A detection system and device for tensile properties of pipes

Abstract

The present invention provides a detection system and device for the tensile performance of a pipe. A tensile fixing module, a strain gauge sensor module fixed on the pipe, an image acquisition and processing module fixed on the tensile fixing module, and a central processing module provided with a decision tree unit are provided. A decision tree model is constructed through the decision tree unit. The decision tree model provides a first control signal according to the material and wall thickness of the pipe. The central processing module controls the tensile force and bending force provided by the tensile fixing module in real time in combination with the first control signal, the tensile electrical signal and the image signal. The present invention can monitor the situation of the pipe in the tensile process in real time and comprehensively, and adjust the tensile force and bending force in time according to the real-time situation, so that the detection process is closer to the actual use situation, so that the detection data is closer to the actual value, and the beneficial effect of safety and stability is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of pipeline detection, and in particular to a detection system and device for the tensile performance of a pipe. Background Art

[0002] Pipes are pipeline materials used to transport various fluids, such as liquids and gases. Pipes are often used in various fields such as chemical industry, environment, and new energy, especially in petrochemical industry, thermal power plant, and shipping industry. The performance of pipes is directly related to the safety of personnel around the pipes, maintenance cycle, and maintenance costs. The materials required for pipelines are not only about the cooling and heat insulation performance, but also the adhesion strength of the material itself under different pipe types and various orientations. Therefore, it is necessary to test the tensile strength of the pipes.

[0003] At present, for the tensile strength test of pipes, the traditional tensile test method mainly focuses on the performance of pipes under axial tensile force. In actual applications, pipes may be subjected to a variety of complex stress combinations, and local deformation and fracture phenomena are not tested. At present, large errors will be generated when testing the tensile force of pipes, resulting in the inability to accurately obtain the key mechanical performance indicators such as the actual yield strength and tensile strength of the pipes, causing the measured data to deviate from the true value. The data measured by the traditional test method not only cannot fully reflect the performance of the pipes in actual conditions, but also cannot simulate the actual use during the test process, resulting in too large an error value in the test data, which brings some safety issues. Summary of the invention

[0004] The present invention provides a system and device for detecting the tensile performance of a pipe, which are used to solve the defects in the prior art that the performance of the pipe under actual conditions cannot be fully reflected when detecting the tensile strength, and the actual usage conditions cannot be simulated during the detection process, resulting in too large an error value of the detection data, thus causing some safety problems.

[0005] The present invention provides a system for detecting the tensile properties of a pipe, comprising:

[0006] A stretching and fixing module, the stretching and fixing module is used to fix the pipe and provide the pipe with a stretching force and a bending force;

[0007] A strain gauge sensor module is fixed on the pipe and used to detect the tensile data of the pipe in real time. The tensile data includes strain data and tension data. The strain gauge sensor module calculates the tensile data collected by the formula The stress of the pipe is calculated and a stress-strain curve is drawn, and the strain gauge sensor module converts the stress-strain curve into a tensile electrical signal;

[0008] An image acquisition and processing module, the image acquisition and processing module is fixed to the stretching and fixing module, the image acquisition and processing module is used to acquire the image of the pipe in the stretching stage in real time and convert the image into an image signal, process the image signal to obtain strain distribution data, and display the strain distribution on the surface of the pipe by drawing a strain cloud diagram;

[0009] A central processing module, wherein the central processing module is connected to the stretching and fixing module, and the central processing module is electrically connected to the strain gauge sensor module and the image acquisition and processing module respectively. The central processing module is provided with a decision tree unit, and a decision tree model is constructed by the decision tree unit. The decision tree model provides a first control signal according to the material and wall thickness of the pipe. The central processing module controls the stretching and fixing module in real time in combination with the first control signal, the stretching electrical signal and the image signal. The stretching and fixing module adjusts the stretching force and the bending force provided to the pipe in real time.

[0010] According to a system for detecting the tensile properties of a pipe provided by the present invention, the tensile fixing module includes a tensile force driving structure and a bending force driving structure. The tensile force driving structure generates a tensile force through a screw mechanism driven by a motor, and the bending force driving structure generates a bending force through a bending mechanism driven by a hydraulic motor. The tensile force driving structure and the bending force driving structure are independently controlled to generate a composite force formed by the tensile force and the bending force.

[0011] According to a system for detecting the tensile performance of a pipe provided by the present invention, the tensile fixing module includes an inclined fixing structure, which includes at least an inclined fixing rod. The inclined fixing structure is used to fix one end of the pipe and gradually increase the pressing force on the pipe during the stretching process of the pipe.

[0012] According to a system for detecting the tensile properties of a pipe provided by the present invention, the strain gauge sensor module is wirelessly connected to the central processing module, the strain gauge sensor module includes a plurality of semiconductor strain gauge sensors, the semiconductor strain gauge sensors are arranged at intervals on the surface of the pipe, and the semiconductor strain gauge sensors collect the tensile data of the pipe in real time during the stretching process.

[0013] According to a pipe tensile performance detection system provided by the present invention, the image acquisition processing module uses a weight coefficient of The image is grayscaled, where Y is the grayscale value, and R, G and B are the pixel values ​​of the red, green and blue channels respectively, and the noise of the image is processed by median filtering, specifically, each pixel point in the image is traversed through a nested loop structure to obtain the value of each pixel in the neighborhood of the pixel point, after obtaining the neighborhood pixel value centered on the current pixel, the pixel values ​​are sorted by the built-in sorting function in Python, and the value of the current pixel point is replaced by the sorted median value.

[0014] According to a pipe tensile property detection system provided by the present invention, the image acquisition processing module is also used to enhance the image, specifically, by finding the minimum and maximum pixel values ​​in the image through the min and max functions of the numpy library in Python, and traversing each pixel of the image, and enhancing the pixel through the formula Stretch, where pixel is the original pixel value and new_pixel is the pixel value after stretching. is the minimum pixel value of the image, is the maximum pixel value of the image, and the enhanced image is displayed through the cv2.imshow function.

[0015] According to a pipe tensile performance detection system provided by the present invention, the decision tree unit constructs the decision tree model through the CART algorithm, the decision tree unit includes a data set, the data set includes the material and wall thickness of the pipe, and the CART algorithm constructs the decision tree model by formula

[0016]

[0017] The Gini index is calculated as, where D is the dataset, K is the total number of categories, is the proportion of samples belonging to the kth class, and the data set D is divided into sub-datasets D1 and D2, where, is a set of samples whose feature A takes the value of a. is a set of samples whose feature A value is not a. Calculate the Gini index after partitioning, where represents the number of samples in the dataset D, represents the number of samples of the sub-dataset D1 obtained after division, It represents the number of samples of the sub-dataset D2 obtained after the division. For each of the sub-datasets, the sub-datasets are further divided to create new child nodes.

[0018] According to a pipe tensile performance detection system provided by the present invention, the decision tree unit evaluates the decision tree model through accuracy, recall rate and F1-score indicators, wherein: , , , the number of samples of the decision tree model is adjusted accordingly according to the F1-score indicator evaluation result, and the calculation is repeated until the F1-score indicator evaluation result reaches the set indicator value.

[0019] According to a detection system for the tensile performance of a pipe provided by the present invention, the central processing module controls the stretching and fixing module in real time in combination with the first control signal, the stretching electrical signal and the image signal, and the stretching and fixing module adjusts the tensile force and the bending force provided to the pipe in real time, specifically comprising:

[0020] The central processing module drives the stretching and fixing module according to the first control signal, and the first control signal includes the initial control parameters of the stretching and fixing module corresponding to the pipe. During the stretching process, the central processing module calculates the actual stress according to the stretching electrical signal and compares it with the target stress set in the first control signal. If there is a deviation between the actual stress and the target stress, the central processing module controls the stretching and fixing module to make the actual stress equal to the target stress, and the central processing module analyzes the image signal and the stretching electrical signal in real time. If the pipe is deformed, the central processing module adjusts the stretching force and the bending force provided by the stretching and fixing module.

[0021] The present invention also provides a device for detecting the tensile properties of a pipe, which comprises the above-mentioned system for detecting the tensile properties of a pipe.

[0022] The present invention provides a detection system and device for the tensile performance of a pipe. The system and device are provided with a tensile fixing module, which is used to fix the pipe and provide tensile force and bending force for the pipe. The strain gauge sensor module fixed on the pipe is used to detect the tensile data of the pipe in real time and convert the stress-strain curve into a tensile electrical signal. The image acquisition processing module fixed on the tensile fixing module is used to collect the image of the pipe in the tensile stage in real time and convert the image into an image signal. The central processing module is provided with a decision tree unit, and a decision tree model is constructed through the decision tree unit. The decision tree model provides a first control signal according to the material and wall thickness of the pipe. The central processing module controls the tensile fixing module in real time in combination with the first control signal, the tensile electrical signal and the image signal. The tensile fixing module adjusts the tensile force and the bending force provided to the pipe in real time. The present invention can monitor the situation of the pipe in the tensile process in real time and comprehensively, and timely adjust the tensile force and the bending force according to the real-time situation, so that the detection process is closer to the actual use situation, so that the detection data is closer to the actual value, and the beneficial effect of safety and stability is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0024] Figure 1 is a schematic structural diagram of a system for detecting the tensile properties of a pipe provided in an embodiment of the present invention;

[0025] Figure 2 It is a flow chart of an image processing method of a system for detecting the tensile properties of a pipe provided in an embodiment of the present invention.

[0026] Reference numerals:

[0027] 1. Detection system; 10. Stretching and fixing module; 11. Strain gauge sensor module; 12. Image acquisition and processing module; 13. Central processing module. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] The present invention provides a detection system for the tensile performance of a pipe, comprising: a stretching and fixing module, the stretching and fixing module is used to fix the pipe and provide the pipe with a stretching force and a bending force, the stretching and fixing module comprises a stretching force driving structure and a bending force driving structure, the stretching force driving structure generates a stretching force through a screw mechanism driven by a motor, the bending force driving structure generates a bending force through a bending mechanism driven by a hydraulic motor, and the composite force formed by the stretching force and the bending force is generated by independently controlling the stretching force driving structure and the bending force driving structure, and in addition, the stretching and fixing module comprises an inclined fixing structure, the inclined fixing structure at least comprises an inclined fixing rod, the inclined fixing structure is used to fix one end of the pipe and gradually increase the pressing force on the pipe during the stretching process of the pipe.

[0030] The detection system of the tensile performance of the pipe of the present invention comprises: a strain gauge sensor module, the strain gauge sensor module is fixed on the pipe and used to detect the tensile data of the pipe in real time, the tensile data includes strain data and tension data, the strain gauge sensor module uses the formula according to the collected tension data The stress of the pipe is calculated and a stress-strain curve is drawn. The strain gauge sensor module converts the stress-strain curve into a tensile electrical signal. The strain gauge sensor module is wirelessly connected to the central processing module. The strain gauge sensor module includes a plurality of semiconductor strain gauge sensors. The semiconductor strain gauge sensors are arranged at intervals on the surface of the pipe. The semiconductor strain gauge sensors collect the tensile data of the pipe in real time during the stretching process.

[0031] Among them, the detection system of the tensile performance of the pipe of the present invention also includes: an image acquisition and processing module, which is fixed to the stretching fixing module, and is used to collect the image of the pipe in the stretching stage in real time and convert the image into an image signal, process the image signal to obtain strain distribution data, and display the strain distribution on the surface of the pipe by drawing a strain cloud map.

[0032] Among them, the detection system of the tensile performance of the pipe of the present invention also includes: a central processing module, the central processing module is connected to the stretching and fixing module, and the central processing module is electrically connected to the strain gauge sensor module and the image acquisition and processing module respectively, the central processing module is provided with a decision tree unit, and a decision tree model is constructed by the decision tree unit. The decision tree model provides a first control signal according to the material and wall thickness of the pipe. The central processing module controls the stretching and fixing module in real time in combination with the first control signal, the stretching electrical signal and the image signal, and the stretching and fixing module adjusts the tensile force and the bending force provided to the pipe in real time.

[0033] Among them, the image acquisition processing module uses the weight coefficient as The image is grayscaled, where Y is the grayscale value, and R, G and B are the pixel values ​​of the red, green and blue channels respectively, and the noise of the image is processed by median filtering, specifically, each pixel point in the image is traversed through a nested loop structure to obtain the value of each pixel in the neighborhood of the pixel point, after obtaining the neighborhood pixel value centered on the current pixel, the pixel values ​​are sorted by the built-in sorting function in Python, and the value of the current pixel point is replaced by the sorted median value.

[0034] In addition, the image acquisition processing module is also used to enhance the image, specifically, by finding the minimum and maximum pixel values ​​in the image through the min and max functions of the numpy library in Python, and traversing each pixel of the image, and converting the pixels into Stretch, where pixel is the original pixel value and new_pixel is the pixel value after stretching. is the minimum pixel value of the image, is the maximum pixel value of the image, and the enhanced image is displayed through the cv2.imshow function.

[0035] The decision tree unit constructs the decision tree model through the CART algorithm, the decision tree unit includes a data set, the data set includes the material and wall thickness of the pipe, and the CART algorithm constructs the decision tree model by formula

[0036]

[0037] The Gini index is calculated as, where D is the dataset, K is the total number of categories, is the proportion of samples belonging to the kth class, and the data set D is divided into sub-datasets D1 and D2, where, is a set of samples whose feature A takes the value of a. is a set of samples whose feature A value is not a. Calculate the Gini index after partitioning, where represents the number of samples in the dataset D, represents the number of samples of the sub-dataset D1 obtained after division, It represents the number of samples of the sub-dataset D2 obtained after the division. For each of the sub-datasets, the sub-datasets are further divided to create new child nodes.

[0038] In addition, the decision tree unit evaluates the decision tree model through accuracy, recall and F1-score indicators, where: , , , the number of samples of the decision tree model is adjusted accordingly according to the F1-score indicator evaluation result, and the calculation is repeated until the F1-score indicator evaluation result reaches the set indicator value.

[0039] The central processing module controls the stretching and fixing module in real time in combination with the first control signal, the stretching electrical signal and the image signal, and the stretching and fixing module adjusts the stretching force and the bending force provided to the pipe in real time, specifically including:

[0040] The central processing module drives the stretching and fixing module according to the first control signal, and the first control signal includes the initial control parameters of the stretching and fixing module corresponding to the pipe. During the stretching process, the central processing module calculates the actual stress according to the stretching electrical signal and compares it with the target stress set in the first control signal. If there is a deviation between the actual stress and the target stress, the central processing module controls the stretching and fixing module to make the actual stress equal to the target stress, and the central processing module analyzes the image signal and the stretching electrical signal in real time. If the pipe is deformed, the central processing module adjusts the stretching force and the bending force provided by the stretching and fixing module.

[0041] The present invention also provides a device for detecting the tensile properties of a pipe, comprising the above-mentioned system for detecting the tensile properties of a pipe.

[0042] The present invention provides a detection system for the tensile performance of a pipe. The system comprises a stretching and fixing module, which is used to fix the pipe and provide a tensile force and a bending force for the pipe. The strain gauge sensor module fixed on the pipe is used to detect the tensile data of the pipe in real time and convert the stress-strain curve into a tensile electrical signal. The image acquisition and processing module fixed on the stretching and fixing module is used to collect the image of the pipe in the stretching stage in real time and convert the image into an image signal. The central processing module is provided with a decision tree unit, and a decision tree model is constructed through the decision tree unit. The decision tree model provides a first control signal according to the material and wall thickness of the pipe. The central processing module controls the stretching and fixing module in real time in combination with the first control signal, the tensile electrical signal and the image signal. The stretching and fixing module adjusts the tensile force and the bending force provided to the pipe in real time. The present invention can monitor the situation of the pipe in the stretching process in real time and comprehensively, and timely adjust the tensile force and the bending force according to the real-time situation, so that the detection process is closer to the actual use situation, so that the detection data is closer to the actual value, and the beneficial effect of safety and stability is achieved.

[0043] In this embodiment, please refer to Figure 1 and Figure 2 , which are respectively a structural schematic diagram of a detection system for the tensile properties of a pipe provided in an embodiment of the present invention and a flow chart of an image processing method of a detection system for the tensile properties of a pipe provided in an embodiment of the present invention.

[0044] Specifically, in this embodiment, Figure 1 and Figure 2As shown, a system 1 for detecting the tensile properties of a pipe provided by an embodiment of the present invention includes: a tensile fixing module 10, wherein the tensile fixing module 10 is used to fix the pipe and provide tensile force and bending force for the pipe, wherein the tensile fixing module 10 includes a tensile force driving structure and a bending force driving structure, wherein the tensile force driving structure generates tensile force through a screw mechanism driven by a motor, and the bending force driving structure generates bending force through a bending mechanism driven by a hydraulic motor, and the tensile force driving structure and the bending force driving structure are independently controlled to generate a composite force formed by the tensile force and the bending force. Specifically, the tensile force driving structure and the bending force driving structure can independently control the loading rate and loading size to simulate tensile and bending composite stresses of different proportions.

[0045] In this embodiment, it is best that the stretching and fixing module 10 includes an inclined fixing structure, which includes at least an inclined fixing rod. The inclined fixing structure is used to fix one end of the pipe and gradually increase the pressing force on the pipe during the stretching process of the pipe. Since the pressing force of the pipe gradually increases during the stretching process, the pipe is fixed more firmly and is not easy to fall off, thereby improving safety.

[0046] In this embodiment, the pipe tensile performance detection system 1 of the present invention is further provided with a strain gauge sensor module 11, which is fixed on the pipe and used to detect the tensile data of the pipe in real time. The tensile data includes strain data and tension data. The strain gauge sensor module 11 calculates the tensile data collected by the formula The stress of the pipe is calculated and a stress-strain curve is drawn, and the strain gauge sensor module 11 converts the stress-strain curve into a tensile electrical signal.

[0047] Among them, in this embodiment, the strain gauge sensor module 11 is wirelessly connected to the central processing module 13, and the strain gauge sensor module 11 includes a plurality of semiconductor strain gauge sensors, which are arranged at intervals on the surface of the pipe. Preferably, the plurality of semiconductor strain gauge sensors are arranged in an array on the surface of the pipe, and the semiconductor strain gauge sensor is fixedly connected to the pipe by epoxy resin glue. Specifically, the epoxy resin glue is evenly applied to the base of the semiconductor strain gauge sensor and the surface of the pipe, and then the semiconductor strain gauge sensor is attached to the pipe. During the pasting process, care should be taken to avoid the generation of bubbles to ensure that the semiconductor strain gauge sensor is tightly attached to the surface of the pipe. The semiconductor strain gauge sensor collects the stretching data of the pipe in real time during the stretching process. Because the semiconductor strain gauge sensor has a higher sensitivity, it can more keenly detect the slight strain changes of the pipe and obtain more accurate stretching data.

[0048] In this embodiment, the best, the detection system 1 for the tensile performance of the pipe of the present invention also includes an image acquisition and processing module 12, and the image acquisition and processing module 12 is fixed to the stretching fixing module 10. The image acquisition and processing module 12 is used to collect the image of the pipe in the stretching stage in real time and convert the image into an image signal, and process the image signal to obtain strain distribution data. The strain distribution on the surface of the pipe is displayed by drawing a strain cloud map. The image acquisition and processing module 12 can capture in real time the local fracture or deformation of the pipe that may exist in the stretching process, which is conducive to more accurately obtaining the key mechanical performance indicators such as the real yield strength and tensile strength of the pipe, so that the measured data is closer to the real value.

[0049] In this embodiment, please refer to Figure 2 The image processing flow of the pipe tensile performance detection system 1 provided in the embodiment of the present invention is as follows:

[0050] S1: grayscale the image using a weight coefficient Y=0.299R+0.587G+0.114B;

[0051] S2: traverse each pixel point in the image through a nested loop structure to obtain the value of each pixel in the neighborhood of the pixel point;

[0052] S3: Find the minimum and maximum pixel values ​​in the image using the min and max functions of the numpy library in Python, stretch the image, and display the enhanced image using the cv2.imshow function;

[0053] S4: Use feature point matching algorithm to find corresponding feature points in the images before and after deformation;

[0054] S5: Calculate the displacement vector of each feature point, and obtain the displacement field distribution on the surface of the pipe by analyzing the displacement vectors of multiple feature points.

[0055] In this embodiment, the above-mentioned image acquisition processing flow is specifically described. The image acquisition processing module 12 uses the weight coefficient as The image is grayscaled, wherein Y is the grayscale value, and R, G and B are the pixel values ​​of the red, green and blue channels respectively, and the noise of the image is processed by median filtering, specifically, each pixel point in the image is traversed through a nested loop structure to obtain the value of each pixel in the neighborhood of the pixel point, and after obtaining the neighborhood pixel value centered on the current pixel, the pixel values ​​are sorted by the built-in sorting function in Python, and the value of the current pixel point is replaced by the sorted median value. The image acquisition and processing module 12 converts the acquired color image into a grayscale image, which reduces the data volume and processing complexity of the image, while improving the speed of image processing. In addition, the noise in the image is removed, thereby improving the quality of the image.

[0056] In this embodiment, the image acquisition processing module 12 has the function of enhancing the image, specifically, searching for the minimum and maximum pixel values ​​in the image through the min and max functions of the numpy library in Python, and traversing each pixel of the image, and converting the pixel into Stretch, where pixel is the original pixel value and new_pixel is the pixel value after stretching. is the minimum pixel value of the image, is the maximum pixel value of the image, and the enhanced image is displayed by the cv2.imshow function. By enhancing the denoised image, the feature points and texture information on the surface of the pipe are highlighted, which is more convenient for subsequent image matching and analysis.

[0057] In this embodiment, it is best to use a feature point extraction algorithm to extract representative feature points from the images before and after deformation. These feature points should have good stability and uniqueness, and can be accurately identified and matched in different images. Feature points can be extracted by the algorithm SIFT (Scale-Invariant Feature Transform) or SURF (Speeded Up Robust Features).

[0058] In addition, based on the extracted feature points, a feature point matching algorithm is used to find the corresponding feature points in the images before and after deformation. The feature point matching algorithm has high accuracy and robustness, and can accurately match feature points in the presence of noise, deformation, etc. in the image. The best feature point matching can be achieved through the nearest neighbor algorithm or RANSAC (Random Sample Consensus).

[0059] The best solution is to calculate the displacement vector of each feature point based on the matching feature point pairs. By analyzing the displacement vectors of multiple feature points, the displacement field distribution on the pipe surface can be obtained. Based on the displacement field data, the strain calculation formula is used to calculate the strain distribution on the pipe surface, where the strain calculation formula is Green strain. Then, the calculated strain distribution data is analyzed to study the local deformation and fracture behavior of the pipe. Specifically, the strain distribution on the pipe surface can be intuitively displayed by drawing a strain cloud map, and parameters such as the strain concentration area and the maximum strain value can be analyzed to evaluate the mechanical properties and reliability of the pipe.

[0060] In this embodiment, the detection system 1 for the tensile performance of the pipe of the present invention also includes a central processing module 13, which is connected to the stretching and fixing module 10, and the central processing module 13 is electrically connected to the strain gauge sensor module 11 and the image acquisition and processing module 12 respectively. The central processing module 13 is provided with a decision tree unit, and a decision tree model is constructed by the decision tree unit. The decision tree model provides a first control signal for the stretching and fixing module 10 according to the material and wall thickness of the pipe. The central processing module 13 controls the stretching and fixing module 10 in real time in combination with the first control signal, the stretching electrical signal and the image signal, and the stretching and fixing module 10 adjusts the tensile force and the bending force provided to the pipe in real time.

[0061] The decision tree unit constructs the decision tree model through the CART algorithm, the decision tree unit includes a data set, the data set includes the material and wall thickness of the pipe, and the CART algorithm constructs the decision tree model by formula

[0062]

[0063] The Gini index is calculated as, where D is the dataset, K is the total number of categories, is the proportion of samples belonging to the kth class, and the data set D is divided into sub-datasets D1 and D2, where, is a set of samples whose feature A takes the value of a. is a set of samples whose feature A value is not a. Calculate the Gini index after partitioning, where represents the number of samples in the dataset D, represents the number of samples of the sub-dataset D1 obtained after division, It represents the number of samples of the sub-dataset D2 obtained after the division. For each of the sub-datasets, the sub-datasets are further divided to create new child nodes.

[0064] Among them, the data set is to collect various relevant characteristic data and corresponding tensile performance results of the pipe by conducting a large number of pipe tensile tests. These characteristic data include the material (such as metal, plastic, etc.), outer diameter, inner diameter, wall thickness, manufacturing process (such as extrusion, injection molding, etc.), stretching speed, test temperature, etc. of the pipe. The tensile performance results are mainly key indicators such as yield strength, tensile strength, and elongation at break of the pipe. For example, a data set is constructed in which each record contains the above characteristic information of the pipe and the corresponding tensile performance indicators. Assuming that 1,000 sets of tensile test data of different pipes are collected, these data will serve as the basis for the subsequent decision tree model. In addition, from the many collected pipe features, the features that have a greater impact on the tensile performance are selected as the input features of the decision tree. Through calculation, it is found that the material and wall thickness of the pipe have a higher information gain for the yield strength of the pipe. Therefore, these two features are mainly considered as nodes when constructing the decision tree model.

[0065] In addition, it is best to use the preprocessed data set to train the constructed decision tree model, and divide the data set into a training set and a validation set. Usually 70%-80% of the data is used as the training set, and the remaining data is used as the validation set. In this embodiment, 70% is used as the training set and 30% is used as the validation set. During the training process, the decision tree model adjusts the tree structure and node division rules by learning the relationship between the pipe characteristics and the tensile performance in the training set. For example, during the training process, the decision tree model will continuously optimize the node division boundaries according to the tensile performance corresponding to the pipes of different materials and wall thicknesses in the training set, so that the decision tree can more accurately predict the tensile performance of the pipe.

[0066] In this embodiment, the trained model is evaluated using a validation set, wherein the decision tree unit evaluates the decision tree model using the accuracy, recall and F1-score indicators, wherein: , , , the number of samples of the decision tree model is adjusted accordingly according to the F1-score indicator evaluation result, and the calculation is repeated until the F1-score indicator evaluation result reaches the set indicator value and satisfies the stop condition, so that a hierarchical decision tree model is constructed. Specifically, if the model is used to predict whether the tensile performance of the pipe is qualified, the accuracy is calculated through the validation set, that is, the proportion of qualified and unqualified pipes is correctly predicted. If the accuracy reaches a high level, such as above 80%, it means that the decision tree model has good performance.

[0067] The central processing module 13 controls the stretching and fixing module 10 in real time in combination with the first control signal, the stretching electrical signal and the image signal, and the stretching and fixing module 10 adjusts the stretching force and the bending force provided to the pipe in real time, specifically including:

[0068] The central processing module 13 drives the stretching and fixing module 10 according to the first control signal, and the first control signal includes the initial control parameters of the stretching and fixing module 10 corresponding to the pipe. During the stretching process, the central processing module 13 calculates the actual stress according to the stretching electrical signal, and compares it with the target stress set in the first control signal. If there is a deviation between the actual stress and the target stress, the central processing module 13 controls the stretching and fixing module 10 to make the actual stress equal to the target stress, and the central processing module 13 analyzes the image signal and the stretching electrical signal in real time. If the pipe is deformed, the central processing module 13 adjusts the stretching force and the bending force provided by the stretching and fixing module 10.

[0069] Specifically, when a new pipe needs to be tested for its tensile properties, the relevant characteristics of the pipe (material, wall thickness) are input into the model. The decision tree model will predict the tensile properties of the pipe (such as yield strength, tensile strength, elongation at break, etc.) based on the trained rules. For example, for a new pipe, its material and wall thickness characteristics are input, and the decision tree model can quickly predict whether the yield strength of the pipe meets the standard requirements, and then provide a first control signal based on the relevant characteristics of the pipe.

[0070] In addition, if deformation of the pipe is found through analysis of image signals and stretching electrical signals, the central processing module 13 adjusts the stretching force and bending force provided by the stretching fixing module 10. In the case of local deformation, the deformation can be alleviated by adjusting the clamping force of the local clamp or changing the stretching direction. If the pipe is found to have a bending tendency, an appropriate corrective force can be applied on the opposite side of the bending direction, and the size and direction of the stretching force can be adjusted at the same time to restore the pipe to a normal stretching state. During the entire adjustment process, stress changes are continuously monitored to ensure that the stress is always near the target stress and within a safe range.

[0071] In this embodiment, the process of testing the tensile properties of the pipe is as follows:

[0072] Fixing the pipe to the stretching and fixing module 10;

[0073] Arrange a plurality of semiconductor strain gauge sensors of the strain gauge sensor module 11 in an array on the surface of the pipe and fix them with epoxy resin glue;

[0074] The central processing module 13 provides a first control signal to the stretching and fixing module 10 according to the material and wall thickness corresponding to the pipe in the decision tree model and drives the stretching and fixing module 10 to operate. The stretching and fixing module 10 provides corresponding stretching force and bending force for the pipe according to the first control signal;

[0075] The strain gauge sensor module 11 detects the tensile data of the pipe in real time, obtains the stress of the pipe and draws a stress-strain curve, and converts the stress-strain curve into a tensile electrical signal;

[0076] The image acquisition and processing module 12 acquires the image of the pipe in the stretching stage in real time, performs grayscale processing, noise reduction processing and image enhancement on the image, and then converts the image into an image signal;

[0077] The central processing module 13 controls the stretching and fixing module 10 in real time in combination with the first control signal, the stretching electrical signal and the image signal. The stretching and fixing module 10 adjusts the stretching force and the bending force provided to the pipe in real time until the pipe breaks.

[0078] The present invention also provides a device for detecting the tensile properties of a pipe, including the above-mentioned system 1 for detecting the tensile properties of a pipe. The specific functions are as described above and will not be repeated here.

[0079] The present invention provides a system and device for detecting the tensile properties of a pipe, wherein a tensile fixing module 10 is provided, wherein the tensile fixing module 10 is used to fix the pipe and provide tensile force and bending force to the pipe, a strain gauge sensor module 11 fixed on the pipe is used to detect the tensile data of the pipe in real time and convert the stress-strain curve into a tensile electrical signal, the image acquisition processing module 12 fixed on the tensile fixing module 10 is used to acquire an image of the pipe in the tensile stage in real time and convert the image into an image signal, the central processing module 13 is provided with a decision tree unit, and a decision tree is constructed through the decision tree unit Model, the decision tree model provides a first control signal for the stretching and fixing module 10 according to the material and wall thickness of the pipe, the central processing module 13 controls the stretching and fixing module 10 in real time in combination with the first control signal, the stretching electrical signal and the image signal, and the stretching and fixing module 10 adjusts the stretching force and the bending force provided to the pipe in real time. The present invention can monitor the situation of the pipe in the stretching process in real time and comprehensively, and adjust the stretching force and the bending force in time according to the real-time situation, so that the detection process is closer to the actual usage, so that the detection data is closer to the actual value, and the beneficial effect of safety and stability is achieved.

[0080] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0081] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0082] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A system for detecting the tensile properties of a pipe, characterized in that: include: A stretching and fixing module, the stretching and fixing module is used to fix the pipe and provide the pipe with a stretching force and a bending force; A strain gauge sensor module is fixed on the pipe and used to detect the tensile data of the pipe in real time. The tensile data includes strain data and tension data. The strain gauge sensor module calculates the tensile data collected by the formula The stress of the pipe is calculated and a stress-strain curve is drawn, and the strain gauge sensor module converts the stress-strain curve into a tensile electrical signal; An image acquisition and processing module, the image acquisition and processing module is fixed to the stretching and fixing module, the image acquisition and processing module is used to acquire the image of the pipe in the stretching stage in real time and convert the image into an image signal, process the image signal to obtain strain distribution data, and display the strain distribution on the surface of the pipe by drawing a strain cloud diagram; A central processing module, wherein the central processing module is connected to the stretching and fixing module, and the central processing module is electrically connected to the strain gauge sensor module and the image acquisition and processing module respectively. The central processing module is provided with a decision tree unit, and a decision tree model is constructed by the decision tree unit. The decision tree model provides a first control signal according to the material and wall thickness of the pipe. The central processing module controls the stretching and fixing module in real time in combination with the first control signal, the stretching electrical signal and the image signal. The stretching and fixing module adjusts the stretching force and the bending force provided to the pipe in real time.

2. The tube tensile performance detection system according to claim 1, characterized in that: The stretching and fixing module includes a stretching force driving structure and a bending force driving structure. The stretching force driving structure generates a stretching force through a screw mechanism driven by a motor, and the bending force driving structure generates a bending force through a bending mechanism driven by a hydraulic motor. The composite force formed by the stretching force and the bending force is generated by independently controlling the stretching force driving structure and the bending force driving structure.

3. The tube tensile performance detection system according to claim 2, characterized in that: The stretching and fixing module includes an inclined fixing structure, which includes at least an inclined fixing rod. The inclined fixing structure is used to fix one end of the pipe and gradually increase the pressing force on the pipe during the stretching process of the pipe.

4. The tube tensile performance detection system according to claim 1, characterized in that: The strain gauge sensor module is wirelessly connected to the central processing module. The strain gauge sensor module includes a plurality of semiconductor strain gauge sensors. The semiconductor strain gauge sensors are arranged at intervals on the surface of the pipe. The semiconductor strain gauge sensors collect the stretching data of the pipe in real time during the stretching process.

5. The tube tensile performance detection system according to claim 1, characterized in that: The image acquisition processing module uses a weight coefficient of The image is grayscaled, where Y is the grayscale value, and R, G and B are the pixel values ​​of the red, green and blue channels respectively, and the noise of the image is processed by median filtering, specifically, each pixel point in the image is traversed through a nested loop structure to obtain the value of each pixel in the neighborhood of the pixel point, after obtaining the neighborhood pixel value centered on the current pixel, the pixel values ​​are sorted by the built-in sorting function in Python, and the value of the current pixel point is replaced by the sorted median value.

6. The tube tensile property detection system according to claim 5, characterized in that: The image acquisition processing module is also used to enhance the image, specifically by finding the minimum and maximum pixel values ​​in the image through the min and max functions of the numpy library in Python, and traversing each pixel of the image, and enhancing the pixel through the formula Stretch, where pixel is the original pixel value and new_pixel is the pixel value after stretching. is the minimum pixel value of the image, is the maximum pixel value of the image, and the enhanced image is displayed through the cv2.imshow function.

7. The tube tensile property detection system according to claim 1, characterized in that: The decision tree unit constructs the decision tree model through the CART algorithm, the decision tree unit includes a data set, the data set includes the material and wall thickness of the pipe, and the CART algorithm constructs the decision tree model by formula The Gini index is calculated as, where D is the dataset, K is the total number of categories, is the proportion of samples belonging to the kth class, and the data set D is divided into sub-datasets D1 and D2, where, is a set of samples whose feature A takes the value of a. is a set of samples whose feature A value is not a. Calculate the Gini index after partitioning, where represents the number of samples in the dataset D, represents the number of samples of the sub-dataset D1 obtained after division, It represents the number of samples of the sub-dataset D2 obtained after the division. For each of the sub-datasets, the sub-datasets are further divided to create new child nodes.

8. The tube tensile property detection system according to claim 7, characterized in that: The decision tree unit evaluates the decision tree model through accuracy, recall and F1-score indicators, where: , , , the number of samples of the decision tree model is adjusted accordingly according to the F1-score indicator evaluation result, and the calculation is repeated until the F1-score indicator evaluation result reaches the set indicator value.

9. The tube tensile property detection system according to claim 1, characterized in that: The central processing module controls the stretching and fixing module in real time in combination with the first control signal, the stretching electrical signal and the image signal, and the stretching and fixing module adjusts the stretching force and the bending force provided to the pipe in real time, specifically including: The central processing module drives the stretching and fixing module according to the first control signal, and the first control signal includes the initial control parameters of the stretching and fixing module corresponding to the pipe. During the stretching process, the central processing module calculates the actual stress according to the stretching electrical signal and compares it with the target stress set in the first control signal. If there is a deviation between the actual stress and the target stress, the central processing module controls the stretching and fixing module to make the actual stress equal to the target stress, and the central processing module analyzes the image signal and the stretching electrical signal in real time. If the pipe is deformed, the central processing module adjusts the stretching force and the bending force provided by the stretching and fixing module.

10. A device for detecting the tensile properties of a pipe, characterized in that: A system for detecting the tensile properties of a pipe comprising at least one of claims 1 to 9.

Images