Method and device for predicting three-dimensional stress-strain curve by using ultrasonic wave

Abstract

The present disclosure relates to a method and device for predicting a three-dimensional stress-strain curve by using an ultrasonic wave. More specifically, the method comprises the steps of: emitting an ultrasonic signal to a local area of a specimen with an ultrasonic transceiver to measure an ultrasonic signal; inputting the measured ultrasonic signal to an artificial intelligence model that has been trained to calculate and output a first parameter on the basis of an input ultrasonic signal; inputting the first parameter to a Ramberg-Osgood model when the first parameter calculated on the basis of the measured ultrasonic signal is output from the artificial intelligence model; scanning the ultrasonic transceiver to repeatedly perform each of the above steps for the next local area of the specimen when a second parameter calculated on the basis of the first parameter is output from the Ramberg-Osgood model; and visualizing a three-dimensional stress-strain curve for the entire area of the specimen when all the second parameters for the entire area of the specimen are collected.

Description

Method and apparatus for predicting 3D stress-strain curves using ultrasound

[0001] The present disclosure relates to a method and apparatus for predicting a three-dimensional stress-strain curve using ultrasound.

[0002] Destructive Tensile Testing (DTT) is the most common material test used to determine the stress-strain relationship of a material. To measure a material's resistance (deformation and fracture), the test is conducted by slowly pulling a specimen (or test coupon) at a uniform speed and with a uniform force until fracture, while simultaneously measuring and recording the applied force and the deformation of the specimen.

[0003] As a result, a stress-strain curve for the material can be obtained. This stress-strain curve is a two-dimensional curve obtained through a tensile test that represents the stress-strain relationship of the material.

[0004] Stress-strain curves are used to indicate mechanical properties of a material, such as the proportional limit, elastic limit, yield point, yield strength, ultimate tensile strength, and fracture point.

[0005] The background art is provided to facilitate understanding of the present disclosure. It should not be understood as an acknowledgment that the matters described in the background art exist as prior art.

[0006] Previously, conducting tensile tests involved the cumbersome process of extracting specimens from the material and reprocessing them to meet specifications. Furthermore, since the specimens are permanently destroyed after the test and cannot be reused, material waste occurs. In other words, there are problems involving not only high inspection costs, such as material and processing fees, but also significant inspection time.

[0007] For this reason, it is also impossible to perform inline inspection during the manufacturing process.

[0008] Meanwhile, to perform a tensile test, the specimen size must be larger than the minimum specification limit (approximately 20 mm in length), and applying it to parts smaller than this size is very difficult as it requires additionally complex equipment and procedures. Furthermore, even if the minimum specification limit is met, conventional tensile testing cannot selectively obtain tensile properties only in areas of local damage or local property changes for parts with local damage at the level of a few mm or materials where physical properties can change locally (e.g., additive manufacturing materials or composite materials). In other words, the two-dimensional stress-strain curve obtained from conventional tensile testing has a limitation in that it provides only the average physical property value over the entire specimen area, thus failing to sufficiently reflect information regarding local damage or changes in physical properties.

[0009] Accordingly, the inventor of the present disclosure has come to invent a method and apparatus capable of predicting a three-dimensional stress-strain curve for a test specimen (inspection body) in a non-destructive manner.

[0010] Accordingly, the problem to be solved by the present disclosure is to provide a method and apparatus for predicting a three-dimensional stress-strain curve using ultrasound, which can reduce the cost or time required for manufacturing and inspection by enabling tensile testing to be performed in a short time using ultrasonic signals without the need to extract a test specimen (inspection body) from a part, as well as enabling the reuse of the test specimen based on a non-destructive method and allowing in-line inspection during the process.

[0011] Meanwhile, another problem to be solved by the present disclosure is to provide a method and apparatus for predicting a three-dimensional stress-strain curve using ultrasound, which can predict a three-dimensional stress-strain curve applicable to localized damage that cannot be detected in existing two-dimensional stress-strain curves or to micro-components of the same size as semiconductors.

[0012] The problems of the present disclosure are not limited to those mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the description below.

[0013] The effects according to the present disclosure are not limited to those exemplified above, and a wider variety of effects are included within the present disclosure. To solve the problems described above, a method for predicting a three-dimensional stress-strain curve using ultrasound according to an embodiment of the present disclosure is provided. The method may include the steps of: measuring an ultrasonic signal by irradiating an ultrasonic signal onto a local area of ​​a test specimen using an ultrasonic transceiver; inputting the measured ultrasonic signal to an artificial intelligence model that has been pre-trained to calculate and output a first parameter based on the input ultrasonic signal; inputting the first parameter to a Lamborgh-Osgood model when the first parameter calculated based on the measured ultrasonic signal is output from the artificial intelligence model; scanning the ultrasonic transceiver to repeat each of the above steps for the next local area of ​​the test specimen when the second parameter calculated based on the first parameter is output from the Lamborgh-Osgood model; and visualizing a three-dimensional stress-strain curve for the entire area of ​​the test specimen when all second parameters for the entire area of ​​the test specimen have been collected.

[0014] According to the features of the present disclosure, the step of measuring the ultrasonic signal may involve simultaneously measuring at least one of the following as at least one ultrasonic parameter: longitudinal wave velocity, attenuation coefficient, back-scattering coefficient, second and third-order acoustic nonlinearity parameters, shear wave velocity, shear wave attenuation coefficient, or shear wave back-scattering coefficient of the measured ultrasonic signal.

[0015] According to the features of the present disclosure, the at least one ultrasonic parameter may represent at least one component of the bulk wave, surface acoustic wave, Lamb wave, or wave mixing of the measured ultrasonic signal.

[0016] According to the features of the present disclosure, the artificial intelligence model may be pre-trained based on a machine learning method or a deep learning method so as to be able to non-destructively predict a stress-strain curve for a local region of the test specimen using at least one ultrasonic parameter.

[0017] According to the features of the present disclosure, the first parameter may be at least one unknown value for the Lamborgh-Osgood model calculated using the at least one ultrasonic parameter.

[0018] According to the features of the present disclosure, the second parameter is a strain ( ), Strain hardening modulus ( ), stress near ultimate tensile strength ( ) and strain corresponding to stress near the ultimate tensile strength ( It may include at least one of ).

[0019] According to the features of the present disclosure, the at least one unknown value comprises at least one of an elastic modulus or a yield stress near each preset strain, wherein the elastic modulus is calculated using a relationship based on ultrasonic longitudinal and transverse wave velocities, and the yield stress near each strain may be calculated using an ultrasonic nonlinear elastic modulus relationship or by using an ultrasonic damping coefficient and a Hall-petch equation together.

[0020] According to the features of the present disclosure, the step of measuring the ultrasonic signal may be to measure the ultrasonic signal based on a contact ultrasonic measurement method or a non-contact ultrasonic measurement method.

[0021] According to the features of the present disclosure, the three-dimensional stress-strain curve may be composed of a new axis that adds an ultrasonic scanning direction to the stress and strain axes of the predicted two-dimensional stress-strain curve.

[0022] To solve the problem described above, a three-dimensional stress-strain curve prediction device using ultrasound according to one embodiment of the present disclosure is provided. The device comprises: a communication interface; a memory; and a processor operably connected to the communication interface and the memory, wherein the processor may be configured to measure an ultrasonic signal by irradiating an ultrasonic signal to a local area of ​​a test specimen using an ultrasonic transceiver, input the measured ultrasonic signal to an artificial intelligence model that has been pre-trained to calculate and output a first parameter based on the input ultrasonic signal, input the first parameter to a Lamborgh-Osgood model when the first parameter calculated based on the measured ultrasonic signal is output from the artificial intelligence model, and when a second parameter calculated based on the first parameter is output from the Lamborgh-Osgood model, scan the ultrasonic transceiver to repeat each of the above operations for the next local area of ​​the test specimen, and when all second parameters for the entire area of ​​the test specimen are collected, visualize a three-dimensional stress-strain curve for the entire area of ​​the test specimen.

[0023] Specific details of other embodiments are included in the detailed description and drawings.

[0024] The present disclosure enables tensile testing to be performed in a short time using ultrasonic signals without the need to extract a test specimen (inspection body) from a part, and also enables the reuse of the test specimen based on a non-destructive method, thereby allowing in-line inspection during the process, which can reduce the cost or time required for manufacturing and inspection.

[0025] Meanwhile, the present disclosure enables the prediction of a three-dimensional stress-strain curve that can be applied to localized damage that cannot be detected in conventional two-dimensional stress-strain curves, or to micro-components such as semiconductors.

[0026] FIG. 1 is a drawing for explaining a comparison between a destructive method-based tensile test according to a comparative example of the present disclosure and a non-destructive method-based tensile test according to one embodiment of the present disclosure.

[0027] FIG. 2 is a schematic diagram illustrating a three-dimensional stress-strain curve prediction service providing system using ultrasound according to one embodiment of the present disclosure.

[0028] FIG. 3 is a block diagram showing the configuration of a three-dimensional stress-strain curve prediction device using ultrasound according to one embodiment of the present disclosure.

[0029] FIG. 4 is a block diagram showing the configuration of a user terminal using a three-dimensional stress-strain curve prediction service using ultrasound according to one embodiment of the present disclosure.

[0030] FIG. 5 is a flowchart schematically illustrating a method for predicting a three-dimensional stress-strain surface using ultrasound according to one embodiment of the present disclosure.

[0031] FIG. 6 is a drawing illustrating an example of a method for measuring ultrasonic parameters according to one embodiment of the present disclosure.

[0032] FIG. 7 is a diagram illustrating a series of procedures for reconstructing a predicted stress-strain curve using ultrasonic parameters measured according to one embodiment of the present disclosure.

[0033] FIG. 8 is a diagram showing the layer configuration of an artificial intelligence model for calculating unknown values ​​of a Lamborgh-Osgood model according to one embodiment of the present disclosure.

[0034] FIG. 9 is a drawing showing an example of comparing the predicted stress-strain curves and the error between the stress-strain curves for each of a material with high stiffness and a material with low stiffness, according to one embodiment of the present disclosure.

[0035] FIG. 10 is a drawing for illustrating examples of three-dimensional stress-strain curves visualized through a method for predicting three-dimensional stress-strain curves using ultrasound according to one embodiment of the present disclosure.

[0036] The advantages and features of the present disclosure and the methods for achieving them will become clear by referring to the embodiments described below in detail together with the accompanying drawings. However, the present disclosure is not limited to the embodiments disclosed below but may be implemented in various different forms. These embodiments are provided merely to ensure that the disclosure is complete and to fully inform those skilled in the art of the scope of the invention, and the present disclosure is defined only by the scope of the claims. In connection with the description of the drawings, similar reference numerals may be used for similar components.

[0037] In this document, expressions such as "have," "can have," "include," or "can include" refer to the existence of the relevant feature (e.g., numerical values, functions, actions, or components, etc.) and do not exclude the existence of additional features.

[0038] In this document, expressions such as “A or B,” “at least one of A or / and B,” or “one or more of A or / and B” may include all possible combinations of items listed together. For example, “A or B,” “at least one of A and B,” or “at least one of A or B” may refer to cases including (1) at least one A, (2) at least one B, or (3) both at least one A and at least one B.

[0039] Expressions such as "first," "second," "first," or "second" used in this document may modify various components regardless of order and / or importance, and are used merely to distinguish one component from another without limiting such components. For example, the first user device and the second user device may represent different user devices regardless of order or importance. For example, without departing from the scope of rights set forth in this document, the first component may be named the second component, and similarly, the second component may be renamed the first component.

[0040] Where it is stated that a certain component (e.g., a first component) is "(operatively or communicatively) coupled with" or "connected to" another component (e.g., a second component), it should be understood that the said certain component may be directly connected to the said other component or connected through another component (e.g., a third component). On the other hand, where it is stated that a certain component (e.g., a first component) is "directly connected" or "directly connected" to another component (e.g., a second component), it may be understood that no other component (e.g., a third component) exists between the said certain component and the said other component.

[0041] As used in this document, the expression “configured to” may be replaced, depending on the context, with, for example, “suitable for,” “having the capacity to,” “designed to,” “adapted to,” “made to,” or “capable of.” The term “configured to” does not necessarily mean “specifically designed to” in hardware. Instead, in some situations, the expression “device configured to” may mean that the device is “capable of” in conjunction with other devices or components. For example, the phrase “processor configured to perform A, B, and C” may mean a dedicated processor for performing those operations (e.g., an embedded processor) or a generic-purpose processor (e.g., a CPU or application processor) capable of performing those operations by executing one or more software programs stored in a memory device.

[0042] The terms used in this document are used merely to describe specific embodiments and are not intended to limit the scope of other embodiments. Singular expressions may include plural expressions unless the context clearly indicates otherwise. Terms used herein, including technical or scientific terms, may have the same meaning as generally understood by those skilled in the art described in this document. Terms used in this document that are defined in general dictionaries may be interpreted as having the same or similar meaning as they have in the context of the relevant technology, and are not to be interpreted in an ideal or overly formal sense unless explicitly defined in this document. In some cases, even terms defined in this document may not be interpreted to exclude the embodiments of this document.

[0043] The features of each of the various embodiments of the present disclosure may be combined or combined with one another, either partially or wholly, and as will be fully understood by those skilled in the art, various technical interlocking and operation are possible, and each embodiment may be implemented independently of one another or together in an interlocking relationship.

[0044] For clarity in the interpretation of this specification, the terms used in this specification are defined below.

[0045] The device referred to as a “service server” below may mean, but is not limited to, a physically independent server according to the present disclosure, a virtual machine, and is intended to encompass all modules, programs, or Docker, etc., operating on a virtual or physical machine.

[0046] The term "three-dimensional stress-strain curve" as used in this specification refers to a three-dimensional representation in which the strain of a test specimen is predicted in response to stress, and may also be expressed as a "three-dimensional stress-strain surface," and the method of representation is not limited.

[0047] Hereinafter, the present disclosure will be described in detail by explaining preferred embodiments of the present disclosure with reference to the attached drawings.

[0048] FIG. 1 is a drawing for explaining a comparison between a destructive method-based tensile test according to a comparative example of the present disclosure and a non-destructive method-based tensile test according to one embodiment of the present disclosure.

[0049] Referring to FIG. 1, (a) shows a series of procedures for obtaining a stress-strain curve by conducting a tensile test based on a destructive method according to a comparative example, and (b) shows a series of procedures for obtaining a three-dimensional stress-strain surface by conducting a tensile test based on a non-destructive method according to one embodiment of the present disclosure.

[0050] First, as illustrated in Fig. 1 (a), the tensile test according to the comparative example is conducted by slowly pulling the test specimen of the material with a uniform speed and force until it breaks, in order to measure the strength and deformation of the material such as elastic modulus, yield strength, and elongation, while simultaneously measuring and recording the applied force, i.e., stress, and the strain of the test specimen in relation to that stress.

[0051] As explained above, the tensile test according to this comparative example requires extracting the test specimen from the material and reprocessing it, and the test specimen is permanently destroyed as a result of the tensile test and cannot be reused. In this case, the size of the test specimen must also be larger than the minimum specification limit (e.g., 20 mm length), and it cannot be applied to parts smaller than this.

[0052] Meanwhile, as shown in Fig. 1(b), in order to obtain a three-dimensional stress-strain surface, three steps must be taken: ultrasonic measurement (step 1), stress-strain curve prediction (step 2), and stress-strain surface visualization (step 3).

[0053] First, in the first step, an ultrasonic signal is measured by irradiating a local area of ​​the test specimen with an ultrasonic transceiver. At this time, the linear and nonlinear parameters of the ultrasonic signal (ultrasonic parameters), which are important for predicting the stress-strain curve, are automatically measured quickly and accurately.

[0054] Subsequently, in the second step, at least one parameter for predicting a two-dimensional stress-strain curve for a local region is calculated based on the measured ultrasonic signal. To this end, a pre-trained artificial intelligence model and a Lamborgh-Osgood model may be used together.

[0055] Subsequently, in the third step, a three-dimensional stress-strain curve is visualized by applying it to the entire area of ​​the specimen along with ultrasonic scanning. As such, as ultrasonic scanning proceeds, a two-dimensional stress-strain curve for each local region can be predicted, and these two-dimensional stress-strain curves can be combined to create a three-dimensional representation, thereby providing a global view of the material properties.

[0056] Below, we will specifically examine the operation of predicting a tensile surface based on a non-destructive method according to one embodiment of the present disclosure.

[0057] FIG. 2 is a schematic diagram illustrating a three-dimensional stress-strain curve prediction service providing system using ultrasound according to one embodiment of the present disclosure.

[0058] Referring to FIG. 2, a three-dimensional stress-strain curve prediction service providing system (1000) using ultrasound according to one embodiment of the present disclosure (hereinafter referred to as the "service providing system") may be a system configured to generate a three-dimensional stress-strain curve by performing a tensile test based on a non-destructive method on a test specimen when a tensile test on the test specimen is requested, and to provide the three-dimensional stress-strain curve. Accordingly, the service providing system (1000) may include a service server (100) and a user terminal (300).

[0059] Here, the service server (100) may correspond to a web server. In order to provide a service (hereinafter referred to as the "prediction service") for predicting a three-dimensional stress-strain curve for a test specimen according to a user's request, this service server (100) may be connected to an ultrasonic measuring device (200), a user terminal (300), and / or a separate external device (not shown) to transmit and receive various notifications, requests, information, data, etc. At this time, the prediction service may be based on a separate web page or platform (application).

[0060] Additionally, the service server (100) can build at least one artificial intelligence model by pre-training it to provide a prediction service.

[0061] Meanwhile, the ultrasonic measuring device (200) may be the ultrasonic transceiver itself or a separate device equipped with an ultrasonic transceiver. The ultrasonic measuring device (200) can irradiate an ultrasonic signal onto a test specimen to measure the ultrasonic signal and transmit it to a service server (100). By doing so, the service server (100) can generate a three-dimensional stress-strain curve based on the measured ultrasonic signal.

[0062] At this time, the ultrasonic transceiver may be configured to be used as both a transmitter and a receiver by having one ultrasonic sensor, or configured to be used as a transmitter and a receiver respectively by having two ultrasonic sensors.

[0063] Meanwhile, the user terminal (300) is a collective term for a terminal possessed by a user who is pre-registered on the service server (100) to receive (use) the prediction service, and there may be at least one. At this time, each user terminal (300) can receive the prediction service by executing a separate web page or platform (application) provided by the service server (100).

[0064] When each user terminal (300) receives a three-dimensional stress-strain curve generated based on an ultrasonic signal measured by an ultrasonic measuring device (200) from a service server (100), the three-dimensional stress-strain curve can be visually displayed on a display module provided in the user terminal (300) or on a separate display device connected (linked) to the user terminal (300) to provide it to the user.

[0065] Each of the user terminals (300) described above may be at least one, and may be a computer, UMPC (Ultra Mobile PC), workstation, netbook, PDA (Personal Digital Assistants), portable computer, web tablet, wireless phone, mobile phone, smartphone, pad, smart watch, wearable terminal, e-book, PMP (portable multimedia player), portable game console, navigation device, black box or digital camera, other mobile communication terminal, etc., capable of installing and running multiple applications (i.e., applications) desired by each user, and is not limited to the terminals mentioned above.

[0066] The service provision system (1000) described above is illustrated as having an ultrasonic measuring device (200) that measures ultrasonic signals and a service server (100) that generates a three-dimensional stress-strain curve based on the measured ultrasonic signals, but it is possible to perform all operations, including measuring ultrasonic signals in the ultrasonic measuring device (200) and generating a three-dimensional stress-strain curve based on the measured ultrasonic signals, without having a separate service server (100). That is, in the latter case, the service provision system (1000) may be composed only of an ultrasonic measuring device (200) and a user terminal (300), and in this case, the ultrasonic measuring device (200) performs all of the series of operations of measuring ultrasonics and predicting a three-dimensional stress-strain curve, and may be named a three-dimensional stress-strain curve prediction device (200).

[0067] In other words, the service provision system (1000) is not limited to the configuration shown in FIG. 2, and may be configured to include other devices (terminals, servers, etc.) or may be configured to exclude some configurations.

[0068] In the following, the service server (100) is excluded, and the present disclosure is described based on a service providing system (1000) composed only of a three-dimensional stress-strain curve prediction device (200) and a user terminal (300).

[0069] FIG. 3 is a block diagram showing the configuration of a three-dimensional stress-strain curve prediction device using ultrasound according to one embodiment of the present disclosure.

[0070] Referring to FIG. 3, the three-dimensional stress-strain curve prediction device (200) may include a communication interface (210), a memory (220), an I / O interface (230), and a processor (240), and each component may communicate with one or more communication buses or signal lines.

[0071] The communication interface (210) can be connected to the user terminal (300) as well as other devices through a wired / wireless communication network to exchange data.

[0072] Meanwhile, a communication interface (210) that enables the transmission and reception of such data includes a wired communication pod (211) and a wireless circuit (212), wherein the wired communication port (211) may include one or more wired interfaces, for example, Ethernet, Universal Serial Bus (USB), FireWire, etc. Additionally, the wireless circuit (212) may transmit and receive data with an external device via an RF signal or an optical signal. Furthermore, wireless communication may use at least one of a plurality of communication standards, protocols, and technologies, such as GSM, EDGE, CDMA, TDMA, Bluetooth, Wi-Fi, VoIP, Wi-MAX, or any other suitable communication protocol.

[0073] The memory (220) may store data for at least one process (algorithm) for providing image processing services or a program that reproduces the process. In addition, the memory (220) may store additional processes for performing other operations, and is not limited thereto.

[0074] Meanwhile, the memory (220) can store various data used in the three-dimensional stress-strain curve prediction device (200), as well as pre-trained models built as needed.

[0075] In various embodiments, the memory (220) may include a volatile or non-volatile recording medium capable of storing various data, commands, and information. For example, the memory (220) may include at least one type of storage medium among a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (e.g., SD or XD memory, etc.), RAM, SRAM, ROM, EEPROM, PROM, network storage, cloud, and blockchain database.

[0076] In various embodiments, the memory (220) may store at least one configuration of an operating system (221), a communication module (222), a user interface module (223), and one or more applications (224).

[0077] An operating system (221) (e.g., embedded operating systems such as LINUX, UNIX, MAC OS, WINDOWS, VxWorks, etc.) may include various software components and drivers for controlling and managing general system operations (e.g., memory management, storage device control, power management, etc.) and may support communication between various hardware, firmware, and software components.

[0078] The communication module (222) can support communication with another device through the communication interface (210). The communication module (222) may include various software components for processing data received by the wired communication port (211) or wireless circuit (212) of the communication interface (210).

[0079] The user interface module (223) can receive requests or inputs from viewers from a keyboard, touch screen, microphone, etc., through the I / O interface (230) and provide a user interface on the display.

[0080] The application (224) may include a program or module configured to be executed by one or more processors (240).

[0081] The I / O interface (230) can connect at least one of an input / output device (not shown) of the service server (100), such as a display, keyboard, touch screen, and microphone, to the user interface module (223). The I / O interface (230) can receive user input (e.g., voice input, keyboard input, touch input, etc.) together with the user interface module (223) and process commands based on the received user input.

[0082] The processor (240) is connected to a communication interface (210), a memory (220), and an I / O interface (230) to control the overall operation of the three-dimensional stress-strain curve prediction device (200) and can execute various commands through an application or program stored in the memory (220).

[0083] The processor (240) may correspond to a computing device such as a CPU (Central Processing Unit) or an AP (Application Processor). Additionally, the processor (240) may be implemented in the form of an Integrated Chip (IC), such as a System on Chip (SoC) that integrates various computing devices. Alternatively, the processor (240) may include a module for computing artificial neural network models, such as a Neural Processing Unit (NPU).

[0084] Specifically, the processor (240) may be configured to input the previously acquired ultrasonic signal to a pre-trained artificial intelligence model that irradiates an ultrasonic signal onto a local area of ​​a test specimen and outputs a first parameter based on the input ultrasonic signal, and then input the first parameter to the Lamborgh-Osgood model when the first parameter calculated based on the acquired ultrasonic signal is output from the artificial intelligence model. Subsequently, when the second parameter calculated based on the first parameter is output from the Lamborgh-Osgood model, the processor (240) may be configured to scan the ultrasonic transceiver and repeat the above operations from the beginning for the next local area of ​​the test specimen, and when all second parameters for the entire area of ​​the test specimen are collected, visualize a three-dimensional stress-strain curve for the entire area of ​​the test specimen.

[0085] Thus, the processor (240) can transmit the generated three-dimensional stress-strain curve to at least one user terminal (300) to provide it to the user.

[0086] Here, the ultrasonic transceiver may include at least one of a contact-type ultrasonic transceiver or a non-contact-type ultrasonic transceiver. A contact-type ultrasonic transceiver refers to the use of contact ultrasonic sensors, such as piezoelectric elements. A non-contact ultrasonic transceiver refers to the use of non-contact ultrasonic sensors, such as lasers. When a contact-type ultrasonic transceiver is used, ultrasonic signal transmission and reception occur while the transceiver is in contact with the test specimen. In contrast, when a non-contact ultrasonic transceiver is used, ultrasonic signal transmission and reception occur while the transceiver is spaced apart from the test specimen by a certain distance.

[0087] That is, when measuring an ultrasonic signal, the processor (240) may be configured to measure the ultrasonic signal based on a contact ultrasonic measurement method or a non-contact ultrasonic measurement method.

[0088] Meanwhile, when measuring an ultrasonic signal, the processor (240) may be configured to simultaneously measure at least one of the following as at least one ultrasonic parameter: longitudinal wave velocity, attenuation coefficient, back-scattering coefficient, second and third-order acoustic nonlinearity parameters, shear wave velocity, shear wave attenuation coefficient, or shear wave back-scattering coefficient of the measured ultrasonic signal.

[0089] Here, at least one ultrasonic parameter may represent at least one component of the bulk wave, surface acoustic wave, Lamb wave, or wave mixing of the measured ultrasonic signal.

[0090] Additionally, the processor (240) can generate a three-dimensional stress-strain curve by configuring a new axis that adds an ultrasonic scanning direction to the stress and strain axes of the predicted two-dimensional stress-strain curve.

[0091] To this end, the processor (240) may build at least one artificial intelligence model by pre-training it. This at least one artificial intelligence model may be pre-trained based on a machine learning method or a deep learning method so as to non-destructively predict a stress-strain curve for a local region of a test specimen using at least one ultrasonic parameter.

[0092] The processor (240) may be configured to selectively execute at least one of at least one artificial intelligence model according to the frequency of the ultrasonic signal investigated through the ultrasonic transceiver.

[0093] Here, the machine learning method may include at least one of Random Forest, Linear / Nonlinear Regression, Support Vector Machines (SVM), K-Nearest Neighbors (KNN), or Decision Trees. In addition, the deep learning method may include at least one of a Multi-Layer Perceptron (MLP), Deep Neural Networks (DNN), Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM), Self-Supervised Learning, Semi-Supervised Learning, Unsupervised Learning, Transfer Learning, Domain Adaptation, Multi-modal based Neural Networks, or Fine-Tuning.

[0094] Meanwhile, the processor (240) may further construct and provide a Lamborgh-Osgood model in which the relationship between the input ultrasonic parameters and the ultrasonic signal is pre-learned. This Lamborgh-Osgood model may be a physical model capable of reconstructing a stress-strain curve with several ultrasonic parameters, and may reconstruct a stress-strain curve based on the following <Equation 1> to <Equation 4>.

[0095]

[0096]

[0097]

[0098]

[0099] Here, is strain, is stress, is the elastic modulus, and are the yield stresses near strains of 0.05% and 0.2%, respectively. is the stress near the ultimate tensile strength, Is Strain corresponding to, It can represent the strain hardening coefficient.

[0100] In particular, the value of the unknown in <Mathematical Equation 4> is , and There can be three, and the stress-strain curve can be reconstructed through <Equation 4>. At this time, instead It is acceptable to use other values ​​within a strain of 0.2%, as shown. Here, and Each value for can be preset and is not limited to.

[0101] The Lamborgh-Osgood model will be explained in detail below with reference to Fig. 7 while explaining Fig. 5.

[0102] FIG. 4 is a block diagram showing the configuration of a user terminal using a three-dimensional stress-strain curve prediction service using ultrasound according to one embodiment of the present disclosure.

[0103] Referring to FIG. 4, the user terminal (300) may include a memory interface (310), one or more processors (320) and a peripheral interface (330). Various components within the user terminal (300) may be connected by one or more communication buses or signal lines.

[0104] The memory interface (310) is connected to the memory (350) and can transmit various data to the processor (320). Here, the memory (350) may include at least one type of storage medium among flash memory type, hard disk type, multimedia card micro type, card type memory (e.g., SD or XD memory, etc.), RAM, SRAM, ROM, EEPROM, PROM, network storage, cloud, and blockchain database.

[0105] In various embodiments, the memory (350) may store a web / app application or program for receiving image processing services. Additionally, the memory (350) may not only store at least one image captured and acquired by an image sensor and information / data for each image, but may also store various information / data acquired through an application or program.

[0106] In various embodiments, memory (350) may store at least one of an operating system (351), a communication module (352), a graphical user interface module (GUI) (353), a sensor processing module (354), a telephone module (355), and an application module (356). Specifically, the operating system (351) may include instructions for processing basic system services and instructions for performing hardware operations. The communication module (352) may communicate with at least one of one or more other devices, computers, and servers. The graphical user interface module (GUI) (353) may process a graphical user interface. The sensor processing module (354) may process sensor-related functions (e.g., processing voice input received through one or more microphones (392)). The telephone module (355) may process telephone-related functions. The application module (356) may perform various functions of a user application, such as electronic messaging, web browsing, media processing, navigation, imaging, and other processing functions. In addition, the user terminal (300) can store one or more software applications (356-1, 356-2) (e.g., service applications) associated with any one type of service in the memory (350).

[0107] In various embodiments, the memory (350) can store a digital assistant client module (357) (hereinafter, DA client module) and accordingly store commands for performing client-side functions of the digital assistant and various user data (358) (e.g., user-customized vocabulary data, preference data, user's electronic address book, etc.).

[0108] Meanwhile, the DA client module (357) can obtain voice input, text input, touch input and / or gesture input from the user through various user interfaces (e.g., I / O subsystem (340)) provided in the user terminal (300).

[0109] Additionally, the DA client module (357) can output data in the form of audiovisual and tactile elements. For example, the DA client module (357) can output data consisting of a combination of at least two of voice, sound, notifications, text messages, menus, graphics, videos, animations, and vibrations. Furthermore, the DA client module (357) can communicate with a digital assistant server (not shown) using a communication subsystem (380).

[0110] In various embodiments, the DA client module (357) may collect additional information about the surrounding environment of the user terminal (300) from various sensors, subsystems, and peripheral devices to construct the context associated with the user input. For example, the DA client module (357) may provide context information along with the user input to a digital assistant server to infer the user's intent. Here, the context information that may accompany the user input may include sensor information, e.g., lighting, ambient noise, ambient temperature, images, videos, etc. of the surrounding environment. As another example, the context information may include the physical state of the user terminal (300) (e.g., device orientation, device location, device temperature, power level, speed, acceleration, motion pattern, cellular signal strength, etc.). As yet another example, the context information may include information related to the software state of the user terminal (300) (e.g., processes running on the user terminal (300), installed programs, past and present network activity, background services, error logs, resource usage, etc.).

[0111] In various embodiments, the memory (350) may include additional or deleted instructions. Furthermore, the user terminal (300) may include additional configurations in addition to the configuration shown in FIG. 4, or exclude some configurations.

[0112] The processor (320) can control the overall operation of the user terminal (300) and can execute various commands to use the prediction service provided by the service server (100) or the three-dimensional stress-strain curve prediction device (200) by running an application or program stored in memory (350).

[0113] The processor (320) may correspond to a computing device such as a CPU (Central Processing Unit) or an AP (Application Processor). Additionally, the processor (320) may be implemented in the form of an integrated chip (IC), such as a System on Chip (SoC) that integrates various computing devices that perform machine learning, such as a Neural Processing Unit (NPU).

[0114] In various embodiments, the processor (320) may receive various information and / or data about the user based on the platform through a user interface screen.

[0115] The peripheral interface (330) is connected to various sensors, subsystems, and peripheral devices and can provide data to enable the user terminal (300) to perform various functions. Here, it can be understood that the user terminal (300) performing a function is performed by the processor (320).

[0116] The peripheral interface (330) may receive data from a motion sensor (360), a light sensor (light sensor) (361), and a proximity sensor (362), thereby enabling the user terminal (300) to perform orientation, light, and proximity detection functions. As another example, the peripheral interface (330) may receive data from other sensors (363) (positioning system—GPS receiver, temperature sensor, biometric sensor), thereby enabling the user terminal (300) to perform functions related to the other sensors (363).

[0117] In various embodiments, the user terminal (300) may include a camera subsystem (370) connected to a peripheral interface (330) and an optical sensor (371) connected thereto, thereby enabling the user terminal (300) to perform various shooting functions such as taking photos and recording video clips.

[0118] In various embodiments, the user terminal (300) may include a communication subsystem (380) connected to a peripheral interface (330). The communication subsystem (380) is composed of one or more wired / wireless networks and may include various communication ports, radio frequency transceivers, and optical transceivers.

[0119] In various embodiments, the user terminal (300) includes an audio subsystem (390) connected to a peripheral interface (330), and the audio subsystem (390) includes one or more speakers (391) and one or more microphones (392), so that the user terminal (300) can perform voice-operable functions, such as voice recognition, voice cloning, digital recording, and telephone functions.

[0120] In various embodiments, the user terminal (300) may include an I / O subsystem (340) connected to a peripheral interface (330). For example, the I / O subsystem (340) may control a touch screen (343) included in the user terminal (300) through a touch screen controller (341).

[0121] For example, the touch screen controller (341) can detect user contact and movement or interruption of contact and movement using any one of a plurality of touch sensing technologies, such as capacitive, resistive, infrared, surface acoustic wave technology, proximity sensor array, etc. As another example, the I / O subsystem (340) can control other input / control devices (344) included in the user terminal (300) through other input controller(s) (342). As an example, other input controller(s) (342) can control one or more pointer devices such as buttons, rocker switches, thumb wheels, infrared ports, USB ports, and styluses.

[0122] FIG. 5 is a flowchart schematically illustrating a method for predicting a three-dimensional stress-strain curve using ultrasound according to an embodiment of the present disclosure, wherein a series of operations can be performed based on a platform built by a service server (100). Hereinafter, while describing each step of FIG. 5, specific operations and details will be described based on examples with reference to FIG. 6 to FIG. 8.

[0123] Referring to FIG. 5, the processor (240) measures the ultrasonic signal by irradiating an ultrasonic signal to a local area of ​​the test specimen (A) using an ultrasonic transceiver (S110).

[0124] As previously explained, the processor (240) may use a contact ultrasonic measurement method or a non-contact ultrasonic measurement method to measure the ultrasonic signal.

[0125] For example, as a contact ultrasonic measurement method, at least one of a pulse-echo method or a through-transmission method can be used, as shown in FIG. 6. FIG. 6 (a) represents the pulse-echo method, and (b) represents the through-transmission method.

[0126] In addition, as a non-contact ultrasonic measurement method, at least one of the following may be used: laser ultrasonics, electromagnetic acoustic transducer (EMAT), air coupled transducer (ACT), water immersion-based focusing method, or planar scanning acoustic microscopy (SAM) method. Here, to use the laser ultrasonics method, an ultrasonic generating unit and a measuring unit utilizing nanosecond, picosecond, femtosecond, or attosecond lasers may be provided, and a Michelson interferometer or Fabry-P Measurement units such as a rot interferometer, laser Doppler vibrometer, photorefractive interferometer, reflectometry, knife-edge, and beam distortion detection may also be included.

[0127] Next, the processor (240) inputs the ultrasonic signal measured by step S110 to a pre-trained artificial intelligence model that calculates and outputs a first parameter based on the input ultrasonic signal (S120).

[0128] Next, when the first parameter calculated from the artificial intelligence model is output by step S120, the processor (240) inputs the first parameter into the Lamborgh-Osgood model (S130).

[0129] At this time, the Lamborgh-Osgood model can reconstruct the stress-strain curve using a first parameter calculated based on at least one ultrasonic parameter.

[0130] Here, the first parameter may include at least one unknown value for the Lamborgh-Osgood model calculated using the at least one ultrasonic parameter. In this case, the modeling between the at least one unknown value and the at least one ultrasonic parameter can be obtained by performing learning (e.g., machine learning) as shown in FIG. 7. Here, in addition to the ultrasonic parameters, ultrasonic scan signals A, B (Brightness scan), and C (Cross-sectional scan) may also be used to calculate the at least one unknown value. Thus, the calculated at least one unknown value may be input back into the Lamborgh-Osgood model to reconstruct the stress-strain curve and produce a final output.

[0131] That is, the processor (240) can calculate at least one unknown value for the Lamborgh-Osgood model using only at least one ultrasonic parameter or ultrasonic signal.

[0132] However, without using machine learning, it is also possible to calculate at least one unknown value for the Lamborgh-Osgood model using a known physical model. For example, The value can be calculated using a relationship based on previously known ultrasonic longitudinal and transverse wave velocities, and me The value can be calculated using the previously known relationship for the ultrasonic nonlinear elastic modulus or by using the ultrasonic attenuation coefficient and the Hall-petch equation together.

[0133] Table 1 and Figure 8 below show examples of using regression analysis and CNN.

[0134] Value Ultrasonic velocity correlation coefficient hundreds of scattering equal damping correlation coefficient non-linear parameter correlation coefficient 2750-0.349-15.0-197 2540-0.313-16.5-212

[0135] Ultrasonic parameters such as the ultrasonic velocity used in above are merely examples, and the previously mentioned ultrasonic A, B, and C scan signals may also be used together.

[0136] The CNN hyperparameters used in Fig. 8 are merely examples, and other hyperparameter values ​​or model structures may be used. For example, although two CNN layers, or Conv layers, are used in Fig. 8, only one may be used.

[0137] Referring to FIG. 8, ultrasonic parameters are input to the input and may be composed of a CNN layer (Conv), an LSTM layer, a fully-connected layer, and a softmax layer. At this time, the hyperparameter numbers shown in FIG. 8 are merely examples and may be changed.

[0138] First, CNN layers are used to more effectively extract features related to stress-strain curves from the input ultrasonic signal. Since the ultrasonic signal is one-dimensional A-scan data, one-dimensional CNN layers are used. The number of CNN layers can be increased considering model training time and performance. Second, LSTM layers are used to more effectively extract features from time-series data such as ultrasonic signals.

[0139] Specifically, the CNN layer may include a first layer and a second layer.

[0140] The first layer may include a convolutional layer and a pooling layer. The convolutional layer performs convolution between the input data and a filter to extract features from the input data. The pooling layer subsamples, or pools, the features extracted through the convolutional layer. This is because it is not necessary to use all the features extracted through the convolutional layer. Since the input data is ultrasound A-scan data, the features extracted by the CNN layer will include the ultrasound velocity, attenuation, frequency, and non-linear components. According to the embodiment, the output of the first layer is 64. In addition, with respect to the first layer, the kernel size can be set to 5, the stride to 1, the zero padding to (4, 0), and the pooling to 2.

[0141] The second layer may include a convolutional layer and a pooling layer, just like the first layer. According to the embodiment, the output of the second layer is 64. And with respect to the second layer, the kernel size may be set to 5, the stride to 1, the zero padding to (4, 0), and the pooling to 2.

[0142] Additionally, LSTM layers can be deployed to prevent overfitting, such as dropout. The LSTM layer includes a hidden layer, and the size of the hidden layer can be set to 1×2×64. In this case, the output of the LSTM layer can be 1×64×64.

[0143] A fully connected layer is a layer that is connected to every neuron of the previous layer. The data output from the fully connected layer is input into the softmax layer.

[0144] The softmax layer applies the softmax function as an activation function to the input data to produce a target output.

[0145] Next, when the second parameter calculated based on the first parameter from the Lamborgh-Osgood model is output by the processor (240), the processor scans the ultrasonic transceiver and repeats steps S110 through S130 previously performed for the next local area of ​​the test specimen (A) (S140).

[0146] Here, scanning means moving the ultrasonic transceiver by a preset distance, and step S140 may involve the ultrasonic transceiver moving a certain distance by the preset distance while performing steps S110 to S130 for each local area.

[0147] As previously explained, the ultrasonic transceiver may be configured to be used as both a transmitter and a receiver by equipping a single ultrasonic sensor, or it may be configured to be used as a transmitter and a receiver, respectively, by equipping two ultrasonic sensors. In the latter case, the meaning of "scanning the ultrasonic transceiver described above" may be interpreted as moving the transmitter and receiver together, or as moving only one of them.

[0148] Next, when all second parameters for the entire area of ​​the test specimen (A) are collected, the processor (240) visualizes a three-dimensional stress-strain curve for the entire area of ​​the test specimen (A) (S150). That is, by sequentially repeating steps S110 through S130 for each local area within the entire area of ​​the test specimen (A), all second parameters for each local area can be collected.

[0149] In other words, in this S150 step, steps S110 through S140 are repeatedly applied to the entire area of ​​the test specimen, and a 3D stress-strain curve is generated by predicting a 2D stress-strain curve based on a second parameter for each predicted local area and constructing it in 3D. At this time, the 3D stress-strain curve can be composed of a new axis by adding an ultrasonic scanning direction to the stress and strain axes of the existing 2D stress-strain curve.

[0150] In the general ultrasonic measurement method described above in Fig. 6, ultrasonic scanning can be performed in combination with a 2-axis scanning stage operated by a mechanical motor, and in the case of a laser ultrasonic method, it can be performed in combination with a laser scanning system using a galvano mirror.

[0151] FIG. 9 is a drawing showing an example of comparing the predicted stress-strain curves and the error between the stress-strain curves for each of a material with high stiffness and a material with low stiffness, according to one embodiment of the present disclosure.

[0152] The graph (20) shown in FIG. 9 represents a stress-strain curve within 0.2% strain of a material with high stiffness (hereinafter referred to as the first stress-strain curve) (21) and a stress-strain curve within 0.2% strain of a material with low stiffness (hereinafter referred to as the second stress-strain curve) (22).

[0153] Referring to FIG. 9, when linearity is high (i.e., non-linearity is low), such as in the first stress-strain curve (21), it is preferable to use a value of less than 0.1% (e.g., 0.05%). Also, when linearity is low (i.e., non-linearity is high), such as in the second stress-strain curve (22), it is preferable to use a value of 0.1% or more (e.g., 0.15%).

[0154] Meanwhile, if it is not known whether the stiffness of the material is large or small, it is best to use a value of 0.1%. Here, the meaning of "best" is that when a stress-strain curve is predicted by performing a tensile test according to the present disclosure, the error between the predicted stress-strain curve and the stress-strain curve obtained by performing a tensile test according to a comparative example is the smallest.

[0155] FIG. 10 is a drawing for explaining examples of stress-strain curves visualized through a three-dimensional stress-strain curve prediction method using ultrasound according to one embodiment of the present disclosure, and shows examples applied to aluminum materials.

[0156] Referring to FIG. 10, for each of the three test specimens (a, (a) to (c)) obtained by performing a tensile test according to the present disclosure as in (g), three-dimensional stress-strain surfaces (each (a) to (c)) and two-dimensional stress-strain curves (each (d) to (f)) obtained by performing a tensile test according to a comparative example are shown.

[0157] In particular, when comparing (c) and (f) for specimens with severe damage, it can be seen that the damage is visibly visible in the three-dimensional stress-strain curve, but not in the two-dimensional stress-strain curve.

[0158] Accordingly, according to the present disclosure, tensile testing can be performed in a short time using ultrasonic signals without the need to extract a test specimen (inspection body) from a part, and the cost or time required for manufacturing and inspection can be reduced by enabling the reuse of the test specimen based on a non-destructive method and allowing in-line inspection during the process. In addition, according to the present disclosure, it is possible to predict a three-dimensional stress-strain curve that is applicable to localized damage that cannot be detected by conventional two-dimensional stress-strain curves or to micro-components such as semiconductors.

[0159] Although embodiments of the present disclosure have been described in more detail with reference to the attached drawings, the present disclosure is not necessarily limited to these embodiments and may be modified in various ways within the scope of the technical spirit of the present disclosure. Accordingly, the embodiments disclosed in the present disclosure are intended to explain, not limit, the technical spirit of the present disclosure, and the scope of the technical spirit of the present disclosure is not limited by these embodiments. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. The scope of protection of the present disclosure shall be interpreted by the claims below, and all technical spirits within an equivalent scope shall be interpreted as being included within the scope of rights of the present disclosure.

[0160] [National R&D projects that supported this invention]

[0161] [Project ID] 2710005537

[0162] [Assignment No.] 00357748

[0163] [Ministry Name] Ministry of Science and ICT

[0164] [Name of Project Management (Specialized) Agency] National Research Foundation of Korea

[0165] [Research Project Name] Individual Basic Research (Ministry of Science and ICT)

[0166] [Project Title] Ultrashort Laser-Based Tungsten Alloy Additive Manufacturing and Inspection Technology

[0167] [Name of Project Performing Organization] Soongsil University Industry-Academic Cooperation Foundation

[0168] [Research Period] April 1, 2024 ~ March 31, 2025

Claims

1. A method for predicting a three-dimensional stress-strain curve using ultrasound, performed by a device, A step of measuring an ultrasonic signal by irradiating an ultrasonic signal onto a local area of ​​a test specimen using an ultrasonic transmitter / receiver; A step of inputting the measured ultrasonic signal into a pre-trained artificial intelligence model that calculates and outputs a first parameter based on the input ultrasonic signal; When a first parameter calculated based on the measured ultrasonic signal is output from the artificial intelligence model, a step of inputting the first parameter into the Lamborgh-Osgood model; When a second parameter calculated based on the first parameter from the Lamborgh-Osgood model is output, a step of scanning the ultrasonic transceiver to repeat each of the above steps for the next local region of the test specimen; and When all second parameters for the entire area of ​​the test specimen are collected, the method includes the step of visualizing a three-dimensional stress-strain curve for the entire area of ​​the test specimen. 3D stress-strain curve prediction method using ultrasound.

2. In Paragraph 1, The step of measuring the above ultrasonic signal is, At least one of the following is simultaneously measured as at least one ultrasonic parameter: longitudinal wave velocity, attenuation coefficient, back-scattering coefficient, second and third-order acoustic nonlinearity parameters, shear wave velocity, shear wave attenuation coefficient, or back-scattering coefficient of the measured ultrasonic signal. 3D stress-strain curve prediction method using ultrasound.

3. In Paragraph 2, The above at least one ultrasonic parameter is, Representing at least one component of a bulk wave, surface acoustic wave, Lamb wave, or wave mixing of the measured ultrasonic signal, 3D stress-strain curve prediction method using ultrasound.

4. In Paragraph 2, The above artificial intelligence model is, Pre-trained based on a machine learning or deep learning method to non-destructively predict a stress-strain curve for a local region of the test specimen using at least one ultrasonic parameter, 3D stress-strain curve prediction method using ultrasound.

5. In Paragraph 2, The above first parameter is, At least one unknown value for the Lamborgh-Osgood model calculated using the at least one ultrasonic parameter, 3D stress-strain curve prediction method using ultrasound.

6. In Paragraph 2, The above second parameter is, Strain ( ), Strain hardening modulus ( ), stress near ultimate tensile strength ( ) and strain corresponding to stress near the ultimate tensile strength ( including at least one of ), 3D stress-strain curve prediction method using ultrasound.

7. In Paragraph 5, The above at least one unknown value is, It includes at least one of an elastic modulus or a yield stress near each preset strain, The above elastic modulus is, It is calculated using a relationship based on ultrasonic longitudinal and transverse wave velocities, and The yield stress near each of the above deformation rates is, Calculated using the relationship of the ultrasonic nonlinear elastic modulus, or calculated using the ultrasonic attenuation coefficient and the Hall-petch equation together, 3D stress-strain curve prediction method using ultrasound.

8. In Paragraph 1, The step of measuring the above ultrasonic signal is, Measuring the ultrasonic signal based on a contact ultrasonic measurement method or a non-contact ultrasonic measurement method, 3D stress-strain curve prediction method using ultrasound.

9. In Paragraph 1, The above three-dimensional stress-strain surface is, It consists of a new axis formed by adding the ultrasonic scanning direction to the stress and strain axes of the predicted two-dimensional stress-strain curve, 3D stress-strain curve prediction method using ultrasound.

10. Communication interface; Memory; and A processor operably connected to the above communication interface and the above memory, comprising: The above processor is, Ultrasonic signals are measured by irradiating a local area of ​​a test specimen with an ultrasonic transmitter and receiver, and The measured ultrasonic signal is input into an artificial intelligence model that has been pre-trained to calculate and output a first parameter based on the input ultrasonic signal, and When a first parameter calculated based on the measured ultrasonic signal is output from the artificial intelligence model, the first parameter is input into the Lamborgh-Osgood model, and when a second parameter calculated based on the first parameter is output from the Lamborgh-Osgood model, the ultrasonic transceiver is scanned to repeat each of the above operations for the next local region of the test specimen. When all second parameters for the entire area of ​​the test specimen are collected, configured to visualize the three-dimensional tensile surface for the entire area of ​​the test specimen, 3D stress-strain curve prediction device using ultrasound.

Images