. A multi-hardness composite testing method, characterized in that, Includes the following steps: Step S1: Collect the raw data of the material to be tested, and preprocess the raw data to obtain the preprocessed data; Step S2: Compare the preprocessed data with the material property database to identify the type of material to be tested and the estimated hardness range. Based on the type of material to be tested and the estimated hardness range, match the optimal combination of detection parameters from the material property database and generate a composite detection logic instruction. Step S3: Based on the composite detection logic instructions, perform composite detection on the material to be tested, and acquire displacement data and image data; Step S4: Calculate the displacement data and the image data to obtain multiple initial hardness values on different scales; The initial hardness values of each of the different scales are then subjected to abnormal data identification and removal processing and multi-source data fusion processing to obtain preliminary hardness values; Step S5: Perform multi-level error adaptive correction on the preliminary hardness value and the initial hardness values of each different scale to obtain the comprehensive hardness value and the corrected hardness values of each different scale.
2. The multi-hardness composite testing method according to claim 1, characterized in that, In step S2, The material property database is established based on national standards GB / T 1172-1999, GB / T 230.1-2018 and the Handbook of Metallic Materials, including hardness ranges, elastic-plastic parameters, surface properties and suitable testing scales for a variety of commonly used metallic materials; Custom new material information can be added to the material property database using either machine-automated learning mode or by manually inputting standard values.
3. The multi-hardness composite testing method according to claim 1, characterized in that, In step S3, based on the composite detection logic instructions, composite detection is performed on the material to be tested, and displacement data and image data are acquired. The specific steps include: According to the composite detection logic instructions The pressure head switching mechanism controls the switching of the composite detection probe to the corresponding pressure head combination. A preset graded test force is applied to the surface of the material under test using a force loading mechanism and held at pressure for a preset holding time. Simultaneously, displacement data and image data are acquired using a time-division multiplexing strategy.
4. The multi-hardness composite testing method according to claim 3, characterized in that, In the time-division multiplexing strategy, the acquisition timing is configured as follows: during the first 10% of the pressure holding phase, the displacement peak is acquired; during the middle 80% of the pressure holding phase, image acquisition and processing are performed; and during the last 10% of the pressure holding phase, secondary displacement verification is performed.
5. The multi-hardness composite testing method according to claim 1, characterized in that, In step S4, the specific steps of the abnormal data identification and removal process include: The Mahalanobis distance algorithm is used to calculate the Mahalanobis distance between the initial hardness values of each different scale and the historical mean vector of multiple scales. If the Mahalanobis distance is greater than the Mahalanobis distance judgment threshold, the initial hardness values of the different scales are determined to be abnormal data, the abnormal data are removed, and a re-detection is triggered.
6. The multi-hardness composite testing method according to claim 5, characterized in that, In step S4, the specific steps of the multi-source data fusion processing include: The initial hardness values of each different scale after removing the outlier data are fused using a random forest regression algorithm to obtain the preliminary hardness value.
7. The multi-hardness composite testing method according to claim 1, characterized in that, In step S5, the multi-level adaptive error correction includes: Vibration and jitter error correction, detection position error correction, and dynamic calibration correction; wherein, the detection position error correction includes: surface curvature correction and contact angle correction.
8. The multi-hardness composite testing method according to claim 1, characterized in that, The multi-hardness composite testing method also includes: outputting the final test result, which includes the corrected hardness values of each of the different scales and the comprehensive hardness value, and displaying the multi-scale comparison curve and the material hardness uniformity analysis result. After the test is completed, the equipment automatically resets and waits for the next test to achieve closed-loop testing.
9. The multi-hardness composite testing method according to any one of claims 1 to 8, characterized in that, If any abnormal conditions occur during the testing process, the adaptive adjustment logic will be automatically triggered. The abnormal conditions include: fluctuation of loading force, poor contact of the pressure head, and abnormal data acquisition. If the loading force fluctuates, the detection is paused, the force loading mechanism is controlled to perform a recalibration operation, and a preset graded test force is reapplied to the surface of the material to be tested; If the pressure head has poor contact, the composite detection probe is controlled to make a fine adjustment in position and re-contact the surface of the material to be tested, and the adjustment is continued until a stable and good contact state is achieved. If data acquisition is abnormal, a data completion mechanism will be triggered to complete the missing data. If completion fails twice in a row, a fault message will be output and the corresponding abnormal information will be recorded.
10. A portable hardness tester, characterized in that, For implementing the multi-hardness composite testing method according to any one of claims 1 to 9, the portable hardness tester includes: a main housing, a composite testing probe, a force loading mechanism, a testing module, a main control unit, a display module, a power supply module, an operation module, and a storage and transmission module; The main unit housing is a portable handheld structure, integrating the display module, the operation module and the power interface. The outer surface of the main unit housing is provided with anti-slip rubber texture, and the interior is provided with an anti-vibration and anti-shake mechanism. The composite detection probe is threadedly connected to the main housing and integrates multiple pressure heads; The force loading mechanism is used to output graded test forces to adapt to the loading requirements of different hardness scales; The detection module integrates a displacement detection unit, an image detection unit, and a position sensing unit to achieve synchronous acquisition of multi-source data. The main control unit integrates a material property database, a hardness conversion database, a composite detection logic algorithm, an error correction algorithm, and a data fusion algorithm. The main control unit is electrically connected to the force loading mechanism, the detection module, the display module, and the storage and transmission module, respectively. The display module is used to display detection mode information, parameter information, multi-scale hardness test results, and error correction process information in real time. The power module has a built-in rechargeable lithium battery and is equipped with a power interface for connecting an external power source to power the portable hardness tester. The operation module integrates button components and touch screen components, and is equipped with a voice broadcast unit to realize human-computer interaction; The storage and transmission module integrates a Bluetooth or WiFi wireless transmission module to realize the storage, tracing and transmission of detection data.