A Tension Monitoring Method and System for a Core Conveyor Device

Abstract

This application relates to the field of tension monitoring, and particularly to a tension monitoring method and system for a die core conveying device. The method includes the steps of: calculating the tension degree of the entire die core; sending the tension degree to a preset matching model to output the corresponding counterweight; the calculation method of the tension degree is: setting a window on the real-time collected image so that the die core area passes through the window, and dividing the entire die core into multiple segments of die cores according to the window; during the conveying process of the overall die core, calculating the bending degree of each segment of die core, and constructing a bending degree sequence of each segment of die core regarding time and bending degree; taking the Pearson correlation coefficient of the bending degree sequences corresponding to each segment of die core with adjacent positions as the similarity, and taking the mean value of the similarity as the tension degree of the entire die core. This application has the effect of automatically adjusting the counterweight of the conveying device according to the change of the die core tension.

Description

Technical Field

[0001] This application relates to the field of tension monitoring, and particularly to a tension monitoring method and system for a die core conveying device. Background Art

[0002] The technological process of a rubber hose includes putting raw materials such as rubber into a mixer for mixing, then extruding and forming through an extruder and cooling, and then entering a winding machine for further processing. The hose forming methods are mainly divided into a coreless method, a hard core method, and a soft core method according to whether a core is used. Their main differences lie in whether a core is used and the material and characteristics of the core material used. The hard core method uses a hard core, which is usually solid and can be made of metal or other rigid materials. The hard core maintains its shape during the vulcanization process of the hose, helping to maintain the inner diameter size of the hose.

[0003] The die cores (pipe cores) of metal hoses are usually set in rolls and need to be unrolled through a bottom pressing automatic lifting and lowering device (die core conveying device) during use. The bottom pressing automatic lifting and lowering device is as Figure 1 shown, including a frame 1, a fixed frame 2, a movable frame 3, a tension pulley 4, a suspension rope 5, and a weight 6. The fixed frame 2 is fixed on the frame 1. A slide rail 7 is provided on the frame 1. The movable frame 3 is slidably connected to the slide rail 7 and can slide vertically to lift and lower. One end of the suspension rope 5 is fixed to the movable frame 3, and the other end is fixed to a hoisting device (not shown) provided at the rear side of the frame 1. The rotation of the hoisting device can lift or lower the movable frame 3. A plurality of weights 6 are provided on the movable frame 3. The hoisting device can adjust the length of the suspension rope 5 by winding and unwinding to pull the movable frame 3 to slide on the slide rail 7. Multiple groups of tension pulleys 4 are provided on both the fixed frame 2 and the movable frame 3. The free end of the rolled die core is pulled to wind the die core around the tension pulleys 4 on the fixed frame 2 and the movable frame 3 one by one. The tension pulley 4 can rotate to convey the die core and make the die core taut during the conveying process. The free end is pulled by the device of the next process to perform the unwinding operation. During the unwinding process, the movement of the die core pulls the tension pulley 4 to rotate, and the die core is tensioned by the weight of the weight 6.

[0004] The above device requires manual setting of the counterweight (adjusting the number of weights) to adjust the tension degree of the die core. However, during the conveying process of the die core, due to factors such as changes in the traction force of the die core or changes in the bending degree of the die core itself, the weight of the initially set die core counterweight is not suitable for the changing application scenarios. If the tension degree of the die core becomes lower, the conveying of the die core needs to be stopped at this time, and the equipment for the subsequent processing procedures of the conveying device also needs to be stopped. After adjusting the counterweight, the unwinding process and subsequent processing are continued, affecting the production efficiency. Summary of the Invention

[0005] To solve the technical problem of how to automatically adjust the counterweight of the conveying device according to the change of the core tension, the present application provides a tension monitoring method and system for core conveying equipment.

[0006] In the first aspect, the present application provides a tension monitoring method for core conveying equipment, adopting the following technical solutions:

[0007] A tension monitoring method for core conveying equipment includes the steps of: calculating the tension degree of the whole core; sending the tension degree to a preset matching model to output the corresponding counterweight; calculating the similarity degree of the bending degree sequences corresponding to each section of the core according to the DTW algorithm, and sending an alarm signal in response to the similarity degree being lower than a preset first threshold and higher than a second threshold; the calculation method of the tension degree is: setting a window on the real-time acquired image to make the core area pass through the window, and dividing the whole core into multiple sections of the core according to the window; during the conveying process of the whole core, calculating the bending degree of each section of the core, and constructing a bending degree sequence of each section of the core regarding time and bending degree; taking the Pearson correlation coefficient of the bending degree sequences corresponding to each adjacent section of the core as the similarity degree, and taking the mean value of the similarity degrees as the tension degree of the whole core.

[0008] The beneficial effect is that during the conveying process of the core, multiple sections of the core arranged in parallel between the fixed frame and the movable frame are in a vertical state or a state close to vertical when tensioned. When there is an untensioned situation, the core will bend, and the greater the bending degree. Therefore, the present application reflects the tension degree of the core by quantifying the bending degree of the core.

[0009] When the tension of the core is low, the tension degrees of multiple sections of the core will show an asynchronous phenomenon. By calculating the similarity degree of the bending degree sequences formed by the tension degrees of each section of the core at the same moment, this asynchronous phenomenon can be quantified. The greater the similarity degree of the two bending degree sequences, the higher the synchronization of the tension degree of the core. On the contrary, the lower the synchronization of the tension degree. At this time, it may be due to low tension that there is an untensioned winding phenomenon. By calculating the similarity degrees of all adjacent core sections and taking their mean value, a comprehensive tension degree index can be obtained, and this index can reflect the overall tension state of the whole core. According to the tension degree of the overall model, the corresponding counterweight is matched according to the matching model to achieve the effect of automatically adjusting the counterweight of the conveying device according to the change of the core tension.

[0010] According to the DTW algorithm, calculate the similarity degree of the bending degree sequences corresponding to each section of the core. If the similarity degree of the bending degree sequences is low, but the similarity degree is high, it means that there is an abnormal core section at this time, and this core section still has not been unfolded after passing through multiple tension wheels. At this time, there may be a phenomenon of core self-structure deformation or self-twisting, and it is necessary to stop the machine for manual processing.

[0011] Optionally, the calculation method of the bending degree is as follows: , where represents the bending degree of the th segment of the die core at time for the th segment of the die core at time represents the curvature of the pixel point on the outer edge of the die core area, represents the total number of pixel points on the outer edge, represents the exponential function with as the base.

[0012] The beneficial effect is that by calculating the average value of the curvature of the pixel points on the outer edge of the die core area and normalizing the average value, a quantified bending degree is obtained. The larger the average value of the curvature of the pixel points on the outer edge of the die core area, the greater the bending degree; conversely, the smaller the bending degree.

[0013] Optionally, the calculation formula of the curvature is as follows: Obtain the gray value of the outer edge pixel points and construct a gray value function; construct a Hessian matrix based on the gray value function and calculate the first eigenvalue and the second eigenvalue of the Hessian matrix; obtain the curvature according to the first eigenvalue and the second eigenvalue; the expression of the curvature is = , where represents the curvature of the pixel point on the outer edge of the die core area, represents the first eigenvalue, represents the second eigenvalue.

[0014] The beneficial effect is that the eigenvalues describe the stretching and compression characteristics of the Hessian matrix and can be used to judge the local shape of a point in the image. The product of the eigenvalues can be regarded as a measure of the local curvature, and the sum of the eigenvalues can be regarded as a measure of the local scale. Therefore, this ratio can provide a quantified curvature.

[0015] Optionally, the calculation formula of the curvature is: = , where represents the curvature of the pixel point on the outer edge of the die core area. The pixel points adjacent to both sides of the pixel point are the pixel point and the pixel point respectively, is the angle between the line connecting the pixel point and the pixel point and the line connecting the pixel point and the pixel point . Indicates a pixel point The connection line with the pixel point Length of the connection line

[0016] The beneficial effect is as follows: Another method for calculating curvature is provided. The circumscribed circle is determined according to the pixel points on both sides of any pixel, and then the curvature of the pixel point is calculated based on the method of the curvature of the three-point circumscribed circle

[0017] Optionally, the method for calculating the degree of bending is as follows: Set the long side close to the inner edge of the die core area on the window as the target side, calculate the shortest distance from the pixel points on the outer edge of the die core area to the target side, and use the normalized result of the average value or the maximum value of the shortest distance as the degree of bending

[0018] The beneficial effect is as follows: Calculate the length of the perpendicular line segment from the points on the outer edge of the die core area to the target side. When the die core is more bent, the tension degree of the die core is smaller at this time. The normalized result of the average value or the maximum value of the above shortest distance is larger, and the degree of bending is larger at this time. The possibility that the die core is not tensioned is higher at this time. On the contrary, the degree of bending Is smaller, and the possibility that the die core is tensioned is higher at this time

[0019] Optionally, the calculation formula for the degree of bending is , where Indicates The bending degree of the die core in the th segment at the Indicates the th segment of the die core at the moment, Indicates the area of the window area

[0020] The beneficial effect is as follows: When the die core is more bent, the tension degree of the die core is smaller at this time, and the area of the die core area in the window is larger. On the contrary, when the die core is straighter, the tension degree of the die core is larger, and the area of the die core area in the window is smaller. Therefore, the bending degree of the die core can be quantified by the proportion of the die core area in the window. Specifically, the larger the value of the bending degree, the higher the bending degree of the die core, and the higher the possibility that the die core is not tensioned at this time; the smaller the value of the bending degree, the lower the bending degree of the die core, and the higher the possibility that the die core is tensioned at this time

[0021] Optionally, the setting method of the matching model is as follows: Use the acquisition time of the historical acquisition image and the tension degree of the entire die core as the training data, label the training data, the label is the counterweight setting, input the labeled training data into the neural network model, train the neural network model, and when the training reaches the preset training stop condition, stop training to obtain the matching model

[0022] In the second aspect, the present application provides a tension monitoring system for a die core conveying device, and adopts the following technical solutions

[0023] A tension monitoring system for a core conveying device, comprising: a processor and a memory, the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the tension monitoring method for the core conveying device described above is implemented.

[0024] The beneficial effect is: generating a computer program for the above-mentioned tension monitoring method for the core conveying device and storing it in the memory to be loaded and executed by the processor. Thus, making a system according to the memory and the processor is convenient to use.

[0025] The present application has the following technical effects:

[0026] 1. The degree of tension of the core is reflected by quantifying the degree of bending of the core. By calculating the similarity of the degree of tension of each section of the core at the same moment to form a bending degree sequence, the synchronism of the tension degree of each section of the core can be quantified. When the similarity of the two bending degree sequences is greater, it indicates that the synchronism of the tension degree of the core is higher. On the contrary, the synchronism of the tension degree is lower. At this time, it may be due to low tension that there is a phenomenon of non-tight winding. By calculating the similarity of all adjacent core sections and taking their average value, a comprehensive tension degree index can be obtained, and this index can reflect the overall tension state of the entire core. According to the tension degree of the overall model and the matching model, the corresponding counterweight is matched to achieve the effect of automatically adjusting the counterweight of the conveying device according to the change of the core tension.

[0027] 2. According to the DTW algorithm, calculate the similarity of the bending degree sequences corresponding to each section of the core. If the similarity of the bending degree sequences is low but the degree of similarity is high, it indicates that there is an abnormal core section at this time, and this core section still has not been unfolded after passing through multiple tension wheels. At this time, there may be a phenomenon of the above-mentioned core structure deformation or self-twisting, and it is necessary to stop the machine for manual processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features, and advantages of the exemplary embodiments of the present application will become easy to understand. In the drawings, several embodiments of the present application are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts.

[0029] Figure 1 It is a background technical drawing showing the structure of the conveying device.

[0030] Figure 2 It is a method flow chart of a tension monitoring method for a core conveying device according to an embodiment of the present application.

[0031] Figure 3It is a method flowchart for calculating the tension degree in a tension monitoring method for a core conveying device according to an embodiment of the present application.

[0032] Figure 4 It is a schematic plan view showing multiple core segments and multiple windows in a tension monitoring method for a core conveying device according to an embodiment of the present application.

[0033] Explanation of reference numerals: 1, frame; 2, fixed frame; 3, movable frame; 4, tension pulley; 5, suspension rope; 6, weight; 7, slide rail. Detailed implementation manners

[0034] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present application.

[0035] It should be understood that when the claims, the description and the drawings of the present application use terms such as "first" and "second", they are only used to distinguish different objects, rather than to describe a specific order. The terms "including" and "comprising" used in the description and claims of the present application indicate the presence of the described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or their combinations.

[0036] An embodiment of the present application discloses a tension monitoring method for a core conveying device. Referring to Figure 2 , it includes step S1 - step S2, specifically as follows:

[0037] S1: Calculate the tension degree of the entire core, send the tension degree to a preset matching model, and output the corresponding counterweight.

[0038] Referring to Figure 3 , the calculation method of the tension degree includes step S10 - step S12, specifically as follows:

[0039] S10: Set a window on the real-time acquired image so that the core area passes through the window, and divide the entire core into multiple core segments according to the window.

[0040] In one embodiment, a camera is set to shoot the video stream of the conveying device in motion, and frames are extracted from the video stream to obtain the actually acquired image. The frequency of frame extraction can be 30 frames per second, and the frequency of frame extraction can be adjusted according to the actual application scenario, which will not be elaborated here.

[0041] The shooting angle of this application is preferably the shooting angle of the perspective between the front view angle and the left view angle of the conveying device, and can capture Figure 1 multiple segments of die core images juxtaposed between the fixed frame 2 and the movable frame 3 in Figure 4 as shown. The die core area to be analyzed is intercepted through the window, and the window is Figure 4 a rectangular frame formed by a dotted line in

[0042] In one embodiment, the number of windows is one, and multiple die core areas are within the same window. The number of windows can also be multiple, and the number of windows is the same as the number of segments of the die core, with each segment of the die core corresponding to one window.

[0043] During the conveying process of the die core, the multiple segments of die core juxtaposed between the fixed frame 2 and the movable frame 3 are vertical or nearly vertical in the tensioned state. When there is a situation of non-tension, the die core will bend, and the greater the degree of bending. Therefore, this application reflects the tension degree of the die core by quantifying the bending degree of the die core.

[0044] S11: During the conveying process of the overall die core, calculate the bending degree of each segment of the die core, and construct a bending degree sequence of each segment of the die core regarding time and bending degree.

[0045] Determine the pixel points of the outer edge or inner edge of the die core area through image processing techniques such as edge detection algorithms.

[0046] In one embodiment, the calculation method of the bending degree is: , where represents the bending degree of the th segment of the die core at time. For the th segment of the die core at time, represents the curvature of the pixel point on the outer edge of the die core area,

[0047] represents the mean value of the curvatures of the pixel points on the outer edge of the die core area. Here, the outer edge of the die core area can also be replaced by the inner edge of the die core area. represents an exponential function with as the base, used to normalize to obtain the bending degree . The greater the mean value of the curvatures of the pixel points on the outer edge of the die core area, the greater the bending degree , and vice versa, the smaller the bending degree .

[0048] Among them, the curvature of a pixel point can be calculated through the Hessian matrix. Specifically, the gray value of the outer edge pixel point is obtained to construct a gray value function. The gray value function is a function that describes the change in image brightness and shows the brightness distribution in different regions of the image.

[0049] Construct the Hessian matrix according to the gray value function; the Hessian matrix is a matrix composed of second-order partial derivatives and is used to describe the local curvature of the function. The expression of the Hessian matrix can be: , where are all for the image at and directions of the second-order partial derivatives. The construction of the Hessian matrix is a prior art and will not be elaborated here.

[0050] Calculate the first eigenvalue and the second eigenvalue of the Hessian matrix; obtain the curvature according to the first eigenvalue and the second eigenvalue. The expression of the curvature is = , where represents the curvature of the pixel point on the outer edge of the core region, represents the first eigenvalue, represents the second eigenvalue.

[0051] The eigenvalue describes the stretching and compression characteristics of the Hessian matrix and can be used to judge the local shape of a certain point in the image. The product of the eigenvalues can be regarded as a measure of the local curvature, and the sum of the eigenvalues can be regarded as a measure of the local scale. Therefore, this ratio can provide a quantified curvature.

[0052] The calculation method of the curvature can also be to determine the circumscribed circle according to the pixel points on both sides of any pixel, and then calculate the curvature of the pixel point based on the method of the curvature of the three-point circumscribed circle. The calculation formula of the curvature is: = , where represents the curvature of the pixel point on the outer edge of the core region, the pixel points adjacent to both sides of the pixel point are respectively the pixel point and the pixel point , is the included angle between the line connecting the pixel point and the pixel point and the line connecting the pixel point and the pixel point , represents the pixel point and the pixel point The length of the connection line. The method based on the curvature of the circumcircle of three points is a prior art and will not be elaborated here.

[0053] In one embodiment, the calculation method of the degree of bending can also be: set the long side of the window close to the inner edge of the die core area as the target side, calculate the shortest distance from the pixel points on the outer edge of the die core area to the target side, and use the average value of the shortest distances or the normalized result of the maximum value of the shortest distances as the degree of bending.

[0054] In one embodiment, taking the calculation of the maximum value of the above-mentioned shortest distance as an example, the mathematical expression of the degree of bending is: , where represents the degree of bending of the th section of the die core at the th moment. For the th section of the die core at the th moment, represents the shortest distance value from the first pixel point on the outer edge of the die core area to the target side, represents the shortest distance value from the second pixel point on the outer edge of the die core area to the target side, represents the shortest distance value from the th pixel point on the outer edge of the die core area to the target side. In this application, the shortest distance from a point to a line refers to the length of the perpendicular line segment from the point to the line. The formula for calculating the shortest distance from a point to a line segment can be used for calculation. This calculation is a prior art and will not be elaborated here. represents the maximum value function, that is, taking the maximum value in . The purpose is to perform normalization processing on .

[0055] Calculate the length of the perpendicular line segment from the points on the outer edge of the die core area to the target side. When the die core is more bent, the tension degree of the die core is smaller at this time, is larger, and the degree of bending is larger at this time. The possibility that the die core is not tensioned is higher. On the contrary, the degree of bending is smaller, and the possibility that the die core is tensioned is higher.

[0056] In one embodiment, the calculation formula of the degree of bending is: , where represents the degree of bending of the th section of the die core at the th moment, represents the area of the Denote the area of the window region. When the die core is more bent, the tension degree of the die core is smaller at this time, and the area of the die core region within the window is larger. On the contrary, when the die core is straighter, the tension degree of the die core is larger, and the area of the die core region within the window is smaller. Therefore, the bending degree of the die core can be quantified by the proportion of the die core region area within the window. Specifically, the larger the value of the bending degree, the higher the bending degree of the die core, and the higher the possibility that the die core is not tensioned at this time; the smaller the value of the bending degree, the lower the bending degree of the die core, and the higher the possibility that the die core is tensioned at this time.

[0057] S12: Use the Pearson correlation coefficient of the bending degree sequences corresponding to each adjacent die core segment as the similarity, and use the mean value of the similarities as the tension degree of the overall die core.

[0058] The calculation formula for the similarity can be: , where denotes the time series and the sequence similarity, denotes the Pearson correlation coefficient, denotes the sequence at time , denotes the sequence at time . The closer the value of the Pearson correlation coefficient is to 1, the more positively correlated the sequences and the sequence are. The closer the value of the Pearson correlation coefficient is to 0, the weaker the correlation between the sequences and the sequence . The closer the similarity between the sequences and the sequence is to 1, the more synchronous the operation of the two die core segments is, and the more likely the die core is tensioned at this time. On the contrary, it means that the operation of the two die core segments is less synchronous, and the higher the possibility of non-tensioning at this time.

[0059] When the tension of the die core is low, the tension degrees of multiple die core segments will show an asynchronous phenomenon. By calculating the similarity degree of the bending degree sequences composed of the tension degrees of each die core segment at the same moment, this asynchronous phenomenon can be quantified. The greater the similarity between the two bending degree sequences, the higher the synchrony of the tension degree of the die core. On the contrary, the lower the synchrony of the tension degree. At this time, it may be due to low tension that there is a phenomenon of non-tensioning winding. By calculating the similarities of all adjacent die core segments and taking their mean value, a comprehensive tension degree index can be obtained, and this index can reflect the overall tension state of the entire die core.

[0060] In one embodiment, the method for setting the matching model is as follows: The acquisition time of the historical acquisition images and the tension degree of the entire die core are used as training data, and labels are assigned to the training data. The label is the counterweight setting. The training data with labels is input into the neural network model, and the neural network model is trained. When the training reaches the preset training stop condition, the training is stopped to obtain the matching model.

[0061] Among them, the neural network model can be a BP (Back Propagation) neural network model, and the loss function is the mean square error loss function.

[0062] Among them, the label being the counterweight setting refers to the weight of the weight 6 and the number of turns of the winch. The weight of the weight 6 is fixed, and by adjusting Figure 1 the length of the middle lifting rope 5 released, that is, adjusting the number of turns of the winch, the height of the movable frame 3 is adjusted.

[0063] In the initial state, the die core is tensioned under the combined action of the gravity of the weight 6 and the traction force of the lifting rope 5. When the tension of the die core changes, by controlling the number of turns of the winch to adjust the length of the lifting rope 5, the height of the movable frame 3 can be adjusted. For example, when the die core is not tensioned, the winch rotates to unwind the rope, reducing the height of the movable frame 3 to tension the die core.

[0064] In other embodiments, a jacking device such as a hydraulic cylinder can also be provided below the movable frame 3. By adjusting the extension length of the cylinder shaft of the hydraulic cylinder, the height of the movable frame 3 is adjusted to achieve the effect of adjusting the tension degree of the die core.

[0065] S2: Calculate the similarity degree of the bending degree sequences corresponding to each section of the die core according to the DTW algorithm. In response to the similarity degree being lower than the preset first threshold and higher than the second threshold, an alarm signal is issued.

[0066] Furthermore, the dynamic time warping (DTW) algorithm can be used to calculate the change in the tension degree of a section of the die core at different positions at different times. At this time, it can be detected whether the structure of this section of the die core is deformed by itself (such as partial cracking, etc.) or twisted by itself (one part of the die core is wound or twisted with another part, resulting in abnormal unfolding or operation).

[0067] Specifically, the similarity of the bending degree sequences corresponding to each die core segment is calculated according to the DTW algorithm. If the similarity of the bending degree sequences is low but the similarity degree is high, it indicates that an abnormal die core segment appears at this time, and the die core segment still has not been unfolded after passing through multiple tension wheels 4. At this time, the above-mentioned phenomenon of the die core's own structural deformation or self-twisting may occur, and it is necessary to stop the machine for manual processing. Exemplarily, if the similarity of the bending degree sequences approaches 0 (such as lower than the preset first threshold of 0.3) and the similarity degree approaches 1 (such as higher than the second threshold of 0.8), it indicates that the die core has deformation or twisting, and an alarm signal is sent to prompt manual processing. Otherwise, the conveying operation continues.

[0068] The dynamic time warping algorithm is a method for measuring the similarity between two time bending degree sequences, which is applicable to the situation where the lengths of the two time bending degree sequences are different or there is a displacement on the time axis. It is a prior art and will not be elaborated here.

[0069] The embodiment of the present application also discloses a tension monitoring system for a die core conveying device, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, a tension monitoring method for a die core conveying device according to the present application is implemented.

[0070] The above system also includes other components well-known to those skilled in the art such as a communication bus and a communication interface. Their settings and functions are known in the art, so they will not be elaborated here.

[0071] In the present application, the foregoing memory can be any tangible medium that contains or stores a program, and the program can be used by or in combination with an instruction execution system, device, or device. For example, the computer-readable storage medium can be any suitable magnetic storage medium or magneto-optical storage medium, such as resistive random access memory RRAM (Resistive Random Access Memory), dynamic random access memory DRAM (Dynamic Random Access Memory), static random access memory SRAM (Static Random Access Memory), enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), high bandwidth memory HBM (High Bandwidth Memory), hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that can be used to store the required information and can be accessed by an application program, module, or both. Any such computer storage medium can be part of the device or accessible or connectable to the device.

[0072] Although this specification has shown and described multiple embodiments of the present application, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will think of many changes, alterations, and alternative ways without departing from the spirit and idea of the present application. It should be understood that various alternatives to the embodiments of the present application described herein may be employed in the practice of the present application.

[0073] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereby. Therefore, all equivalent changes made according to the structure, shape, and principle of the present application shall be covered within the protection scope of the present application.

Claims

1. A tension monitoring method for a core conveying device, characterized in that: Includes steps: Calculate the tension of the entire mold core; send the tension to the preset matching model and output the corresponding counterweight; The calculation method of the tension degree is as follows: a window is set on the real-time acquisition image, the mold core area passes through the window, and the whole mold core is divided into multiple mold core segments according to the window; during the transportation process of the whole mold core, the bending degree of each mold core segment is calculated, and the bending degree sequence of each mold core segment with respect to time and bending degree is constructed; the Pearson correlation coefficient of the bending degree sequence corresponding to each mold core segment with adjacent positions is used as the similarity, and the mean of the similarity is used as the tension degree of the whole mold core; The matching model is set up by taking the acquisition time of the historical acquisition image and the tension degree of the whole mold core as training data, labeling the training data, the label is the counterweight setting, inputting the labeled training data into the neural network model, training the neural network model, and stopping the training to obtain the matching model when the training reaches the preset training stop condition; The similarity of the bending degree sequence corresponding to each segment of the mold core is calculated according to the DTW algorithm; In response to the similarity being lower than a preset first threshold and the similarity being higher than a second threshold, an alarm signal is issued.

2. The tension monitoring method for core conveying equipment according to claim 1, characterized in that: The degree of curvature is calculated as: ,in express Moment The bending degree of the segment core is The moment Segment mold core, Represents the pixel points on the outer edge of the core area The curvature of Represents the total number of pixels on the outer edge, Indicates An exponential function with base .

3. The tension monitoring method for core conveying equipment according to claim 2, characterized in that: The curvature is calculated as: Get the gray value of the outer edge pixel and construct a gray value function; Construct a Hessian matrix according to the gray value function, and calculate the first eigenvalue and the second eigenvalue of the Hessian matrix; The curvature is obtained based on the first eigenvalue and the second eigenvalue; the expression of the curvature is = ,in, Represents the pixel points on the outer edge of the core area The curvature of represents the first eigenvalue, represents the second eigenvalue.

4. The tension monitoring method for a core conveying device according to claim 2, characterized in that: The curvature is calculated as: = ,in, Represents the pixel points on the outer edge of the core area The curvature, pixel The adjacent pixels on both sides are pixel points and pixels , Pixel With pixels The lines and pixels With pixels The angle between the lines of Represents pixel With pixels The length of the connection.

5. The tension monitoring method for core conveying equipment according to claim 1, characterized in that: The degree of curvature is calculated as follows: the long side of the window close to the inner edge of the core area is set as the target side, the shortest distance between the pixel point on the outer edge of the core area and the target side is calculated, and the normalized result of the mean or maximum value of the shortest distance is used as the degree of curvature.

6. The tension monitoring method for core conveying equipment according to claim 1, characterized in that: The calculation formula for the degree of bending is: ,in express Moment The bending degree of the segment core, Indicates the time The area of ​​the segment core, Indicates the area of ​​the window region.

7. A tension monitoring system for a core conveying device, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a tension monitoring method for a core conveying device according to any one of claims 1-6 is implemented.

Images