Intelligent detection method and system for impact performance of forging blank

Abstract

The invention relates to the field of data processing, in particular to an intelligent detection method and system for the impact performance of a forging blank, and the method comprises the steps: firstly, carrying out micro impact and thrust experiments on a sample forging, obtaining a basic data set containing parameters such as geometric dimensions, a force displacement curve and impact energy, and carrying out grouping storage according to material types and shapes; secondly, constructing an inflection point size correlation model and a target fracture energy and impact energy correlation model based on each group of data, and respectively representing correlation relations between inflection point coordinates and geometric dimensions and between fracture energy and impact energy; the model and the identity label of the forge piece group are associated and stored in a database. And when a detection request is received, a micro-impact experiment is executed on a requested forge piece, and impact performance parameters are predicted through the correlation model in the database. According to the method, efficient and accurate detection of the impact performance of the forge piece is achieved through combination of experimental data and an intelligent model, and the method is suitable for forge piece blanks of different material types and shapes.

Description

Technical Field

[0001] The present invention relates to the field of material mechanical property detection, and in particular to an intelligent detection method and system for the impact performance of a forging blank. Background Art

[0002] Currently, mainstream approaches to impact testing forgings rely on analyzing image data from the forging's surface and cross-section after fracture, using computer vision algorithms to extract characteristic parameters such as crack length, propagation direction, and fracture morphology, and then assess impact performance (e.g., the technology disclosed in patent document CN117451539A). These approaches have significant limitations: First, destructive testing of fractured forgings fails to preserve forging integrity, making them unsuitable for testing high-value or irreplaceable forgings. Second, the image analysis process involves manual annotation and feature extraction algorithm design, resulting in low data processing efficiency and poor model generalization. With the increasing application of computer technology in materials testing, the industry urgently needs a forging impact testing solution based on non-destructive testing data, integrated with computer programs and intelligent algorithms. This approach, through automated processing and modeling of digital data from micro-impact tests (such as force-displacement curves and crack propagation area measurements), enables rapid impact performance prediction, thus avoiding the destructive flaws and data processing inefficiencies of traditional methods. Summary of the Invention

[0003] In view of the above technical problems, the technical solution adopted by the present invention is: According to a first aspect of the present invention, there is provided an intelligent detection method for the impact performance of a forging blank, the method comprising the following steps: S100, performing a micro-impact test and a fracture test on each forging in the sample forgings, obtaining micro-impact record data and fracture record data of the forging, and constructing a basic data set based on the record data of all forgings; wherein the sample forgings include a group of forgings having the same material type but different shape types, and each forging group includes forgings having different geometric dimension data; the basic data set includes multiple record sets, each record set corresponds to a forging group, and includes an identity identifier of the forging group, micro-impact record data, and fracture record data, the micro-impact record data includes geometric dimension data, a force-displacement curve, a crack extension area measurement value, and a temperature increase value, the fracture record data includes geometric dimension data and an impact work measurement value, and the force-displacement curve is used to characterize the numerical correspondence between the impact force and the pit depth.

[0004] S200, based on the micro-impact recording data and the impact recording data corresponding to each forging group, obtain the inflection point size association model and the target fracture energy impact work association model corresponding to the forging group; wherein the inflection point size association model is used to characterize the association relationship between the inflection point coordinates and the geometric dimension data, and the target fracture energy impact work association model is used to characterize the association relationship between the fracture energy and the impact work.

[0005] S300: Associating the identity identifier, inflection point size correlation model, and target fracture energy impact work correlation model of each forging group, and storing them in the current database.

[0006] S400 , when a forging performance test request is received, a micro-impact test is performed on the requested forging and the impact energy of the requested forging is predicted based on an inflection point size correlation model and a target fracture energy impact energy correlation model in a current database.

[0007] According to a second aspect of the present invention, there is provided an intelligent detection system for the impact properties of a forging blank, the system comprising: A sample processing module is configured to perform micro-impact experiments and fracture experiments on sample forgings, obtain micro-impact record data and fracture record data, and construct a basic data set; wherein the sample forgings include a group of forgings with the same material type but different shape types, and each forging group includes forgings with different geometric dimension data; the basic data set includes multiple record sets, each record set corresponds to a forging group, and includes an identity identifier of the forging group, micro-impact record data and fracture record data, the micro-impact record data includes geometric dimension data, a force-displacement curve, a crack extension area measurement value and a temperature increase value, the fracture record data includes geometric dimension data and an impact work measurement value, and the force-displacement curve is used to characterize the numerical correspondence between the impact force and the pit depth.

[0008] The model building module is configured to obtain the inflection point size association model and the target fracture energy impact work association model corresponding to each forging group based on the micro-impact record data and the impact record data corresponding to the forging group; wherein the inflection point size association model is used to characterize the association relationship between the inflection point coordinates and the geometric dimension data, and the target fracture energy impact work association model is used to characterize the association relationship between the fracture energy and the impact work.

[0009] The data storage module is configured to associate the identity identification, inflection point size association model and target fracture energy impact work association model of each forging group and store them in the current database.

[0010] The prediction module is configured to, when receiving a forging performance test request, perform a micro-impact test on the requested forging and predict the impact energy of the requested forging based on the inflection point size correlation model and the target fracture energy impact energy correlation model in the current database.

[0011] The present invention has at least the following beneficial effects: The intelligent detection method for the impact performance of forging blanks provided by the embodiment of the present invention integrates the multi-dimensional physical performance data obtained from micro-impact experiments and fracture experiments, and breaks through the limitations of traditional destructive detection by constructing an inflection point size correlation model and a target fracture energy impact work correlation model. The method collects dynamic parameters such as force-displacement curves and crack extension areas in real time through non-destructive micro-impact experiments, combines key indicators such as impact work obtained from the fracture experiment, and uses intelligent algorithms to establish a quantitative correlation between material performance parameters. Compared with traditional detection technology, this method realizes a technological innovation from "post-destructive analysis" to "real-time non-destructive prediction", which not only significantly improves detection efficiency and reduces experimental costs, but also provides key technical support for intelligent quality control of forging production lines through automated data processing and model prediction.

[0012] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0014] Figure 1 The present invention provides a flowchart of an intelligent detection method for the impact performance of forging blanks. DETAILED DESCRIPTION

[0015] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts shall fall within the scope of protection of the present invention.

[0016] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention pertains. The terms used herein in the specification of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0017] It should be noted that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe the steps as sequential processes, many of the steps can be performed in parallel, concurrently, or simultaneously. In addition, the order of the steps can be rearranged. A process can be terminated when its operation is completed, but can also have additional steps not included in the accompanying drawings. A process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0018] The embodiment of the present invention provides an intelligent detection method for the impact performance of forging blanks, such as Figure 1 As shown, the method includes the following steps: S100 , performing a micro-impact test and a fracture test on each forging of the sample forgings, obtaining micro-impact record data and fracture record data of the forging, and constructing a basic data set based on the record data of all forgings.

[0019] In an embodiment of the present invention, the sample forgings include a group of forgings having the same material type but different shapes, and each forging group includes forgings having different geometric size data.

[0020] In an embodiment of the present invention, the material type can be determined based on the material that can be used to prepare the forging, for example, it can include material types such as low carbon steel, aluminum alloy and cast iron. The shape type can be determined based on common forging blanks, for example, it can include shape types such as square, cylindrical, flat, etc. Among them, the geometric dimension data of each forging is used to characterize the geometric dimension specifications of the forging, and can be determined based on the corresponding shape type. For example, the geometric dimension data of square forgings and flat forgings include data such as length, width and height, and the geometric dimension data of cylindrical forgings include length and diameter, etc. The number of geometric dimension data of each shape type can be set based on actual needs, and it is sufficient to roughly cover the common forging geometric dimension data.

[0021] Furthermore, in an embodiment of the present invention, the basic data set may include multiple record sets, specifically stored in the form of structured record sets. Each record set corresponds to a forging group and includes the identity of the forging group, micro-impact record data of each forging, and fracture record data.

[0022] In an embodiment of the present invention, the micro-impact recording data may include geometric dimension data, a force-displacement curve, a crack extension area measurement value and a temperature increase value, and the fracture recording data may include geometric dimension data and an impact work measurement value, wherein the force-displacement curve is used to characterize the numerical correspondence between the impact force and the pit depth.

[0023] In an embodiment of the present invention, the identity of each forging is generated using a composite coding rule based on a combination of standardized codes for material type and shape type. Specifically, the material type and shape type are each encoded using independent character systems. Material type coding complies with national standards such as GB / T 13304.1-2008 "Steel Classification," while shape type coding is based on internal enterprise standards based on the geometric characteristics of the forgings. This creates a standardized coding mapping table that is stored in the system database. For example, the material type is coded using letters (e.g., A-Z), while the shape type is coded using numbers (e.g., 0-9). The combination of the two forms a unique identity. In one illustrative example, if a forging's material type is 45# steel, its standardized material code is "C," and its shape type is cylindrical, corresponding to the shape code "1," the identity combination for the forging group is "C1." Different material types (e.g., aluminum alloy, titanium alloy) or shape types (e.g., disc, rectangle) are assigned different codes, ensuring the uniqueness and scalability of the coding system.

[0024] In the embodiment of the present invention, the micro-impact test refers to a pulsed energy loading test performed on the required forging using a drop hammer impact tester (such as INSTRON 9250HV) at 23±2°C. The specific implementation method is as follows: (1) Impact point layout Set N impact positions (N≥3) along the length of the forging. The distance between any two adjacent impact positions is a preset value ΔL (unit: mm), where ΔL meets the following conditions: When the forging length L≤200mm, ΔL=10mm; When the forging length L>200mm, ΔL=20mm; All impact positions are within ±5mm of the center axis of the forging to ensure uniform transmission of impact energy.

[0025] (2) Experimental parameters Impact speed: 10±0.5m / s, precise speed adjustment is achieved through servo motor closed-loop control.

[0026] Impact energy: Dynamically adjusted according to the yield strength σs of the forging material, satisfying E=0.1×σs×A, where E is the impact energy and A is the cross-sectional area of ​​the impact position.

[0027] Sampling frequency: The sampling frequency of the force-displacement curve is ≥10kHz, and a 16-bit ADC module is configured to ensure data acquisition accuracy.

[0028] (3) Step-by-step loading control During the micro-impact test, an incremental impact force loading strategy is used to apply a corresponding impact force to each impact position to form a pit at that position. The impact force of each impact is increased by a fixed value △F (unit: N) compared to the previous one. This fixed value can be pre-set according to the experimental requirements. For example, the first impact force is F1, the second impact force is F1+△F, the third impact force is F1+2△F, and so on, to ensure that the load increment of each impact remains constant. During the specific experimental process, impact forces are applied to the forgings in sequence. The latter impact force is increased by a pre-set fixed value △F on the basis of the previous impact force. The increment of each impact force is precisely adjusted by controlling the loading system to achieve a step-by-step loading test on the forgings.

[0029] (4) Data collection The pit depth at each punching position is collected and the infrared thermal image data of the impact area are recorded synchronously for temperature field analysis.

[0030] In the embodiment of the present invention, the force-displacement curve is a two-dimensional curve with the impact force as the horizontal axis and the pit depth as the vertical axis. The force-displacement curve of each forging is constructed by using the impact force and pit depth data at each impact position in the micro-impact experiment. The specific implementation method is as follows: A rectangular coordinate system is constructed with the impact force (unit: N) as the horizontal axis and the pit depth (unit: mm) as the vertical axis. The measured impact force value and the corresponding pit depth value at each impact position are mapped to discrete points in the coordinate system. The discrete points are fitted into a continuous curve through a smooth interpolation algorithm to obtain the force-displacement curve of the forging. The force-displacement curve is stored in the form of a digital signal as a two-dimensional array, and the array elements are in the form of (F1, D1), (F2, D2), ..., (F n , D n ) coordinate pairs, which facilitates subsequent feature extraction and numerical calculation by the computer program, and n is the number of impact positions.

[0031] In an embodiment of the present invention, the crack propagation area of ​​each forging can be acquired using non-contact image detection technology. Specifically, a high-resolution image scanner digitizes the impact area of ​​the forging, generating a grayscale image with micron-level accuracy. The scanned image is then processed using a deep learning-based image recognition algorithm. First, a semantic segmentation model such as U-Net is used to automatically identify the crack outline. Then, morphological operations are performed to remove noise interference. Finally, based on pixel statistics and scale calibration, the crack area in the two-dimensional image is converted to its actual physical size, thereby accurately acquiring the crack propagation area value. This value is stored in a standardized data format for subsequent fracture energy calculations and model analysis.

[0032] In the embodiment of the present invention, the temperature increase value of each forging can be obtained through the recorded infrared thermal imaging data.

[0033] In an embodiment of the present invention, the fracture record data of each forging is obtained through a standard fracture test, specifically in accordance with national standards such as GB / T 229-2020 "Charpy Pendulum Impact Test Method for Metallic Materials". During the experiment, the forging is processed into a standard impact specimen (such as a V-notch or U-notch specimen), and the impact load is applied using a pendulum impact testing machine. The load-displacement data during the impact process is collected in real time using a high-precision force sensor and displacement encoder, and the impact absorbed energy value is recorded simultaneously. The final fracture record data obtained includes geometric dimension data and impact energy measurement values. All data are calibrated and stored in a digital format to ensure compatibility and traceability with micro-impact test data.

[0034] S200, based on the micro-impact recording data and the impact recording data corresponding to each forging group, obtain the inflection point size association model and the target fracture energy impact work association model corresponding to the forging group; wherein the inflection point size association model is used to characterize the association relationship between the inflection point coordinates and the geometric dimension data, and the target fracture energy impact work association model is used to characterize the association relationship between the fracture energy and the impact work.

[0035] In the embodiment of the present invention, the fracture energy is the energy absorbed per unit area during the fracture process of the material.

[0036] Furthermore, S200 may specifically include: S201 : Processing the force-displacement curve of the forging group to obtain the inflection point coordinates and the area under the curve of the force-displacement curve, and obtaining the fracture energy corresponding to each forging of the forging group based on a pre-stored fracture energy calculation formula.

[0037] In the present embodiment, an inflection point is a point in the force-displacement curve where the curvature undergoes a sudden change, corresponding to mechanical behavior such as material yield or fracture. Inflection point identification utilizes a second-order derivative sign change detection algorithm known in the art (e.g., the finite-difference inflection point detection method disclosed in the Experimental Course in Material Mechanics). By calculating the first and second derivatives of the curve, the location where the curvature sign changes is identified. The specific implementation steps are omitted here as they pertain to the prior art.

[0038] In embodiments of the present invention, the area under the curve represents the energy absorbed by the material and can be calculated using numerical integration methods (such as the trapezoidal method or Simpson's method) or piecewise integration methods. For example, when using the trapezoidal method, the force-displacement curve is discretized into a number of data points, and the total area is obtained by summing the areas of the trapezoids formed by adjacent points. The specific calculation steps refer to the standard method for energy integration in GB / T228.1-2021 "Tension Test of Metallic Materials" and are not further described here.

[0039] In the embodiment of the present invention, the pre-stored fracture energy calculation formula satisfies the following conditions: Gc=Ac / Af•(1+k•△T); wherein Gc is the fracture energy, the unit is J / m 2 ,Ac is the area under the curve, unit is N•mm. Af is the crack extension area, unit is mm 2 k is the temperature correction factor, which can be determined based on the material type. △T is the temperature increase value, in K.

[0040] In this embodiment of the present invention, the fracture energy calculation formula introduces a temperature correction term (1 + k•ΔT) to quantify the effect of temperature rise during the impact process on the material's fracture energy. This is applicable to micro-impact testing scenarios with impact frequencies exceeding 10 Hz or cumulative temperature rises exceeding 10 K. The temperature correction factor is stored in the system database and automatically matched to the material temperature-dependent performance data specified in GB / T229-2020, "Charpy Pendulum Impact Test Method for Metallic Materials."

[0041] S202 , obtaining an estimated impact energy of each forging by inputting the fracture energy of each forging into a pre-stored initial fracture energy-impact energy correlation model.

[0042] In an embodiment of the present invention, the pre-stored initial fracture energy-impact work correlation model can be an empirical formula, a theoretical model, or a simple model derived from historical data. In one exemplary embodiment, the initial fracture energy-impact work correlation model can be expressed as: Epre = a × Gc + b, where Epre is the estimated impact work value, and a and b are fitting coefficients.

[0043] In an illustrative embodiment, the pre-stored initial fracture energy-impact work correlation model can be obtained based on the basic data set initially received. Specifically, the parameters of the initial fracture energy-impact work correlation model can be solved using the least squares method through the fracture energy and impact work measurement values ​​corresponding to a certain forging group initially received, and the model accuracy can be verified using the determination coefficient and mean square error to obtain the initial fracture energy-impact work correlation model.

[0044] S203 , performing regression analysis on all inflection point coordinates and geometric dimension data of the forging group, and establishing an inflection point dimension correlation model between the inflection point coordinates and the geometric dimension data.

[0045] In an embodiment of the present invention, the inflection point dimension association model includes a force dimension association model and a displacement dimension association model. The force dimension association model is used to characterize the correlation between impact force and geometric dimension data, and the displacement dimension association model is used to characterize the correlation between displacement and geometric dimension data.

[0046] In an embodiment of the present invention, an existing regression analysis method can be used to establish an inflection point dimension association model between inflection point coordinates and geometric dimension data. For example, a regression analysis is performed on the inflection point coordinates (including force coordinates and displacement coordinates) and the corresponding geometric dimension data of a forging group. The specific steps are as follows: first, the force coordinates, displacement coordinates, and geometric dimension data are standardized and preprocessed. Then, a multivariate linear regression or polynomial regression algorithm is used to fit the mapping relationship between the inflection point coordinates and the geometric dimension data. The model parameters are solved using the least squares method, and the model accuracy is verified using the coefficient of determination and mean square error. Finally, an inflection point dimension association model is established to characterize the association between the inflection point coordinates and the geometric dimension data. This model can predict the inflection point of the force-displacement curve of the forging based on its dimensions.

[0047] In an exemplary embodiment, the inflection point size correlation model can be expressed as: y=β0+β1x1+…+β z x z +γ; where y is the force coordinate or displacement coordinate, x z is the zth dimension parameter in the geometric dimension data, z is the number of dimension parameters in the geometric dimension data, β0, β1, ..., β z is the regression coefficient, and γ is the error term.

[0048] S204, based on the impact energy estimation value and the impact energy measurement value corresponding to the forging group, obtain the correction coefficient of the forging group, and correct the initial fracture energy impact energy correlation model based on the correction coefficient to obtain the corresponding correction result, and use the correction result as the target fracture energy impact energy correlation model of the forging group.

[0049] In an exemplary embodiment of the present invention, the correction coefficient h of each forging group is the ratio of the average of all impact energy measurements corresponding to the forging group to the average of all impact energy evaluation values, that is, h=AvgEmea / AvgEpre.

[0050] Wherein, AvgEmea is the average of all impact energy measurement values ​​corresponding to the forging group, and AvgEpre is the average of all impact energy evaluation values ​​corresponding to the forging group.

[0051] In another embodiment of the present invention, the correction coefficient of each forging group is obtained based on the least squares method, and the obtained correction coefficient h satisfies the following conditions: .

[0052] in, is the impact energy measurement value of the u-th forging in the forging group, u ranges from 1 to Q, Q is the number of forgings in the forging group, is the impact energy evaluation value of the u-th forging in the forging group.

[0053] In one embodiment of the present invention, the target fracture energy impact work correlation model can be expressed as: Ecorr=h•Epre, wherein Ecorr is the corrected impact work evaluation value, and Epre is the impact work evaluation value obtained based on the initial fracture energy impact work correlation model.

[0054] In another embodiment of the present invention, the target fracture energy and impact work correlation model can be expressed as: , where h1=AvgEmea / AvgEpre, .

[0055] In an embodiment of the present invention, a specific method for determining the target fracture energy impact work correlation model may be based on error indicators such as mean square error, relative error, absolute error, etc., and the specific determination method may be existing technology.

[0056] Those skilled in the art are aware that, in actual application scenarios, the target fracture energy and impact work correlation model can be dynamically updated during use, and the specific updating method can adopt the existing method. For example, after 100 valid data are processed cumulatively, the correction coefficient is automatically recalculated using the new data. If the prediction accuracy of the new model is higher than the prediction accuracy of the original model by a preset value, it is updated to the current valid model.

[0057] S300: Associating the identity identifier, inflection point size correlation model, and target fracture energy impact work correlation model of each forging group, and storing them in the current database.

[0058] In an embodiment of the present invention, the identity identifier, inflection point size correlation model, and target fracture energy impact work correlation model of each forging group may be stored as a record, and the specific storage method may be an existing method.

[0059] S400 , when a forging performance test request is received, a micro-impact test is performed on the requested forging and the impact energy of the requested forging is predicted based on an inflection point size correlation model and a target fracture energy impact energy correlation model in a current database.

[0060] Furthermore, S400 specifically includes: S401 , performing a micro-impact test on the requested forging to obtain micro-impact record data of the requested forging.

[0061] The specific implementation of performing the micro-impact test on the requested forging may refer to the above content.

[0062] S402 : Processing the force-displacement curve in the micro-impact recording data of the requested forging to obtain corresponding inflection point coordinates and the area under the curve, and using the obtained inflection point coordinates as coordinates to be processed.

[0063] S403, using the identity identifier of the requested forging as input, performing a query operation in the current database, if the corresponding identity identifier is found, that is, the same identity identifier is found, executing S404.

[0064] In an embodiment of the present invention, if the corresponding identity identifier cannot be found in the search, a prompt message such as "there is no identity identifier corresponding to the requested data in the database" may be output.

[0065] S404: Input the geometric dimension data corresponding to the global request into the inflection point dimension association model corresponding to the retrieved identity identifier to obtain the corresponding inflection point coordinates as reference coordinates.

[0066] S405 , obtaining the difference between the coordinate to be processed and the reference coordinate. If the difference is less than a set value, executing S406 ; otherwise, outputting a prompt message indicating that an abnormality exists in the force-displacement curve in the requested data.

[0067] In this embodiment of the present invention, the difference between the coordinates to be processed and the reference coordinates satisfies the following conditions: △d=(((F p -F c ) / F c ) 2 + ((δ p -δ c ) / δ c ) 2 ) 1 / 2 ; Among them, △d is the difference between the coordinate to be processed and the reference coordinate, F p is the force coordinate in the coordinate to be processed, F c is the force coordinate in the reference coordinate, δ p is the displacement coordinate in the coordinate to be processed, δ c is the displacement coordinate in the reference coordinate.

[0068] In the embodiments of the present invention, the set value may be determined based on at least one of the following: (1) the measurement accuracy and safety factor of the force sensor and displacement sensor; (2) the historical statistical distribution of the coordinate differences of the normal forging force-displacement curve (e.g., the 95th percentile); (3) the mechanical property tolerance requirements of the target forging material; and (4) the coordinate difference threshold value obtained through finite element simulation. Generally speaking, in high-precision testing scenarios, the set value will be smaller, such as 0.05; while in tests with ordinary precision requirements, the set value may be relaxed to 0.1.

[0069] The force-displacement curve has key characteristic points, such as the elastic limit, yield point, and fracture point. By comparing the coordinates of these key points between the curve to be processed and the reference curve, if there is a significant deviation in the force or displacement values ​​at a key point, it can be determined that an anomaly exists at that key point and in the surrounding area. For example, if the yield point displacement value of the curve to be processed is much earlier than that of the reference curve, it means that the material may begin to yield at a lower displacement. The area near the yield point is the location of the anomaly, which may indicate a problem with the material's internal structure or performance.

[0070] In an embodiment of the present invention, the output prompt information includes but is not limited to "the force-displacement curve data is abnormal, and the difference value exceeds the set range", and the currently calculated difference value and the set value are displayed at the same time, so that the operator can quickly understand the abnormal situation. In addition, prompts can be displayed in a prominent pop-up window on the user interface, and the system log will record the time of the exception, the source of the requested data, and the specific exception information. In addition, alerts can be sent to relevant technical personnel via email or instant messaging tools to ensure that the abnormal situation is handled promptly.

[0071] S406, based on a pre-stored fracture energy calculation formula, obtaining the fracture energy corresponding to the requested forging, and inputting the obtained fracture energy of the requested forging into a target fracture energy impact work association model corresponding to the queried identity identifier to obtain the impact work corresponding to the requested forging.

[0072] Those skilled in the art are aware that impact energy values ​​intuitively reflect a material's ability to absorb energy under impact loads and play a key role in many practical scenarios, particularly in forging performance testing, such as forging material quality assessment, production process optimization, failure analysis and life prediction, new product development and material selection, and quality certification and standard setting. Therefore, based on actual needs, the impact energy obtained in S406 can be used to analyze the impact performance of the requested forging.

[0073] Furthermore, S406 may include storing the output impact energy data in the current database, categorizing it by fields such as identity and detection time to facilitate subsequent query and traceability. Furthermore, the output impact energy data may be shared with related systems such as quality management systems and production scheduling systems via an API interface, enabling collaborative work and decision-making analysis across departments based on this data.

[0074] In summary, the intelligent detection method for the impact performance of forging blanks provided by the embodiments of the present invention, by integrating micro-impact test data with impact test data to construct an inflection point size correlation model and a target fracture energy impact work correlation model, has at least the following advantages: (1) Non-destructive testing and efficient modeling This embodiment of the present invention overcomes the destructive limitations of traditional impact testing. Using only micro-impact test data (including force-displacement curves, crack growth area, and so on) from a small number of forging samples and the corresponding impact data, a high-precision prediction model can be automatically constructed using a computer program. Compared to the cumbersome process of traditional methods that require forgings to be broken open to obtain cross-sectional images, this solution significantly improves testing efficiency while avoiding the destructive loss of high-value forgings.

[0075] (2) Data-driven accurate prediction of impact performance The embodiment of the present invention is based on the deep mining of micro-impact experimental data by machine learning algorithms. The model can automatically extract the mapping relationship between the inflection point characteristics of the force-displacement curve and the forging size. Combined with the dynamic correction model of fracture energy and impact work, it can reduce the impact work prediction error.

[0076] (3) Intelligent and automated testing process The embodiment of the present invention constructs a full-link intelligent detection system of "data acquisition-model training-real-time prediction", which can automatically analyze the digitized force-displacement curve of the micro-impact experiment without manual labeling of feature points, and based on the material type-shape type association model pre-stored in the database, it can achieve second-level prediction of the impact performance of new forgings.

[0077] (4) Support batch inspection of multiple batches of forgings The embodiment of the present invention can construct a corresponding inflection point size correlation model and a target fracture energy impact work correlation model based on the micro-impact test data and fracture data of a small amount of forging sample data. In practical applications, the impact work similar to the fracture data can be obtained based only on the micro-impact test data of the forging, so as to analyze the detection performance of the forging, thereby improving the detection efficiency and data processing efficiency of the forging detection performance.

[0078] Based on the same inventive concept, an embodiment of the present invention provides an intelligent detection system for the impact performance of forging blanks, the system comprising: A sample processing module is configured to perform micro-impact experiments and fracture experiments on sample forgings, obtain micro-impact record data and fracture record data, and construct a basic data set; wherein the sample forgings include a group of forgings with the same material type but different shape types, and each forging group includes forgings with different geometric dimension data; the basic data set includes multiple record sets, each record set corresponds to a forging group, and includes an identity identifier of the forging group, micro-impact record data and fracture record data, the micro-impact record data includes geometric dimension data, a force-displacement curve, a crack extension area measurement value and a temperature increase value, the fracture record data includes geometric dimension data and an impact work measurement value, and the force-displacement curve is used to characterize the numerical correspondence between the impact force and the pit depth.

[0079] In an embodiment of the present invention, the sample processing module is configured to perform a micro-impact test and a fracture test on a sample forging, which means that the sample processing module is configured to perform the micro-impact test and the fracture test on the sample forging by sending corresponding control instructions to the experimental equipment, for example, including: sending impact parameter control instructions (impact velocity, energy, etc.) to a drop hammer impact tester (such as INSTRON 9250HV) through a GPIB / USB interface; receiving experimental data such as force-displacement curves (sampling frequency ≥10kHz) and infrared thermal images through a 16-bit ADC data acquisition card; and storing the experimental data in a structured record set according to the forging group classification.

[0080] The model building module is configured to obtain the inflection point size association model and the target fracture energy impact work association model corresponding to each forging group based on the micro-impact record data and the impact record data corresponding to the forging group; wherein the inflection point size association model is used to characterize the association relationship between the inflection point coordinates and the geometric dimension data, and the target fracture energy impact work association model is used to characterize the association relationship between the fracture energy and the impact work.

[0081] The data storage module is configured to associate the identity identification, inflection point size association model and target fracture energy impact work association model of each forging group and store them in the current database.

[0082] The prediction module is configured to, when receiving a forging performance test request, perform a micro-impact test on the requested forging and predict the impact energy of the requested forging based on the inflection point size correlation model and the target fracture energy impact energy correlation model in the current database.

[0083] In an embodiment of the present invention, the modules of the system interact in the following ways: the sample processing module transmits the structured record set to the input queue of the model construction module; the model trained by the model construction module is registered to the data storage module through the API interface; the prediction module pulls the model from the data storage module and performs impact energy prediction in combination with real-time experimental data.

[0084] The system can be used to perform Figure 1 The method shown in the embodiment shown, therefore, for the functions that can be realized by each functional module of the device, please refer to Figure 1 The description of the illustrated embodiment is omitted for brevity.

[0085] An embodiment of the present invention also provides an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are configured to execute the method described in the embodiment of the present invention.

[0086] An embodiment of the present invention further provides a computer-readable storage medium storing computer-executable instructions, wherein the computer instructions are used to execute the method described in the embodiment of the present invention.

[0087] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the present invention can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in the present invention can be achieved. This is not limited herein.

[0088] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. An intelligent detection method for the impact performance of forging blanks, characterized in that: The method comprises the following steps: S100, performing a micro-impact test and a fracture test on each forging in the sample forgings, obtaining micro-impact record data and fracture record data of the forging, and constructing a basic data set based on the record data of all forgings; wherein the sample forgings include a group of forgings having the same material type but different shapes, each forging group including forgings having different geometric dimension data; the basic data set includes multiple record sets, each record set corresponding to a forging group, and including an identity identifier of the forging group, micro-impact record data, and fracture record data; the micro-impact record data includes geometric dimension data, a force-displacement curve, a crack extension area measurement value, and a temperature increase value; the fracture record data includes geometric dimension data and an impact energy measurement value; the force-displacement curve is used to characterize the numerical correspondence between impact force and pit depth; S200, based on the micro-impact recording data and the impact recording data corresponding to each forging group, obtaining an inflection point size correlation model and a target fracture energy impact work correlation model corresponding to the forging group; wherein the inflection point size correlation model is used to characterize the correlation relationship between the inflection point coordinates and the geometric dimension data, and the target fracture energy impact work correlation model is used to characterize the correlation relationship between the fracture energy and the impact work; S300, associating the identity identifier, inflection point size correlation model, and target fracture energy impact work correlation model of each forging group, and storing them in the current database; S400 , when a forging performance test request is received, a micro-impact test is performed on the requested forging and the impact energy of the requested forging is predicted based on an inflection point size correlation model and a target fracture energy impact energy correlation model in a current database.

2. The method according to claim 1, characterized in that S200 specifically includes: S201, processing the force-displacement curve of the forging group to obtain the inflection point coordinates and the area under the curve of the force-displacement curve, and obtaining the fracture energy corresponding to each forging in the forging group based on a pre-stored fracture energy calculation formula; S202, inputting the fracture energy of each forging into a pre-stored initial fracture energy-impact energy correlation model to obtain an estimated impact energy value of the forging; S203, performing regression analysis on all inflection point coordinates and geometric dimension data of the forging group, and establishing an inflection point dimension correlation model between the inflection point coordinates and the geometric dimension data; S204, based on the impact energy estimation value and the impact energy measurement value corresponding to the forging group, obtain the correction coefficient of the forging group, and correct the initial fracture energy impact energy correlation model based on the correction coefficient to obtain the corresponding correction result, and use the correction result as the target fracture energy impact energy correlation model of the forging group.

3. The method according to claim 1, characterized in that The identity of each forging group is determined based on the corresponding material type and shape type.

4. The method according to claim 3, characterized in that in, S400 specifically includes: S401, performing a micro-impact test on the requested forging to obtain micro-impact record data of the requested forging; S402, processing the force-displacement curve in the micro-impact recording data of the requested forging to obtain the corresponding inflection point coordinates and the area under the curve, and using the obtained inflection point coordinates as coordinates to be processed; S403, using the identity of the requested forging as input, performing a query operation in the current database. If the corresponding identity is found, executing S404; S404: Input the geometric dimension data corresponding to the requested forging into the inflection point dimension association model corresponding to the retrieved identity identifier to obtain the corresponding inflection point coordinates as reference coordinates; S405, obtaining the difference between the coordinate to be processed and the reference coordinate. If the difference is less than the set value, executing S406; otherwise, outputting a prompt message indicating that the force-displacement curve of the requested forging is abnormal. S406, based on a pre-stored fracture energy calculation formula, obtaining the fracture energy corresponding to the requested forging, and inputting the obtained fracture energy of the requested forging into a target fracture energy impact work association model corresponding to the queried identity identifier to obtain the impact work corresponding to the requested forging.

5. The method according to claim 4, characterized in that The difference between the coordinates to be processed and the reference coordinates meets the following conditions: △d=(((F p -F c ) / F c ) 2 + ((δ p -δ c ) / δ c ) 2 ) 1 / 2 ; Among them, △d is the difference between the coordinate to be processed and the reference coordinate, F p is the force coordinate in the coordinate to be processed, F c is the force coordinate in the reference coordinate, δ p is the displacement coordinate in the coordinate to be processed, δ c is the displacement coordinate in the reference coordinate.

6. The method according to claim 2, characterized in that The correction factor for each forging group meets the following conditions: h=AvgEmea / AvgEpre; Wherein, h is the correction coefficient of the forging group, AvgEmea is the mean of all impact energy measurements corresponding to the forging group, and AvgEpre is the mean of all impact energy evaluation values ​​corresponding to the forging group.

7. The method according to claim 2, characterized in that The correction factor for each forging group meets the following conditions: ； Where h is the correction factor of the forging group, is the impact energy measurement value of the u-th forging in the forging group, u ranges from 1 to Q, Q is the number of forgings in the forging group, is the impact energy evaluation value of the u-th forging in the forging group.

8. The method according to claim 2, characterized in that The pre-stored fracture energy calculation formula satisfies the following conditions: Gc=Ac / Af•(1+k•△T); where Gc is the fracture energy, Ac is the area under the curve, Af is the crack propagation area, k is the temperature correction coefficient, and △T is the temperature increase value.

9. An intelligent detection system for the impact performance of forging blanks, characterized in that: The system comprises: a sample processing module configured to perform micro-impact and impact tests on sample forgings, obtain micro-impact record data and impact record data, and construct a basic data set; wherein the sample forgings include a group of forgings having the same material type but different shapes, each forging group including forgings having different geometric dimension data; the basic data set includes multiple record sets, each record set corresponding to a forging group, and including an identity identifier of the forging group, micro-impact record data, and impact record data; the micro-impact record data includes geometric dimension data, a force-displacement curve, a crack extension area measurement value, and a temperature increase value; the impact record data includes geometric dimension data and an impact energy measurement value; the force-displacement curve is used to characterize the numerical correspondence between impact force and pit depth; a model building module configured to obtain an inflection point size correlation model and a target fracture energy impact work correlation model corresponding to each forging group based on the micro-impact record data and the impact record data corresponding to the forging group; wherein the inflection point size correlation model is used to characterize the correlation relationship between the inflection point coordinates and the geometric dimension data, and the target fracture energy impact work correlation model is used to characterize the correlation relationship between the fracture energy and the impact work; a data storage module configured to associate the identity identifier, the inflection point size association model, and the target fracture energy impact work association model of each forging group, and store the associations in a current database; The prediction module is configured to, when receiving a forging performance test request, perform a micro-impact test on the requested forging and predict the impact energy of the requested forging based on the inflection point size correlation model and the target fracture energy impact energy correlation model in the current database.

Images