Steel belt nondestructive testing method and medium

Abstract

The invention discloses a steel belt nondestructive testing method and a medium. The method comprises the following steps: inputting identification information and attribute information of a to-be-detected steel belt; performing scanning detection on the to-be-detected steel strip to obtain detection data corresponding to a plurality of detection positions; analyzing the detection data to correspondingly obtain characteristic data of the to-be-detected steel strip; and outputting the characteristic data of the steel strip to be detected. According to the method, detection can be carried out without damaging the steel belt, multiple different positions of the steel belt can be detected, the detection comprehensiveness is improved, the analysis model is constructed between the detection data and the output characteristic data, the detection accuracy and the detection efficiency can be improved, the identification information of the steel belt is directly associated and output, and the detection accuracy is improved. And the automation degree of industrial production detection is improved.

Description

Technical Field

[0001] This application relates to the field of steel strip testing technology, and in particular to a non-destructive testing method and medium for steel strips. Background Technology

[0002] Currently, domestic steel companies primarily rely on tensile testing in physicochemical laboratories to obtain performance parameters for the mechanical testing of thin steel products such as strip steel and plate steel. The main drawbacks of this existing method are: 1. Large data lag: Physicochemical measurement values ​​are only available a day after production is completed. 2. Incomplete data: Only the mechanical properties of the initial and final portions of a single coil of strip are available, failing to capture the overall quality of the entire coil. 3. Wasteful shearing: For example, during production, if the machine stops or operates at low speed, a section of "suspected substandard" strip needs to be sheared. How much to cut is unknown, leading to the common practice of cutting as much as possible, resulting in waste. 4. Requires 24-hour human intervention, resulting in low efficiency and contradicting the trends of unmanned and intelligent manufacturing.

[0003] Therefore, it is necessary to improve the efficiency, comprehensiveness, and accuracy of steel strip inspection.

[0004] Application content The main technical problem addressed by this application is to provide a non-destructive testing method and medium for steel strips, thereby solving the problems of improving the efficiency and accuracy of steel strip testing in existing technologies.

[0005] To address the aforementioned technical problems, this application provides a non-destructive testing method for steel strips, comprising the following steps: First, inputting the identification and attribute information of the steel strip to be tested; Second, scanning and probing the steel strip to be tested to obtain detection data corresponding to multiple detection locations; Third, analyzing the detection data to obtain characteristic data of the steel strip to be tested; Fourth, outputting the characteristic data of the steel strip to be tested.

[0006] In some embodiments, in the first step, the identification information of the steel strip to be tested includes the grade, coil number, supplier information, and / or smelting number of the steel strip to be tested, and the attribute information of the steel strip to be tested includes model information, specification information, material information, and / or application information.

[0007] In some embodiments, in the second step, scanning and detecting the steel strip to be detected includes using a detection probe to locally excite the steel strip, acquiring electromagnetic characteristic signals, then amplifying and filtering the signals, and then transmitting them to an electromagnetic detection device via a cable.

[0008] In some embodiments, in the second step, acquiring detection data corresponding to multiple detection locations includes pre-determining predetermined detection locations of the steel strip to be detected based on the identification information and attribute information of the steel strip to be detected.

[0009] In some embodiments, in the second step, when there is a large amount of detection data, the detection data is further filtered. The filtering method includes: prioritizing the performance of test data at different locations of the steel strip as follows: priority of test data at the middle of the steel strip > priority of test data at the tail of the steel strip > priority of test data at the head of the steel strip; or / and, priority of test data obtained at the latest test time > priority of test data obtained at previous test times. In some embodiments, the method for analyzing the detection data in the third step includes stepwise regression or artificial neural network methods.

[0010] In some embodiments, the stepwise regression method includes: Step S301: First, acquire multiple detection data for the steel strip; Step S302: Obtain the actual mechanical property data of the steel strip through testing and analysis; Step S303: Using the detection data as the input independent variable and the mechanical characteristic data as the output dependent variable, combine these data to construct one or more regression calculation formulas; Step S304: Test the significance of the regression coefficients in each regression formula, delete independent variables with insignificant coefficients, recombine the independent variables, and then return to step S303 for iterative testing; Step S305: If the significance of the regression coefficient in the regression calculation formula meets the test criteria, then retain the corresponding independent variable and return to step S304 to continue testing the significance of the regression coefficients of other independent variables that have not yet been tested, until all independent variables have been tested. Step S306: Finally, the desired optimal regression formula is obtained.

[0011] In some embodiments, the neural network method includes: Step S401: First, obtain multiple test data for the steel strip; Step S402: Obtain the actual mechanical property data of the steel strip through testing and analysis; Step S403: Construct a neural network model, using the detection data as the input independent variable of the neural network model and the mechanical characteristic data as the output dependent variable of the neural network model; Step S404: Initialize the weights and thresholds corresponding to each layer of the neural network model; Step S405: Using the existing detection data and mechanical characteristic data as sample data, input them into the neural network model, calculate the output layer error and hidden layer error in the neural network model, and correct the output layer error and the weights and thresholds of the hidden layer until all sample data have completed training; Step S406: Determine whether the trained neural network model meets the conditions for training completion. If it does, the training is completed, and the neural network model is used for practical applications. If it does not meet the conditions, return to step S404 to rebuild and train again until a neural network model that can be used for practical applications is obtained.

[0012] In some embodiments, the output characteristic data of the steel strip to be tested includes: longitudinal and transverse tensile strength, longitudinal and transverse yield strength, longitudinal and transverse elongation after fracture, Vickers hardness and / or confidence interval.

[0013] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the aforementioned method.

[0014] The beneficial effects of this application are as follows: This application discloses a non-destructive testing method and medium for steel strips. The method includes: inputting the identification information and attribute information of the steel strip to be tested; scanning and probing the steel strip to be tested to obtain detection data corresponding to multiple detection positions; analyzing the detection data to obtain characteristic data of the steel strip to be tested; and outputting the characteristic data of the steel strip to be tested. This method does not require damaging the steel strip for testing, and can detect multiple different positions of the steel strip, improving the comprehensiveness of the detection. By constructing an analysis model between the detection data and the output characteristic data, it is beneficial to improve the accuracy and efficiency of the detection. Directly associating the output with the identification information of the steel strip improves the automation level of industrial production testing. Attached Figure Description

[0015] Figure 1 This is a flowchart of an embodiment of the nondestructive testing method for steel strips in this application; Figure 2 This is a flowchart of an embodiment of the stepwise regression analysis method in the nondestructive testing method for steel strips in this application; Figure 3 This is a flowchart of an embodiment of the neural network analysis method in the nondestructive testing method for steel strips in this application; Figure 4 This is a schematic diagram of an embodiment of a computer-readable storage medium in this application; Figure 5 This is a schematic diagram illustrating the composition of an embodiment of the nondestructive testing system for steel strips in this application; Figure 6This is a schematic diagram illustrating the composition of an embodiment of the nondestructive testing system for steel strips in this application; Figure 7 This is a schematic diagram of the robotic arm control components in an embodiment of the non-destructive testing system for steel strips in this application; Figure 8 This is a schematic diagram of the internal composition of each unit in the embodiment of the nondestructive testing system for steel strip in this application; Figure 9 This is a schematic diagram of the switch connection composition in an embodiment of the non-destructive testing system for steel strip in this application; Figure 10 This is a schematic diagram of the main control interface in an embodiment of the non-destructive testing system for steel strip in this application; Figure 11 This is a schematic diagram of the detector circuit composition in an embodiment of the nondestructive testing system for steel strip in this application; Figure 12 This is a schematic diagram of the single-channel detection data acquisition process in the embodiment of the non-destructive testing system for steel strip in this application. Detailed Implementation

[0016] To facilitate understanding of this application, a more detailed description is provided below with reference to the accompanying drawings and specific embodiments. Preferred embodiments of this application are shown in the drawings. However, this application can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the disclosure of this application.

[0017] It should be noted that, unless otherwise defined, all technical and scientific terms used in this specification have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of this application. The term "and / or" as used in this specification includes any and all combinations of one or more of the associated listed items.

[0018] The embodiments will now be described in detail with reference to the accompanying drawings.

[0019] Figure 1 A flowchart of an embodiment of the nondestructive testing method for steel strips in this application is provided, including the following steps: S1: First step, input the identification information and attribute information of the steel strip to be detected; S2: The second step is to scan and detect the steel strip to be detected, and obtain detection data corresponding to multiple detection positions; S3: The third step is to analyze the detection data to obtain the characteristic data of the steel strip to be tested. S4: Fourth step, output the characteristic data of the steel strip to be tested.

[0020] First, it should be stated that the steel strip in this application can also be applied to the testing of other metal sheets, such as aluminum strips, copper strips, etc. Furthermore, in addition to coilable steel strips, non-coilable steel sheets can also be included within the scope of this application.

[0021] In the first step S1, the identification information of the steel strip to be tested includes the grade, coil number, supplier information, and smelting number of the steel strip to be tested. The attribute information of the steel strip to be tested includes model information, specifications such as length, width, and thickness, material information, and application information.

[0022] The identification and attribute information of the steel strip to be tested can be obtained by calling the ERP software corresponding to the steel strip to be tested and entering the corresponding testing system. For example, by scanning the identification QR code of the steel strip to be tested, the database in the ERP software can be queried to obtain the identification and attribute information of the steel strip to be tested. In this way, this information can be digitally entered into the testing system.

[0023] In the second step S2, the steel strip to be tested is scanned and detected, including using a detection probe to locally excite the steel strip, obtain electromagnetic characteristic signals, then amplify and filter the signals, and then transmit them to the electromagnetic detection equipment via a cable.

[0024] Therefore, the predetermined detection positions of the steel strip to be detected can be determined in advance based on the identification and attribute information of the steel strip to be detected. These positions include multiple positions such as the head, tail, and middle of the steel strip. Then, specific detection position points can be selected. These position points can be randomly selected through a predetermined coordinate system and the spatial three-dimensional model of the steel strip to be detected. Alternatively, to enhance the comparability of detection between steel products in the same batch, the same detection positions can be determined.

[0025] Therefore, during the transmission and transportation of steel belts in roller conveyor equipment, it is necessary to detect the running speed and running position of the steel belt. This allows for the real-time acquisition of the detected position information of the steel belt, which can be directly mapped and associated with the detection data and directly output.

[0026] Depending on the detection method used by the probe, there are various types of detection data, including Barkhausen noise detection data and multi-frequency eddy current detection data. Tables 1 and 2 below provide explanations of the meanings of the characteristic parameters corresponding to the detection data.

[0027] Table 1. Characteristic parameters of Barkhausen noise

[0028] Table 2 Characteristic Parameters of Multi-Frequency Eddy Current

[0029] Furthermore, in the second step S2, when there is a large amount of detection data, the detection data can be further filtered. Filtering methods include: Method 1: The performance priority of test data at different locations of the steel strip is ranked as follows: the priority of test data at the middle of the steel strip > the priority of test data at the tail of the steel strip > the priority of test data at the head of the steel strip. That is, the priority of test data at the middle of the steel strip is the highest, followed by the priority of test data at the tail of the steel strip, and finally the priority of test data at the head of the steel strip is the lowest.

[0030] Method 2: The priority of detection data obtained at the latest test time is > the priority of detection data obtained at previous test times.

[0031] Furthermore, location information takes precedence over time information; that is, the detection data at the optimal location is found first, and if the locations are the same, the detection data at the latest time is found.

[0032] In the third step S3, the methods for analyzing the detection data include stepwise regression or artificial neural network methods for modeling. These methods analyze the input detection data and then output the characteristic data of the steel strip to be tested. This characteristic data includes various mechanical property parameters of the steel strip, such as hardness Hv, yield strength Rp, tensile strength Rm, and elongation A.

[0033] Taking yield strength RP as an example, if multiple test data obtained from electromagnetic testing are Mk (k=1,2,…,i,…41), then RP=f(M1,M2,…,Mk,…M41), where f is the mapping between the multiple test data Mk and RP obtained by the modeling analysis method. There are various methods to obtain f, such as stepwise regression or artificial neural network methods for modeling analysis. The accuracy of the modeling analysis can be judged by the difference between the value obtained from physicochemical experiments (true value) and the estimated value calculated by the constructed model.

[0034] Furthermore, in this application, for the stepwise regression method, such as Figure 2 As shown, it includes: Step S301: First, obtain multiple test data for the steel strip; Step S302: Obtain the actual mechanical property data of the steel strip through testing and analysis; Step S303: Using the detection data as the input independent variable and the mechanical characteristic data as the output dependent variable, combine these data to construct one or more regression calculation formulas; Step S304: Test the significance of the regression coefficients in each regression formula, delete independent variables with insignificant coefficients, recombine the independent variables, and then return to step S303 for iterative testing; Step S305: If the significance of the regression coefficient in the regression calculation formula meets the test criteria, then retain the corresponding independent variable and return to step S304 to continue testing the significance of the regression coefficients of other independent variables that have not yet been tested, until all independent variables have been tested. Step S306: Finally, the desired optimal regression formula is obtained.

[0035] Furthermore, in this application, for the neural network method, such as Figure 3 As shown: Step S401: First, obtain multiple test data for the steel strip; Step S402: Obtain the actual mechanical property data of the steel strip through testing and analysis; Step S403: Construct a neural network model, using the detection data as the input independent variable of the neural network model and the mechanical characteristic data as the output dependent variable of the neural network model; Step S404: Initialize the weights and thresholds corresponding to each layer of the neural network model; Step S405: Using the existing detection data and mechanical characteristic data as sample data, input them into the neural network model, calculate the output layer error and hidden layer error in the neural network model, and correct the output layer error and the weights and thresholds of the hidden layer until all sample data have completed training; Step S406: Determine whether the trained neural network model meets the conditions for training completion. If it does, the training is completed, and the neural network model is used for practical applications. If it does not meet the conditions, return to step S404 to rebuild and train again until a neural network model that can be used for practical applications is obtained.

[0036] If the actual output Ym matches the expected output Om, training ends; if they don't, backpropagation begins. During backpropagation, the total error E between the actual output Ym and the expected output Om is calculated and distributed across all layers and nodes, adjusting the weights and thresholds of each layer and node. The network with adjusted weights and thresholds is then used as a new model for retraining. If the trained result meets the expected requirements, training ends; otherwise, backpropagation resumes. This process is repeated until the preset number of iterations is reached or the performance function value is less than the preset error accuracy, at which point training stops.

[0037] Furthermore, in Figure 1In step S4, the output characteristic data of the steel strip to be tested includes: longitudinal and transverse tensile strength, longitudinal and transverse yield strength RP0.2, longitudinal and transverse yield strength RP1.0, longitudinal and transverse elongation after fracture, Vickers hardness, and confidence interval. It may also include test time, coil number, test procedure, and head, middle, and tail position information, which are further correlated with these characteristic data of the steel strip.

[0038] Table 3. Detection Output Information for Steel Strip

[0039] Based on the inventive concept of the above embodiments, this application also provides a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, performs the steps of the method described in any of the above embodiments. The following is in conjunction with... Figure 4 This describes the execution process of the above embodiments on a computer-readable storage medium.

[0040] like Figure 4 As shown, it illustrates the computer-readable storage medium of this application. If the aforementioned synchronization method is implemented as a software functional unit and sold or used as an independent product, it can be stored in the computer-readable storage medium 200. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions / computer programs to cause an Internet of Things device (which may be a personal computer, server, or network terminal, etc.) or processor to execute all or part of the steps of the methods of various embodiments of this application. The aforementioned storage medium includes various media such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks, as well as electronic terminals such as computers, mobile phones, laptops, tablets, cameras, and dedicated devices that have the aforementioned storage media.

[0041] The execution process of program data in a computer-readable storage medium can be described with reference to the above-described method embodiments of this application, and will not be repeated here.

[0042] Therefore, this application discloses a non-destructive testing method and medium for steel strips. The method includes: inputting the identification information and attribute information of the steel strip to be tested; scanning and probing the steel strip to be tested to obtain detection data corresponding to multiple detection positions; analyzing the detection data to obtain characteristic data of the steel strip to be tested; and outputting the characteristic data of the steel strip to be tested. This method does not require damaging the steel strip for testing, and can detect multiple different positions of the steel strip, improving the comprehensiveness of the detection. By constructing an analysis model between the detection data and the output characteristic data, it is beneficial to improve the accuracy and efficiency of the detection. Directly associating the output with the identification information of the steel strip improves the automation level of industrial production testing.

[0043] Furthermore, based on the same inventive concept, this application also provides an embodiment of a non-destructive testing system for steel strips, such as... Figure 5 and Figure 6 As shown, it includes: an electromagnetic detection unit 1, a robotic arm unit 2, a steel belt transmission unit 3, and a control unit 4; wherein, the electromagnetic detection unit includes an electromagnetic detection probe, which is used to locally excite the steel belt, acquire electromagnetic characteristic signals, then amplify and filter the signals, and then transmit them to the detection controller through a cable; and also includes power distribution facilities 5.

[0044] Specifically, the robotic arm unit includes a robotic arm fixing platform and a six-axis robotic arm. An electromagnetic detection probe is set at the end of the six-axis robotic arm. The robotic arm fixing platform is used to fix the six-axis robotic arm. The six-axis robotic arm drives the electromagnetic detection probe to move up and down and laterally. The steel strip conveying unit includes a roller device for conveying the steel strip and a roller encoder, wherein the roller encoder is used to detect the conveying speed and position of the steel strip; The control unit includes a control host, a robotic arm controller and a detection controller that are communicatively connected to the control host. The robotic arm controller is used to control the operation of the six-axis robotic arm, and the detection controller is used to control the detection of the electromagnetic probe and the transmission operation of the steel belt in the steel belt transmission unit.

[0045] Furthermore, the robotic arm fixing platform in the robotic arm unit includes a base and a support frame. Its main function is to fix the robotic arm, and it adopts a bolt-type fixing structure to avoid damaging the ground environment.

[0046] exist Figure 5 and Figure 6In this application scenario, the control host is located in control room 5 and consists of a display interface and a processing host. The robotic arm is installed next to the roller equipment, with a separate support column for fixation, thus not occupying the maintenance space of the roller equipment, avoiding interference with other systems, and facilitating maintenance. The detection equipment is clamped and connected by the end of the robotic arm, and can rotate to detect the steel strip to be tested at different angles. The system wiring is routed underground, following general wiring specifications for automated equipment, prioritizing safety and ease of maintenance. The robotic arm controller and the detection controller are integrated in the electrical control cabinet, located in the side area of ​​the roller equipment.

[0047] Specifically, for the six-axis robotic arm unit, the control probe can move forward and backward, flip left and right, lift up and down, and rotate, achieving three-dimensional spatial motion to adapt to different working conditions. Combined with a motion control system, it precisely positions the steel strip and adjusts the lifting height in real time for plates of different thicknesses to ensure the accuracy of the inspection data. Specifically, the robotic arm can drive the probe to lift and move laterally, ensuring the distance between the probe and the steel strip surface is 5±0.5mm. Simultaneously, the probe's lifting function protects it when production stops. The probe's lateral movement ensures it remains in the correct position for picking up and inspecting steel strips of different widths.

[0048] This six-axis robotic arm consists of a base, an arm, and a hand. The base enables the robotic arm to move horizontally left and right, while the arm and hand work together to enable the robotic arm to move vertically and horizontally forward and backward. The hand is connected to an electromagnetic testing device, which can perform performance testing on stainless steel strips of different heights and inclinations.

[0049] This six-axis robotic arm has six degrees of freedom: rotation, pitch, yaw, and base rotation. Based on the weight of the testing equipment and the on-site operating environment, the robotic arm used has an effective load of 5 kg and a working radius of 954 mm.

[0050] Table 4: Structural Parameters of a Six-Axis Robotic Arm

[0051] In the field of application, the robotic arm needs to achieve a large pitch angle to achieve a contact effect with the detection plate surface. The corresponding axis motion range parameters are as follows: Table 5: Robotic Arm Parameters

[0052] like Figure 7As shown, the control of a six-axis robotic arm includes an upper-level human-machine interface and control software, namely the ROS (Robot Operating System), the robotic arm controller, and motion execution units. The control software performs motion planning and collision detection, and can also add voice processing and image processing functions. The data interface transmits control signals to the execution unit motors in the robotic arm, the motors execute the actions, and the motion trajectory is fed back in real time.

[0053] Combination Figure 8 As shown, for the control unit, the host in the control room is connected to the electrical cabinet via Ethernet. The detection equipment (i.e., the detection controller) inside the electrical cabinet is electrically connected to the electromagnetic detection probe, encoder, barcode scanner, and laser rangefinder on site. The electromagnetic detection probe is used to obtain electromagnetic detection data on the surface of the steel strip. The encoder is used to detect the running speed and position of the steel strip, thereby obtaining the accurate location for detection. The barcode scanner scans the product number on the steel strip to obtain the product information, which facilitates direct correlation between the product information and the detection results. The laser rangefinder measures the distance between the electromagnetic detection probe and the steel strip.

[0054] Furthermore, electromagnetic detection data obtained from electromagnetic detection equipment is further processed through model identification, such as... Figures 1 to 3 The illustrated embodiment completes the output of test performance data to the host computer. The electromagnetic testing equipment comprehensively utilizes various electromagnetic non-destructive testing technologies, including Barkhausen and multi-frequency eddy current methods, achieving good results in testing the properties of ferromagnetic materials and conductors through sample testing. The electromagnetic testing equipment can identify changes in material conditions and determine its mechanical properties, such as hardness, yield strength, tensile strength, and residual stress. The testing probe can be contact or non-contact, and the testing procedure is fully automated, allowing for integration into the manufacturing process.

[0055] exist Figure 8 The control unit also includes components for controlling the robotic arm. Specifically, an emergency stop switch and start / stop buttons are installed in the control room, and the robotic arm control module (i.e., robotic arm controller) is connected to the electrical cabinet. This module is electrically connected to the six-axis robotic arm on site to realize the electrical control of the robotic arm. In addition, the robotic arm control module is also connected to a switch via a PLC module. The switch receives video images captured by the camera on site and then displays the robotic arm operation on the corresponding monitor of the industrial control computer.

[0056] The switch is also connected to the electromagnetic detection equipment network, enabling the linkage between probe detection and probe operation control. The PLC module controls the operation of the probe movement device, communicating with the human-machine interface software via TCP / IP to achieve software control of the robotic arm's feeding mechanism.

[0057] Figure 9 This further demonstrates that the system interconnects with the detection and acquisition industrial control computer (controlling the detection probe), the upper-level monitoring host located in the control room, network cameras, robotic arm controllers, and other systems through a switch, forming a corresponding communication network architecture. This lays the infrastructure foundation for achieving coordinated and precise control of the various components of the system.

[0058] In addition, the host computer also includes a server and human-machine interface software. The server includes a database for data storage, and the human-machine interface software includes a front-end interface and a back-end program. The back-end program includes communication with various on-site modules, database development, and other external data communication. The front-end interface adopts a visual design and has functions such as remote access monitoring, user permission settings, dynamic monitoring of indicators, and video playback.

[0059] Specifically, such as Figure 10 The human-machine interface shown includes a steel strip information input display area P1, a robotic arm control area P2, a detection data curve display area P3, and a mechanical performance display area P4. It is evident that this interface enables the control and display of the electromagnetic detection unit 1, robotic arm unit 2, steel strip transmission unit 3, and control unit 4, facilitating integrated and unified control management, and the overall display and output of detection data.

[0060] like Figure 11 As shown, the circuit of the detection controller in the electrical cabinet consists of a base plate and a multi-channel plug-in board. The base plate is responsible for controlling, powering, and acquiring data from the multi-channel devices, as well as communicating with the upper-level monitoring host. Each plug-in board corresponds to one channel device, and each channel device is connected to a coil for data acquisition.

[0061] In multi-channel I2C communication designs, repetitive communication operations result in high time redundancy. Using ten signal lines to drive ten data lines separately would severely limit the sampling rate. This system adopts a single clock line design, with one clock signal controlling ten slave devices. This effectively reduces I2C communication time, saves component costs, and improves efficiency and sampling rate.

[0062] The modular design using plug-in boards facilitates individual debugging of each module and allows for the selection of different channels to form a probe system based on the required number of probe channels, thus ensuring the versatility of the baseboard to a certain extent.

[0063] An AD5933 impedance measurement chip is mounted on the board to realize eddy current detection. This chip uses an on-chip integrated frequency generator and a 12-bit, 1 MSPS analog-to-digital converter (ADC). The signal generated by the frequency generator excites an external complex impedance. The response signal of the external impedance is sampled by the on-chip ADC and then processed by the on-chip DSP using a Discrete Fourier Transform (DFT). The DFT algorithm returns a real part data word and an imaginary part data word at each frequency. The register contents can be read into the microcontroller via serial I2C.

[0064] Specifically, the detection controller's acquisition mode sends 44-bit frame data every 10ms, including: frame header, frame length, data, and checksum. CRC8 is used to verify data correctness. A maximum of 10 channels of data can be sent. The data includes both imaginary and real parts. Features also include: controlling the chip to repeatedly acquire data, adding a watchdog timer to prevent program crashes, and the microcontroller communicating with the AD5933 via the I2C protocol. The microcontroller controls the AD5933 chip's impedance data measurement and sends the acquired AD5933 impedance data to the host via serial communication. Figure 12 The process of transmitting data via I2C protocol based on serial port is further demonstrated.

[0065] The following table (Table 6) presents eddy current testing data from 8 channels. Each channel collects data with a complex structure, including one real part and one imaginary part. Regression or neural network analysis, as described above, was then performed on these collected test data to obtain the mechanical property values ​​for each channel, including: yield strength Rp, tensile strength Rm, hardness Hv, and elongation A. The creation time of the test sampling data and the analysis data is also provided.

[0066] Table 6: 8-channel eddy current detection and processing information

[0067] This application discloses a non-destructive testing system for steel strip, comprising an integrated setup: an electromagnetic detection unit, a robotic arm unit, a steel strip transmission unit, and a control unit. The electromagnetic detection unit includes an electromagnetic detection probe that locally excites the steel strip, acquires electromagnetic characteristic signals, and transmits them to the detection controller via cable. The robotic arm unit includes a fixed platform and a six-axis robotic arm. The electromagnetic detection probe is located at the end of the robotic arm, which drives the probe to move vertically and laterally. The steel strip transmission unit includes rollers for transmitting the steel strip and roller encoders for detecting the transmission speed and position of the steel strip. The control unit includes a control host and a robotic arm controller and a detection controller that are communicatively connected to the control host. This system enables real-time non-destructive testing of the transmitting steel strip, offering high efficiency and accuracy. The system is easy to deploy and improves the automation level of steel strip production testing.

[0068] The above are merely embodiments of this application and do not limit the scope of this patent application. Any equivalent structural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of this application.

Claims

1. A non-destructive testing method for steel strips, characterized in that, Including the following steps: The first step is to input the identification and attribute information of the steel strip to be tested; The second step is to scan and detect the steel strip to be detected, and obtain detection data corresponding to multiple detection positions; The third step is to analyze the test data to obtain the characteristic data of the steel strip to be tested. The fourth step is to output the characteristic data of the steel strip to be tested.

2. The non-destructive testing method for steel strips according to claim 1, characterized in that, In the first step, the identification information of the steel strip to be tested includes the grade of the steel strip, the coil number, the supplier information, and / or the smelting number, and the attribute information of the steel strip to be tested includes the model information, specification information, material information, and / or application information.

3. The non-destructive testing method for steel strips according to claim 1, characterized in that, In the second step, the steel strip to be tested is scanned and detected, including using a detection probe to locally excite the steel strip, obtain electromagnetic characteristic signals, then amplify and filter the signals, and then transmit them to the electromagnetic detection equipment via a cable.

4. The non-destructive testing method for steel strips according to claim 3, characterized in that, In the second step, detection data corresponding to multiple detection locations are acquired, including determining the predetermined detection locations of the steel strip to be detected based on the identification information and attribute information of the steel strip to be detected.

5. The non-destructive testing method for steel strip according to claim 1, characterized in that, In the second step, when there is a large amount of detection data, the detection data is further filtered, and the filtering method includes: The performance priority of test data at different locations of the steel strip is as follows: priority of test data at the middle of the steel strip > priority of test data at the tail of the steel strip > priority of test data at the head of the steel strip. Alternatively / and, the priority of detection data obtained at the latest test time is greater than the priority of detection data obtained at previous test times.

6. The non-destructive testing method for steel strip according to claim 1, characterized in that, In the third step, the methods for analyzing the detection data include stepwise regression or artificial neural network methods.

7. The non-destructive testing method for steel strips according to claim 6, characterized in that, The stepwise regression method includes: Step S301: First, acquire multiple detection data for the steel strip; Step S302: Obtain the actual mechanical property data of the steel strip through testing and analysis; Step S303: Using the detection data as the input independent variable and the mechanical characteristic data as the output dependent variable, combine these data to construct one or more regression calculation formulas; Step S304: Test the significance of the regression coefficients in each regression formula, delete independent variables with insignificant coefficients, recombine the independent variables, and then return to step S303 for iterative testing; Step S305: If the significance of the regression coefficient in the regression calculation formula meets the test criteria, then retain the corresponding independent variable and return to step S304 to continue testing the significance of the regression coefficients of other independent variables that have not yet been tested, until all independent variables have been tested. Step S306: Finally, the desired optimal regression formula is obtained.

8. The non-destructive testing method for steel strips according to claim 6, characterized in that, The neural network method includes: Step S401: First, obtain multiple test data for the steel strip; Step S402: Obtain the actual mechanical property data of the steel strip through testing and analysis; Step S403: Construct a neural network model, using the detection data as the input independent variable of the neural network model and the mechanical characteristic data as the output dependent variable of the neural network model; Step S404: Initialize the weights and thresholds corresponding to each layer of the neural network model; Step S405: Using the existing detection data and mechanical characteristic data as sample data, input them into the neural network model, calculate the output layer error and hidden layer error in the neural network model, and correct the output layer error and the weights and thresholds of the hidden layer until all sample data have completed training; Step S406: Determine whether the trained neural network model meets the conditions for training completion. If it does, the training is completed, and the neural network model is used for practical applications. If it does not meet the conditions, return to step S404 to rebuild and train again until a neural network model that can be used for practical applications is obtained.

9. The non-destructive testing method for steel strip according to claim 1, characterized in that, The output characteristic data of the steel strip to be tested includes: longitudinal and transverse tensile strength, longitudinal and transverse yield strength, longitudinal and transverse elongation after fracture, Vickers hardness and / or confidence interval.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-9.

Images