Nondestructive testing system and method for differential pressure flowmeter welds based on ultrasonic phased array

Abstract

The present invention relates to the technical field of weld detection. The present invention discloses a nondestructive detection system and method for a differential pressure flowmeter weld based on an ultrasonic phased array. The system and method include determining a dynamic feature of a weld according to first flow data, determining a static feature of the weld according to first static data, and then determining a correlation value between the dynamic feature of the weld and the static feature of the weld, optimizing initial acoustic wave parameters according to a preset intelligent optimization algorithm and the correlation value to obtain optimal acoustic wave parameters, and finally obtaining second flow data and second static data based on the optimal acoustic wave parameters, inputting the second flow data and the second static data into a pre-constructed weld detection model to obtain a weld detection result. In the process of detecting the internal weld of a differential pressure flowmeter, the present invention determines a correlation value according to actual working conditions, and then optimizes the initial acoustic wave parameters according to the correlation value, so that the accuracy of weld detection is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of weld detection, and more specifically, to a differential pressure flowmeter weld nondestructive detection system and method based on ultrasonic phased array. Background Art

[0002] In industrial automation process instrumentation, differential pressure flow meters have two main uses: collecting flow data of various materials in the industrial automation process and measuring the cumulative amount of various materials in industrial production. Nondestructive testing (NDT) technology is a type of technology that detects and evaluates internal or surface defects without destroying the integrity of the material or structure. Nondestructive testing (NDT) technology includes ultrasonic testing, X-ray testing, magnetic particle testing, and visual testing. Ultrasonic phased array testing is an advanced nondestructive testing (NDT) technology that uses a phased array ultrasonic probe to change the propagation direction and focal position of the ultrasonic wave through electronic control, thereby achieving multi-angle scanning and precise detection of materials. In the existing technology, there is little content related to detecting the internal welds of differential pressure flow meters through ultrasonic phased arrays.

[0003] In the prior art, a Chinese patent with publication number CN118533972A provides an ultrasonic-based intelligent weld detection method and system. By collecting ultrasonic feedback information at the weld, and after verification, data conversion and feature extraction model training, a weld feature adjustment index is generated, and a matching analysis is performed with the preset threshold, and a weld detection and adjustment evaluation strategy is formulated based on the results.

[0004] Although the existing technology has achieved weld detection through ultrasound, there are still great challenges in detecting the internal welds of differential pressure flowmeters. Due to factors such as the complex internal structure of the differential pressure flowmeter, diverse material thicknesses, and possible fluid interference in the operating environment, ultrasonic parameters are particularly critical in the detection process. However, the existing technology makes it difficult to optimize ultrasonic parameters according to actual working conditions, resulting in difficulty in effectively improving the accuracy of weld detection.

[0005] In view of this, the present invention proposes a differential pressure flowmeter weld nondestructive testing system and method based on ultrasonic phased array to solve the above problems. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, the present invention provides a differential pressure flowmeter weld nondestructive testing system and method based on ultrasonic phased array.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] In the first aspect, a nondestructive testing method for a differential pressure flowmeter weld based on an ultrasonic phased array is provided, comprising:

[0009] Acquire first flow data and first static data based on initial acoustic wave parameters, determine dynamic characteristics of the weld according to the first flow data, and determine static characteristics of the weld according to the first static data;

[0010] Determine the correlation value between the dynamic characteristics of the weld and the static characteristics of the weld, optimize the initial acoustic wave parameters according to the preset intelligent optimization algorithm and the correlation value, and obtain the optimal acoustic wave parameters;

[0011] The second flow data and the second static data are acquired based on the optimal acoustic wave parameters, and the second flow data and the second static data are input into a pre-built weld detection model to obtain a weld detection result.

[0012] Further, the method for determining the dynamic characteristics of the weld according to the first flow data includes:

[0013] The corresponding image complexity is calculated based on the fluid image set, and the fluid membership is calculated according to the fluid characteristic data. The image complexity and fluid membership are used as dynamic features of the weld.

[0014] Furthermore, the method for calculating the corresponding image complexity based on the fluid image set includes:

[0015] The image gradient value corresponding to the fluid image set is calculated according to the preset Laplace operator, the contrast value of the fluid image set is obtained according to the preset gray-level co-occurrence matrix, and the edge density value of the fluid image set is obtained according to the preset Canny operator. The image gradient value, contrast value and edge density value are weightedly summed to obtain the image complexity.

[0016] Furthermore, the method for calculating the fluid membership according to the fluid characteristic data includes:

[0017] ;

[0018] Where FMD is the fluid membership, is the average fluid velocity, is the standard value of fluid velocity, is the average pressure gradient, is the standard value of pressure gradient, is the average value of fluid viscosity, is the standard value of fluid viscosity, is the standard deviation of fluid velocity, is the standard deviation of the pressure gradient, is the standard deviation of fluid viscosity, is a natural constant, , , are all weight factors.

[0019] Further, the method for determining the static characteristics of the weld according to the first static data includes:

[0020] An initial weld profile is obtained by performing polynomial fitting according to the first static data, and then the initial weld profile is adjusted according to spline curve fitting to obtain a target weld profile, and the target weld profile is used as a static feature of the weld.

[0021] Furthermore, the method for determining the correlation value between the dynamic characteristics of the weld and the static characteristics of the weld includes:

[0022] The static characteristics of the weld are input into the pre-built dynamic conversion model to obtain the target dynamic characteristics, and the Spearman coefficient between the dynamic characteristics of the weld and the target dynamic characteristics is calculated, and the Spearman coefficient is used as the correlation value.

[0023] Furthermore, the method for constructing the dynamic conversion model includes:

[0024] Acquire a sample data set, wherein the sample data set includes historical weld static features and historical target dynamic features;

[0025] Divide the sample data set into a sample training set and a sample test set, and build a regression network;

[0026] The historical weld static features in the sample training set are used as the input data of the regression network, and the historical target dynamic features in the sample training set are used as the output data of the regression network. The regression network is trained to obtain an initial regression network for predicting the initial target dynamic features.

[0027] The initial regression network is tested using a sample test set, and the initial regression network with an output smaller than a preset error value is used as the dynamic conversion model.

[0028] Furthermore, the intelligent optimization algorithm is a sparrow search algorithm, and the initial sound wave parameters are optimized according to the preset intelligent optimization algorithm and the correlation value. The method for obtaining the optimal sound wave parameters includes:

[0029] S201: Randomly generate a group of sparrow individuals, each sparrow individual represents an initial sound wave parameter, and randomly generate an initial position for each sparrow individual ;

[0030] S202: Construct the fitness function value of each sparrow individual :

[0031] ;

[0032] In the formula, is the correlation value corresponding to the initial sound wave parameters;

[0033] S203: Perform exploration operations on some sparrow individuals and update their initial sound wave parameters ;

[0034] ;

[0035] In the formula, For the A sparrow in The position of the generation, For the A sparrow in The position of the generation, is the maximum number of iterations, is a random angle, It is an exponential function with e as the base, where e is a natural constant;

[0036] S204: Perform a following operation on the remaining sparrow individuals and update their initial sound wave parameters ;

[0037] S205: Calculate the fitness function value of each sparrow individual after the update , update the current optimal solution;

[0038] S206: When the preset maximum number of iterations is reached Stop searching when , and output the initial sound wave parameters corresponding to the current optimal solution, which is the optimal sound wave parameters.

[0039] In a second aspect, a differential pressure flowmeter weld nondestructive testing system based on an ultrasonic phased array is provided, which is used to implement the above-mentioned differential pressure flowmeter weld nondestructive testing method based on an ultrasonic phased array, including:

[0040] A feature determination module: used for acquiring first flow data and first static data based on initial acoustic wave parameters, determining a dynamic feature of the weld according to the first flow data, and determining a static feature of the weld according to the first static data;

[0041] Parameter optimization module: used to determine the correlation value between the dynamic characteristics of the weld and the static characteristics of the weld, optimize the initial acoustic wave parameters according to the preset intelligent optimization algorithm and the correlation value, and obtain the optimal acoustic wave parameters;

[0042] Weld detection module: used to obtain the second flow data and the second static data based on the optimal acoustic wave parameters, input the second flow data and the second static data into the pre-built weld detection model, and obtain the weld detection result.

[0043] In a third aspect, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned ultrasonic phased array-based differential pressure flowmeter weld nondestructive testing method when executing the computer program.

[0044] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed, the above-mentioned ultrasonic phased array-based differential pressure flowmeter weld nondestructive testing method is implemented.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] The present invention first determines the dynamic characteristics of the weld according to the first flow data, determines the static characteristics of the weld according to the first static data, and then determines the correlation value between the dynamic characteristics of the weld and the static characteristics of the weld, optimizes the initial acoustic wave parameters according to the preset intelligent optimization algorithm and the correlation value, obtains the optimal acoustic wave parameters, and finally obtains the second flow data and the second static data based on the optimal acoustic wave parameters, inputs the second flow data and the second static data into the pre-built weld detection model, and obtains the weld detection result. In the process of detecting the internal weld of the differential pressure flowmeter, the present invention determines the correlation value according to the actual working conditions, and then optimizes the initial acoustic wave parameters according to the correlation value, so that the accuracy of the weld detection is effectively improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a schematic diagram of the process of the nondestructive testing method of the differential pressure flowmeter weld based on the ultrasonic phased array in the present invention;

[0048] Figure 2 It is a structural schematic diagram of a differential pressure flowmeter weld nondestructive testing system based on ultrasonic phased array in the present invention;

[0049] Figure 3 It is a flow chart of a method for calculating corresponding image complexity based on a fluid image set in the present invention;

[0050] Figure 4 It is a schematic diagram of the process of determining the correlation value between the dynamic characteristics of the weld and the static characteristics of the weld in the present invention. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] Example 1

[0053] See also Figure 1 As shown, this embodiment discloses a nondestructive testing method for a differential pressure flowmeter weld based on an ultrasonic phased array, comprising:

[0054] S10: acquiring first flow data and first static data based on initial acoustic wave parameters, determining dynamic characteristics of the weld according to the first flow data, and determining static characteristics of the weld according to the first static data;

[0055] In this embodiment, the initial acoustic wave parameters refer to the key detection parameters used in the ultrasonic detection process inside the differential pressure flowmeter. The initial acoustic wave parameters include acoustic wave frequency, beam angle, scanning speed, and focusing depth, etc. It can be understood that the initial acoustic wave parameters are pre-set by technical personnel in this field based on expert experience.

[0056] In an optional embodiment, the first flow data refers to data related to the fluid state near the weld inside the differential pressure flowmeter. The first flow data includes a fluid image set and fluid property data. The fluid image set refers to an image set of the fluid near the weld obtained by ultrasonic phased array imaging technology. The fluid property data refers to important physical and dynamic parameters obtained by ultrasonic phased array technology that reflect the fluid state near the weld. The fluid property data includes fluid velocity values, pressure gradient values, and fluid viscosity values. Taking the fluid velocity value as an example, the larger the fluid velocity value, the more likely the type of weld is a butt weld or a pipe girth weld. The butt weld is used to connect two parallel plates or pipes. If the welding is incomplete, it may cause fluid leakage or abnormal increase in local flow rate. Similarly, differential pressure flowmeters often involve pipe girth welds, which are used to connect the ends of pipes. Due to the presence of pipe girth welds, the fluid velocity value increases.

[0057] The method for determining the dynamic characteristics of a weld according to first flow data comprises:

[0058] The corresponding image complexity is calculated based on the fluid image set, and the fluid membership is calculated according to the fluid characteristic data. The image complexity and fluid membership are used as dynamic features of the weld.

[0059] like Figure 3 As shown, the method for calculating the corresponding image complexity based on the fluid image set includes:

[0060] The image gradient value corresponding to the fluid image set is calculated according to the preset Laplace operator, the contrast value of the fluid image set is obtained according to the preset gray-level co-occurrence matrix, and the edge density value of the fluid image set is obtained according to the preset Canny operator. The image gradient value, contrast value and edge density value are weightedly summed to obtain the image complexity.

[0061] It should be noted that the Laplace operator refers to a mathematical tool based on the second-order derivative of the image, which is used to detect edges and mutation areas in the image, and to detect rapidly changing areas in the fluid image set, such as the edges and mutations of the fluid near the weld. The gray-level co-occurrence matrix refers to a tool for analyzing image texture features. It can be used to analyze the texture characteristics of the fluid near the weld and reveal subtle changes in the fluid state by statistically analyzing the gray value relationship between adjacent pixels in the fluid image set. The Canny operator refers to an edge detection algorithm that can extract edge information from the fluid image set, and is used to accurately locate changes in the fluid boundary near the weld.

[0062] In this embodiment, image complexity is used to characterize the dynamic change characteristics of the fluid near the weld. For example, when the weld is a butt weld, the defects of the butt weld include incomplete penetration, cracks or holes. Taking the defects of incomplete penetration and holes as examples, when the defect is incomplete penetration, it is manifested in the fluid image set as a sudden change in grayscale near the weld boundary, but the change area is concentrated and the amplitude is small, the contrast change is concentrated in the weld edge area, and the overall texture difference is not significant, etc. The corresponding image gradient value, contrast value and edge density value are relatively small; and when the defect is a hole, it is manifested in the fluid image set as a strong grayscale change in the hole edge area, and the area range is large, the edge distribution presents a regular ring or locally concentrated distribution, etc., so the corresponding image gradient value, contrast value and edge density value are relatively large. Therefore, the dynamic change characteristics of the fluid near the weld can be quantitatively analyzed by image complexity, so as to distinguish different types of defects.

[0063] Methods for calculating fluid membership based on fluid property data include:

[0064] ;

[0065] Where FMD is the fluid membership, is the average fluid velocity, is the standard value of fluid velocity, is the average pressure gradient, is the standard value of pressure gradient, is the average value of fluid viscosity, is the standard value of fluid viscosity, is the standard deviation of fluid velocity, is the standard deviation of the pressure gradient, is the standard deviation of fluid viscosity, is a natural constant, , , are all weight factors.

[0066] It can be understood that the fluid characteristic data may include multiple groups of sub-characteristic data, each group of sub-characteristic data includes fluid velocity values, pressure gradient values ​​and fluid viscosity values. Taking the fluid velocity value as an example, the average fluid velocity is calculated based on several fluid velocity values, and the standard deviation of the fluid velocity can also be calculated. The standard value of the fluid velocity refers to the fluid velocity when there are no defects in the weld. The same principle applies to the pressure gradient value and the fluid viscosity value. The fluid membership characterizes the severity of the defects in the weld. Specifically, by quantifying the dynamic changes of the fluid velocity value, the pressure gradient value and the fluid viscosity value at the weld, the influence of the weld defects on the fluid state is reflected. The larger the fluid membership, the greater the severity of the defects in the weld and the more significant the influence of the defects on the fluid state.

[0067] In this embodiment, the first static data characterizes the geometric shape characteristics and structural integrity. The first static data includes the weld aspect ratio, the weld width-to-depth ratio and the weld cross-sectional curvature. The weld cross-sectional curvature refers to the curvature of each point on the weld cross-sectional contour line, which is used to quantify the degree of curvature of the cross-sectional shape. The weld cross-sectional curvature reflects the smoothness and regularity of the cross-sectional contour and is an important parameter for evaluating the weld geometry. It can be understood that the weld-related data can be determined by analyzing the amplitude and arrival time of the ultrasonic reflection signal, and this embodiment does not go into details about this.

[0068] The method for determining the static characteristics of a weld according to first static data comprises:

[0069] An initial weld profile is obtained by performing polynomial fitting according to the first static data, and then the initial weld profile is adjusted according to spline curve fitting to obtain a target weld profile, and the target weld profile is used as a static feature of the weld.

[0070] It can be understood that, for example, the first static data includes the weld aspect ratio, the weld width-to-depth ratio, and the weld cross-sectional curvature, then polynomial fitting refers to fitting the first static data of the weld (such as the weld aspect ratio, the weld width-to-depth ratio, the weld cross-sectional curvature, etc.) using a polynomial function to generate a preliminary weld profile curve. The core of this process is to use a mathematical model to capture the geometric characteristics and overall trends of the weld, providing a basis for the subsequent fine adjustment of the spline curve. Spline curve fitting refers to fitting the weld profile data using a piecewise polynomial to generate a smoother and more accurate target weld profile. Compared with polynomial fitting, spline curve fitting can better handle local details and avoid the oscillation problem of high-order polynomials. The above polynomial fitting and spline curve fitting can be implemented by MATLAB software, and this embodiment will not go into too much detail.

[0071] In this embodiment, the dynamic feature extraction of the first flow data and the static feature extraction of the first static data complement each other. The dynamic features of the weld reflect the impact of the weld on the fluid state, and the static features of the weld characterize the weld geometry and integrity. The combination of the two can provide a comprehensive quality assessment.

[0072] S20: determining the correlation value between the dynamic characteristics of the weld and the static characteristics of the weld, optimizing the initial acoustic wave parameters according to the preset intelligent optimization algorithm and the correlation value, and obtaining the optimal acoustic wave parameters;

[0073] Among them, Figure 4 As shown, the method for determining the correlation value between the dynamic characteristics of the weld and the static characteristics of the weld includes:

[0074] The static characteristics of the weld are input into the pre-built dynamic conversion model to obtain the target dynamic characteristics, and the Spearman coefficient between the dynamic characteristics of the weld and the target dynamic characteristics is calculated, and the Spearman coefficient is used as the correlation value.

[0075] The construction method of the dynamic conversion model includes:

[0076] Acquire a sample data set, wherein the sample data set includes historical weld static features and historical target dynamic features;

[0077] Divide the sample data set into a sample training set and a sample test set, and build a regression network;

[0078] The historical weld static features in the sample training set are used as the input data of the regression network, and the historical target dynamic features in the sample training set are used as the output data of the regression network. The regression network is trained to obtain an initial regression network for predicting the initial target dynamic features.

[0079] The initial regression network is tested using a sample test set, and the output of the initial regression network that is less than the preset error value is used as the dynamic conversion model. The initial regression network is a deep neural network model.

[0080] It should be noted that, from the above content, it can be seen that the static characteristics of the weld are used to represent the geometric shape characteristics and overall contour trends of the weld, and the dynamic characteristics of the weld are used to represent the dynamic change characteristics of the fluid near the weld. The static characteristics of the weld and the dynamic characteristics of the weld describe the weld quality from two dimensions: the geometric state and functional performance of the weld. For example, taking the appearance of hole defects in butt welds as an example, the dynamic characteristics of the weld are manifested as a high degree of fluid membership and a large increase in image complexity. The static characteristics of the weld are manifested as local depressions or discontinuities, and the overall contour is not smooth. Therefore, technical personnel in this field can pre-establish the connection between the dynamic characteristics of the weld and the static characteristics of the weld through experiments or expert experience, and pre-train the dynamic conversion model, so that the static characteristics of the weld can be used to predict the target dynamic characteristics, and then match and verify them with the actual detected dynamic characteristics of the weld to determine whether the initial acoustic wave parameters are the optimal acoustic wave parameters.

[0081] It should be added that the Spearman coefficient refers to a rank-based non-parametric statistical method used to evaluate the monotonic correlation between two variables. Its value range is -1 to +1. When the Spearman coefficient is +1, it indicates that the dynamic characteristics of the weld are completely positively correlated with the target dynamic characteristics. When the Spearman coefficient is -1, it indicates that the dynamic characteristics of the weld are completely negatively correlated with the target dynamic characteristics. The calculation of the Spearman coefficient is a prior art and will not be elaborated on in this embodiment.

[0082] The intelligent optimization algorithm is a sparrow search algorithm. The initial sound wave parameters are optimized according to the preset intelligent optimization algorithm and the correlation value. The method for obtaining the optimal sound wave parameters includes:

[0083] S201: Randomly generate a group of sparrow individuals, each sparrow individual represents an initial sound wave parameter, and randomly generate an initial position for each sparrow individual ;

[0084] Among them, sparrow individuals are used to simulate the possible value range of the initial sound wave parameters, and by randomly generating initial positions, it is ensured that these individuals are evenly distributed in the parameter space.

[0085] S202: Construct the fitness function value of each sparrow individual :

[0086] ;

[0087] In the formula, is the correlation value corresponding to the initial sound wave parameters.

[0088] It can be understood that the correlation value measures the degree of match between the initial acoustic wave parameters and the dynamic and static characteristics of the weld, and is used to evaluate the quality of the parameters. By constructing a fitness function, the detection effect of each set of initial acoustic wave parameters is quantified, providing a basis for the subsequent screening of high-quality individuals.

[0089] S203: Perform exploration operations on some sparrow individuals and update their initial sound wave parameters ;

[0090] ;

[0091] In the formula, For the A sparrow in The position of the generation, For the A sparrow in The position of the generation, is the maximum number of iterations, is a random angle, It is an exponential function with e as base, where e is a natural constant.

[0092] S204: Perform a following operation on the remaining sparrow individuals and update their initial sound wave parameters ;

[0093] S205: Calculate the fitness function value of each sparrow individual after the update , update the current optimal solution;

[0094] S206: When the preset maximum number of iterations is reached Stop searching when , and output the initial sound wave parameters corresponding to the current optimal solution, which is the optimal sound wave parameters.

[0095] In S204, the following operation is performed on the remaining sparrow individuals to update their initial sound wave parameters The methods include:

[0096] ;

[0097] In the formula, is a random number, is the location of the current optimal solution.

[0098] In this embodiment, the above steps indicate that the individual sparrows follow and update based on the current optimal solution. In this way, the individual sparrows will gradually move closer to the optimal solution and converge to a better solution.

[0099] This embodiment obtains the dynamic characteristics and static characteristics of the weld to respectively describe the geometric shape characteristics and overall contour trend of the weld (static characteristics of the weld) and the influence of the weld on the fluid state (dynamic characteristics of the weld). Based on the correlation value between the static characteristics and the dynamic characteristics of the weld, the degree of matching between them can be effectively judged, thereby ensuring the scientific nature of the optimization of the initial acoustic wave parameters.

[0100] S30: acquiring second flow data and second static data based on the optimal acoustic wave parameters, and inputting the second flow data and the second static data into a pre-built weld detection model to obtain a weld detection result.

[0101] It can be understood that the second flow data has the same meaning as the first flow data. The second flow data includes a fluid image set and fluid characteristic data, and the second static data includes the weld aspect ratio, weld width-to-depth ratio and weld cross-sectional curvature.

[0102] It should be supplemented that the weld inspection results may be weld defect types and weld defect levels. For example, weld defect types may be incomplete penetration, cracks, and holes, etc., and weld defect levels may include first level, second level, and third level, and the first level > second level > third level.

[0103] The construction method of the weld detection model includes:

[0104] Obtain P groups of training data, where P is a positive integer greater than 1, and the training data includes historical second flow data, historical second static data, and historical weld inspection results;

[0105] The historical second flow data, the historical second static data and the historical weld inspection results are used as sample sets, the sample sets are divided into training sets and test sets, and a classifier is constructed;

[0106] The historical second flow data and the historical second static data in the training set are used as input data, and the historical weld inspection results in the training set are used as output data, and the classifier is trained to obtain an initial classifier;

[0107] The initial classifier is tested using the test set, and a classifier that meets the preset accuracy is output as the weld detection model. The classifier is preferably a naive Bayes model or a support vector machine model.

[0108] It should be noted that the second static data also characterizes the geometric shape characteristics and structural integrity, and is therefore mainly used for the classification of weld defect types, while the second flow data characterizes the impact of the weld on the fluid state, and is used to describe the dynamic change characteristics of the weld and to assist in further determination of the weld defect type and weld defect grade.

[0109] In this embodiment, the dynamic characteristics of the weld are first determined according to the first flow data, the static characteristics of the weld are determined according to the first static data, and then the correlation value between the dynamic characteristics of the weld and the static characteristics of the weld is determined. The initial acoustic wave parameters are optimized according to the preset intelligent optimization algorithm and the correlation value to obtain the optimal acoustic wave parameters. Finally, the second flow data and the second static data are obtained based on the optimal acoustic wave parameters, and the second flow data and the second static data are input into the pre-built weld detection model to obtain the weld detection result. In the process of detecting the internal weld of the differential pressure flowmeter, this embodiment determines the correlation value according to the actual working conditions, and then optimizes the initial acoustic wave parameters according to the correlation value, so that the accuracy of weld detection is effectively improved.

[0110] Example 2

[0111] See also Figure 2 As shown, based on the same inventive concept, this embodiment discloses a differential pressure flowmeter weld nondestructive testing system based on an ultrasonic phased array. For details not provided in this embodiment, please refer to the description of the relevant parts in Embodiment 1. The system includes:

[0112] A feature determination module: used for acquiring first flow data and first static data based on initial acoustic wave parameters, determining a dynamic feature of the weld according to the first flow data, and determining a static feature of the weld according to the first static data;

[0113] In this embodiment, the initial acoustic wave parameters refer to the key detection parameters used in the ultrasonic detection process inside the differential pressure flowmeter. The initial acoustic wave parameters include acoustic wave frequency, beam angle, scanning speed, and focusing depth, etc. It can be understood that the initial acoustic wave parameters are pre-set by technical personnel in this field based on expert experience.

[0114] In an optional embodiment, the first flow data refers to data related to the fluid state near the weld inside the differential pressure flowmeter. The first flow data includes a fluid image set and fluid property data. The fluid image set refers to an image set of the fluid near the weld obtained by ultrasonic phased array imaging technology. The fluid property data refers to important physical and dynamic parameters obtained by ultrasonic phased array technology and reflecting the fluid state near the weld. The fluid property data includes fluid velocity values, pressure gradient values, and fluid viscosity values. Taking the fluid velocity value as an example, the larger the fluid velocity value, the more likely the type of weld is a butt weld or a pipe girth weld. The butt weld is used to connect two parallel plates or pipes. If the welding is incomplete, it may cause fluid leakage or abnormal increase in local flow rate. Similarly, differential pressure flowmeters often involve pipe girth welds, which are used to connect the ends of pipes. Due to the presence of pipe girth welds, the fluid velocity value increases.

[0115] The method for determining the dynamic characteristics of a weld according to first flow data comprises:

[0116] The corresponding image complexity is calculated based on the fluid image set, and the fluid membership is calculated according to the fluid characteristic data. The image complexity and fluid membership are used as dynamic features of the weld.

[0117] Methods for calculating the corresponding image complexity based on a fluid image set include:

[0118] The image gradient value corresponding to the fluid image set is calculated according to the preset Laplace operator, the contrast value of the fluid image set is obtained according to the preset gray-level co-occurrence matrix, and the edge density value of the fluid image set is obtained according to the preset Canny operator. The image gradient value, contrast value and edge density value are weightedly summed to obtain the image complexity.

[0119] Methods for calculating fluid membership based on fluid property data include:

[0120] ;

[0121] Where FMD is the fluid membership, is the average fluid velocity, is the standard value of fluid velocity, is the average pressure gradient, is the standard value of pressure gradient, is the average value of fluid viscosity, is the standard value of fluid viscosity, is the standard deviation of fluid velocity, is the standard deviation of the pressure gradient, is the standard deviation of fluid viscosity, is a natural constant, , , are all weight factors.

[0122] The method for determining the static characteristics of a weld according to first static data comprises:

[0123] An initial weld profile is obtained by performing polynomial fitting according to the first static data, and then the initial weld profile is adjusted according to spline curve fitting to obtain a target weld profile, and the target weld profile is used as a static feature of the weld.

[0124] Parameter optimization module: used to determine the correlation value between the dynamic characteristics of the weld and the static characteristics of the weld, optimize the initial acoustic wave parameters according to the preset intelligent optimization algorithm and the correlation value, and obtain the optimal acoustic wave parameters;

[0125] Among them, the method for determining the correlation value between the dynamic characteristics of the weld and the static characteristics of the weld includes:

[0126] The static characteristics of the weld are input into the pre-built dynamic conversion model to obtain the target dynamic characteristics, and the Spearman coefficient between the dynamic characteristics of the weld and the target dynamic characteristics is calculated, and the Spearman coefficient is used as the correlation value.

[0127] The construction method of the dynamic conversion model includes:

[0128] Acquire a sample data set, wherein the sample data set includes historical weld static features and historical target dynamic features;

[0129] Divide the sample data set into a sample training set and a sample test set, and build a regression network;

[0130] The historical weld static features in the sample training set are used as the input data of the regression network, and the historical target dynamic features in the sample training set are used as the output data of the regression network. The regression network is trained to obtain an initial regression network for predicting the initial target dynamic features.

[0131] The initial regression network is tested using a sample test set, and the output of the initial regression network that is less than the preset error value is used as the dynamic conversion model. The initial regression network is a deep neural network model.

[0132] Weld detection module: used to obtain the second flow data and the second static data based on the optimal acoustic wave parameters, input the second flow data and the second static data into the pre-built weld detection model, and obtain the weld detection result.

[0133] It can be understood that the second flow data has the same meaning as the first flow data. The second flow data includes a fluid image set and fluid characteristic data, and the second static data includes the weld aspect ratio, weld width-to-depth ratio and weld cross-sectional curvature.

[0134] It should be supplemented that the weld inspection results may be weld defect types and weld defect levels. For example, weld defect types may be incomplete penetration, cracks, and holes, etc., and weld defect levels may include first level, second level, and third level, and the first level > second level > third level.

[0135] Example 3

[0136] This embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the nondestructive detection method for differential pressure flowmeter welds based on ultrasonic phased array provided by the above methods is implemented.

[0137] Since the electronic device introduced in this embodiment is an electronic device used to implement the non-destructive testing method for differential pressure flowmeter welds based on ultrasonic phased array in the embodiment of the present application, based on the non-destructive testing method for differential pressure flowmeter welds based on ultrasonic phased array introduced in the embodiment of the present application, technical personnel in this field can understand the specific implementation mode of the electronic device of this embodiment and its various variations, so how the electronic device implements the method in the embodiment of the present application is no longer introduced in detail. As long as technical personnel in this field implement the electronic device used in the non-destructive testing method for differential pressure flowmeter welds based on ultrasonic phased array in the embodiment of the present application, it belongs to the scope of protection of this application.

[0138] Example 4

[0139] The present embodiment discloses a computer-readable storage medium, on which a computer program is stored. When the computer program is executed, the above-mentioned ultrasonic phased array-based differential pressure flowmeter weld nondestructive testing method is implemented.

[0140] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters, weights and thresholds in the formula are set by technicians in this field according to actual conditions.

[0141] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired network or a wireless network. The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media sets. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD) or a semiconductor medium. The semiconductor medium may be a solid-state hard disk.

[0142] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0143] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0144] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only one, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0145] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0146] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0147] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0148] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A nondestructive testing method for differential pressure flowmeter welds based on ultrasonic phased array, characterized in that: include: Acquire first flow data and first static data based on initial acoustic wave parameters, determine dynamic characteristics of the weld according to the first flow data, and determine static characteristics of the weld according to the first static data; Determine the correlation value between the dynamic characteristics of the weld and the static characteristics of the weld, optimize the initial acoustic wave parameters according to the preset intelligent optimization algorithm and the correlation value, and obtain the optimal acoustic wave parameters; Acquire second flow data and second static data based on the optimal acoustic wave parameters, input the second flow data and the second static data into a pre-built weld detection model, and obtain a weld detection result; The first flow data includes a fluid image set and fluid characteristic data. The method for determining the dynamic characteristics of the weld according to the first flow data includes: The corresponding image complexity is calculated based on the fluid image set, the fluid membership is calculated according to the fluid characteristic data, and the image complexity and fluid membership are used as the dynamic characteristics of the weld; The method for calculating the corresponding image complexity based on the fluid image set includes: The image gradient value corresponding to the fluid image set is calculated according to the preset Laplace operator, the contrast value of the fluid image set is obtained according to the preset gray-level co-occurrence matrix, and the edge density value of the fluid image set is obtained according to the preset Canny operator. The image gradient value, contrast value and edge density value are weightedly summed to obtain the image complexity.

2. The nondestructive testing method for differential pressure flowmeter welds based on ultrasonic phased array according to claim 1 is characterized in that: The method for calculating the fluid membership degree according to the fluid characteristic data comprises: FMD= ; Where FMD is the fluid membership, is the average fluid velocity, is the standard value of fluid velocity, is the average pressure gradient, is the standard value of pressure gradient, is the average value of fluid viscosity, is the standard value of fluid viscosity, is the standard deviation of fluid velocity, is the standard deviation of the pressure gradient, is the standard deviation of fluid viscosity, is a natural constant, , , are all weight factors.

3. The nondestructive testing method for differential pressure flowmeter welds based on ultrasonic phased array according to claim 1 is characterized in that: The method for determining the static characteristics of a weld according to first static data comprises: An initial weld profile is obtained by performing polynomial fitting according to the first static data, and then the initial weld profile is adjusted according to spline curve fitting to obtain a target weld profile, and the target weld profile is used as a static feature of the weld.

4. The nondestructive testing method for differential pressure flowmeter welds based on ultrasonic phased array according to claim 3 is characterized in that: The method for determining the correlation value between the dynamic characteristics of the weld and the static characteristics of the weld includes: The static characteristics of the weld are input into the pre-built dynamic conversion model to obtain the target dynamic characteristics, and the Spearman coefficient between the dynamic characteristics of the weld and the target dynamic characteristics is calculated, and the Spearman coefficient is used as the correlation value.

5. The nondestructive testing method for differential pressure flowmeter welds based on ultrasonic phased array according to claim 4 is characterized in that: The method for constructing the dynamic conversion model includes: Acquire a sample data set, wherein the sample data set includes historical weld static features and historical target dynamic features; Divide the sample data set into a sample training set and a sample test set, and build a regression network; The historical weld static features in the sample training set are used as the input data of the regression network, and the historical target dynamic features in the sample training set are used as the output data of the regression network. The regression network is trained to obtain an initial regression network for predicting the initial target dynamic features. The initial regression network is tested using a sample test set, and the initial regression network with an output smaller than a preset error value is used as the dynamic conversion model.

6. The nondestructive testing method for differential pressure flowmeter welds based on ultrasonic phased array according to claim 5 is characterized in that: The intelligent optimization algorithm is a sparrow search algorithm. The initial sound wave parameters are optimized according to the preset intelligent optimization algorithm and the correlation value. The method for obtaining the optimal sound wave parameters includes: S201: Randomly generate a group of sparrow individuals, each sparrow individual represents an initial sound wave parameter, and randomly generate an initial position for each sparrow individual ; S202: Construct the fitness function value of each sparrow individual : = ； In the formula, is the correlation value corresponding to the initial sound wave parameters; S203: Perform exploration operations on some sparrow individuals and update their initial sound wave parameters ; = ； In the formula, For the A sparrow in the The position of the generation, For the A sparrow in the The position of the generation, is the maximum number of iterations, is a random angle, It is an exponential function with e as base, where e is a natural constant; S204: Perform a following operation on the remaining sparrow individuals and update their initial sound wave parameters ; S205: Calculate the fitness function value of each sparrow individual after the update , update the current optimal solution; S206: When the preset maximum number of iterations is reached Stop searching when , and output the initial sound wave parameters corresponding to the current optimal solution, which is the optimal sound wave parameters.

7. A differential pressure flowmeter weld nondestructive testing system based on ultrasonic phased array, which is used to implement the differential pressure flowmeter weld nondestructive testing method based on ultrasonic phased array as described in any one of claims 1 to 6, characterized in that: include: A feature determination module: used for acquiring first flow data and first static data based on initial acoustic wave parameters, determining a dynamic feature of the weld according to the first flow data, and determining a static feature of the weld according to the first static data; Parameter optimization module: used to determine the correlation value between the dynamic characteristics of the weld and the static characteristics of the weld, optimize the initial acoustic wave parameters according to the preset intelligent optimization algorithm and the correlation value, and obtain the optimal acoustic wave parameters; Weld detection module: used to obtain the second flow data and the second static data based on the optimal acoustic wave parameters, input the second flow data and the second static data into the pre-built weld detection model, and obtain the weld detection result.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the ultrasonic phased array-based differential pressure flowmeter weld nondestructive testing method described in any one of claims 1 to 6 is implemented.

Images