A method and experimental system for compensating current density stress of wire bundle electrodes in pipeline steel welding joints

Abstract

The present invention discloses a method and experimental system for compensating current density stress of a wire bundle electrode of a pipeline steel welding joint. The method comprises: measuring a current density distribution matrix and a position matrix of a wire bundle electrode; changing a loading stress and measuring a sequence of current density distribution matrices of a working electrode; constructing a current density stress compensation data set, wherein the data set comprises an input data set and an output data set; establishing an artificial neural network; dividing the input data set and the output data set into three parts: a training set, a test set and a validation set, and performing normalization processing; completing the establishment of an artificial neural network model; establishing a current density stress compensation model of a pipeline steel welding joint under stress based on the artificial neural network model; measuring the current density of a wire bundle electrode of a welding joint under stress-free conditions, and substituting the current density measurement value, its corresponding position and the loading stress amplitude into the stress compensation model to obtain the current density of a certain area of ​​the welding joint under the corresponding loading stress conditions.

Description

Technical Field

[0001] The invention relates to a wire bundle electrode current density stress compensation method, in particular to a wire bundle electrode current density stress compensation method for a pipeline steel welding joint and an experimental system. Background Art

[0002] Wire beam electrodes have significant advantages such as ease of use and the ability to reflect micro-area electrochemical corrosion processes. They are widely used in multiple electrochemical corrosion research fields, including pipeline steel corrosion, and have achieved a series of significant research results. They have become an important testing tool for studying micro-area electrochemical corrosion.

[0003] Both internal and external corrosion of pipeline steel are the result of the synergistic effect of stress and corrosion, and some types of corrosion will be significantly aggravated due to the coupling effect of stress. Therefore, in the testing and research of pipeline steel corrosion, the stress generated by the operating internal pressure is an extremely important and non-negligible external factor and must be taken into consideration. At present, due to the material limitations of its structural characteristics, the wire beam electrode has the disadvantage of being difficult to reflect the force-electrochemical interaction process, which is the main obstacle to its further application in the study of pipeline steel corrosion. Other micro-area electrochemical measurement methods, such as scanning vibration electrodes, are cumbersome to operate, the test equipment is expensive and the synchronization is poor. In order to break through the application barriers of wire beam electrodes in the force-electrochemical interaction environment, the present invention proposes a current density stress compensation method for wire beam electrodes of pipeline steel welded joints, which is used to achieve stress compensation of current density measurement results under stress-free conditions, and further realize the indirect measurement of the current density of the metal surface under stress by the wire beam electrode. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for compensating the current density stress of the wire bundle electrode of pipeline steel welded joints. This method is convenient for measurement. By indirectly compensating the current density distribution of the wire bundle electrode under stress-free conditions, the surface current density of the welded joint under stress load is obtained, accurately reflecting the current density distribution on the surface of the pipeline steel welded joint under the combined action of corrosion and stress load.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is:

[0006] A method for compensating current density stress of a wire bundle electrode of a pipeline steel welding joint comprises the following steps:

[0007] Step 1: Place experimental system 1 and experimental system 2 in the same electrolyte solution concentration, ambient temperature and humidity, and measure the current density distribution matrix I of the wire bundle electrode of experimental system 1. w =[I 1,1 1 (s 1,11 ), I 1,2 1 (s 1,2 1 ),I 1,3 1 (s 1,3 1 ),…,I 1,n 1 (s 1,n 1 );…,I k,1 1 (s k,1 1 ),I k,2 1 (s k,2 1 ),I k,3 1 (s k,3 1 ),…, I k,n 1 (s k,n 1 );…,I n,1 1 (s n,1 1 ),I n,2 1 (s n,2 1 ),I n,3 1 (s n,3 1 ),…,I n,n 1 (s n,n 1 )] and the position matrix S of the wire beam electrode w =[s 1,1 1 ,s 1,2 1 ,s 1,3 1 ,…,s 1,n 1 ;…,s k,1 1 ,s k,2 1 ,s k,3 1 ,…,s k,n 1 ;…,s n,1 1 ,s n,2 1 ,s n,3 1 ,…, sn,n 1 ]; change the loading stress [F1,F2,F3,…,F m ,…,F n ], measuring the working electrode current density distribution matrix sequence of experimental system 2 {I d (F1),I d (F2),…,I d (F m ),…,I d (F n )}, the position matrix of the working electrode is the same as that of the wire bundle electrode, k is any number from 1 to n, and m is any number from 1 to n;

[0008] Step 2: Construct the current density stress compensation data set, which includes the input data set and the output data set. The input data set includes: the position information of the wire bundle electrode array and the current density measurement point of the working electrode [s 1,1 1 ,s 1,2 1 ,s 1,3 1 ,…,s 1,n 1 ;…,s k,1 1 ,s k,2 1 ,s k,3 1 ,…,s k,n 1 ;…,s n,1 1 ,s n,2 1 ,s n,3 1 ,…,s n,n 1 ]、Loading stress of working electrode of welding joint [F1,F2,F3,…,F m ,…,F n ] and the current density distribution information of the wire bundle electrode array of the welded joint under stress-free conditions I w =[I 1,1 1 (s 1,1 1 ),I 1,2 1 (s 1,2 1 ),I 1,3 1 (s 1,3 1 ),…,I 1,n 1 (s 1,n1 );…, I k,1 1 (s k,1 1 ),I k,2 1 (s k,2 1 ),I k,3 1 (s k,3 1 ),…,I k,n 1 (s k,n 1 );…,I n,1 1 (s n,1 1 ),I n,2 1 (s n,2 1 ),I n,3 1 (s n,3 1 ),…, I n,n 1 (s n,n 1 )], the output data set is the current density distribution information of the working electrode of the welding joint {I d (F1),I d (F2),…, I d (F m ),…,I d (F n )};

[0009] Step 3: The position s of the current density measurement point between the wire bundle electrode array and the working electrode k,n 1 , working electrode loading stress F m and the current density I of the wire bundle electrode array without stress k,n 1 (s k,n 1 ) as the input variable, and the current density under the working electrode stress as the output variable I k,n 2 (F m ,s k,n 2 ), establish artificial neural network;

[0010] Step 4: Based on the dataset created in step 2, divide the input dataset and output dataset into three parts: training set, test set, and validation set in a ratio of 70%:15%:15%, and normalize the training set, test set, and validation set.

[0011] Step 5: Train, verify, and test the artificial neural network to meet the following conditions to complete the establishment of the artificial neural network model f(·):

[0012]

[0013] Among them, I k，n 2 (F m , s k，n 2 ) is the current density test value at a certain position of the welding joint under stress, is the predicted value of current density at a certain position of the welding joint, and ε is the preset convergence accuracy of artificial neural network training;

[0014] Step 6: Based on the artificial neural network model f(·), a current density stress compensation model for pipeline steel weld joints under stress is established. The output of the current density stress compensation model for the wire beam electrode of the weld joint is:

[0015] I k,n 2 =f(I k,n 1 ,F m ,s k,n )

[0016] Step 7: Measure the current density of the wire bundle electrode of the weld joint under stress-free conditions, and substitute the current density measurement value and its corresponding position and loading stress amplitude into the stress compensation model to obtain the current density of a certain area of ​​the weld joint under the corresponding loading stress.

[0017] Furthermore, in step 5, the quantum particle swarm algorithm is used to optimize the neural network weight and threshold update process. The optimization objective of the quantum particle swarm algorithm is expressed in the form of a vector:

[0018] [ω ih (1,1),...,ω ih (3,x),θ 1,1 ,...,θ 1,x ,ω ho (1,1),...,ω ho (x,1),θ 2,1 ]

[0019] Among them, ω ih(1,1),…,ω ih (3,1),…,ω ih (1,x),…,ω ih (3,x) is the input weight between the input layer and the hidden layer, ω ho (1,1),ω ho (2,1),…,ω ho (x,1) is the output weight between the hidden layer and the output layer, θ 1,1 ,θ 1,2 ,…,θ 1,x is the hidden layer neuron threshold, θ 2,1 is the output layer neuron threshold.

[0020] Another technical solution adopted by the present invention is: an experimental system for a wire bundle electrode current density stress compensation method for a pipeline steel welding joint, comprising: an experimental system 1 and an experimental system 2;

[0021] Experimental system 1 includes: a multi-channel electrochemical workstation, a wire beam electrode probe, an auxiliary electrode, a reference electrode, a wire beam electrode test module, a corrosion electrolytic cell, a multi-channel high-speed switching device and a host computer;

[0022] Experimental system 2 includes: an electrochemical workstation, a welding joint working electrode, an auxiliary electrode, a reference electrode, a scanning vibration electrode, a three-dimensional motion control device, a corrosion electrolytic cell, an electrometer, a stress loading device, a tension sensor, and a host computer. The stress loading device is loaded at both ends of the working electrode, and the tensile stress load on the working electrode is measured by the tension sensor.

[0023] Furthermore, the wire bundle electrode probe passes through 200 cross-sectional areas of 1mm 2 The array of wire beam electrodes is encapsulated in epoxy resin. The spacing between each wire beam electrode is 1mm. The wire beam electrode is made of metal in the welding zone, base material zone and heat-affected zone respectively. The ratio of the wire beam electrode in the welding zone, the wire beam electrode in the base material zone and the wire beam electrode in the heat-affected zone is 30%:40%:30%. The reference electrode is a saturated silver chloride reference electrode and is equipped with a Luggin capillary. The auxiliary electrode has a working area of ​​2cm 2 of platinum electrodes.

[0024] Furthermore, the working electrode is the same as the wire beam electrode base material, wherein the cross-sectional area involved in corrosion is the same as the cross-sectional area of ​​the wire beam electrode probe in the experimental system 1, and the remaining portion is insulated by silicone rubber encapsulation. The scanning area of ​​the scanning vibration electrode is the corrosion area remaining after the silicone rubber encapsulation.

[0025] Compared with existing wire beam electrode testing methods, the present invention uses an artificial neural network prediction model of the weld joint current density under the synergistic effects of stress and corrosion to compensate for the measured wire beam electrode current density of the weld joint under stress-free conditions. This allows for indirect measurement of the current density distribution of the wire beam electrode in a force-electrochemical interaction environment. This effectively broadens the application scope of wire beam electrodes in the field of electrochemical corrosion testing of pipeline steel and helps obtain pipeline steel electrochemical corrosion test results that are closest to actual working conditions. The artificial neural network-based stress compensation method designed in the present invention has significant advantages in establishing a nonlinear mapping between the wire beam electrode current density and the working electrode current density. It can fully consider multiple input factors that affect stress compensation and has strong adaptive and nonlinear processing capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 The experimental system 1 used in the method of the present invention;

[0027] In the figure, 1 is the host computer, 2 is the multi-channel electrochemical workstation, 3 is the wire beam electrode test module, 4 is the high-speed switch module, 5 is the reference electrode, 6 is the wire beam electrode probe, 7 is the auxiliary electrode, and 8 is the corrosion electrolytic cell.

[0028] Figure 2 This is the wire beam electrode probe used in experimental system 1;

[0029] Figure 3 2 in the experimental system used in the method of the present invention;

[0030] In the figure, 1 is the host computer, 2 is the frequency response analyzer, 3 is the electrochemical workstation, 4 is the electrometer, 5 is the three-dimensional motion control device, 6 is the scanning vibration electrode, 7 is the tension sensor, 8 is the stress loading device, 9 is the corrosion electrolytic cell, 10 is the working electrode of the weld joint, 11 is the auxiliary electrode, and 12 is the reference electrode

[0031] Figure 4 is the working electrode used in experimental system 2;

[0032] Figure 5 This is a flow chart of the wire beam electrode current density stress compensation method based on artificial neural network;

[0033] Figure 6 This is a flow chart for predicting the current density of welded joints under the synergistic effect of stress and corrosion.

[0034] Figure 7 Flowchart of the neural network weight and threshold optimization update process based on quantum particle swarm optimization. DETAILED DESCRIPTION

[0035] The present invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments.

[0036] It should be noted that when a component is considered to be "connected to" another component, it can be directly connected to the other component or there may be an intermediate component. When a component is considered to be "disposed on" another component, it can be directly disposed on the other component or there may be an intermediate component. The terms "top", "bottom", "upper", "lower", "left", "right", "front", "back", and similar expressions used herein are for illustrative purposes only.

[0037] The following embodiments of the present application are described in detail with reference to the accompanying drawings. In the absence of conflict, the following embodiments and features therein may be combined with each other.

[0038] The technical solution adopted by the present invention is: an experimental system for a method for compensating current density stress of a wire bundle electrode of a pipeline steel welding joint includes: an experimental system 1 and an experimental system 2.

[0039] like Figure 1 As shown, the experimental system 1 is used to measure the current density distribution on the surface of the wire beam electrode probe of the weld joint under natural corrosion state without stress, including: a multi-channel electrochemical workstation 1-2, a wire beam electrode probe 1-6, an auxiliary electrode 1-7, a reference electrode 1-8, a wire beam electrode test module 1-3, a corrosion electrolytic cell 1-8, a multi-channel high-speed switching device 1-4 and a host computer 1-1.

[0040] Wire beam electrode probes 1-6 pass through 200 cross-sectional areas of 1mm 2 The array of cylindrical wire bundle electrodes is encapsulated in epoxy resin. The spacing between each wire bundle electrode is 1mm. The wire bundle electrodes are made of metals in the welding zone, base material zone and heat affected zone respectively. The ratio of the wire bundle electrode 1-6-2 in the welding zone, the wire bundle electrode 1-6-3 in the base material zone and the wire bundle electrode 1-6-1 in the heat affected zone is 30%:40%:30%. Figure 2 As shown. The weld zone, base metal zone, and heat-affected zone metal were taken from the spiral weld of pipeline steel. Reference electrodes 1-8 were saturated silver chloride reference electrodes with Luggin capillaries. Auxiliary electrodes 1-7 were 2 cm2 working areas. 2 The host computer is used to process the current density distribution data of the wire beam electrode.

[0041] Using the experimental system 1, the current density distribution matrix of the wire beam electrode probe can be measured as the current density distribution n×n matrix of the weld joint under stress-free conditions, which is:

[0042] Iw =[I 1,1 1 (s 1,1 1 ),I 1,2 1 (s 1,2 1 ),I 1,3 1 (s 1,3 1 ),…,I 1,n 1 (s 1,n 1 ),…,I k,1 1 (s k,1 1 ),I k,2 1 (s k,2 1 ), I k,3 1 (s k,3 1 ),…,I k,n 1 (s k,n 1 ),…,I n,1 1 (s n,1 1 ),I n,2 1 (s n,2 1 ),I n,3 1 (s n,3 1 ),…,I n,n 1 (s n,n 1 )], k is any number from 1 to n;

[0043] The current density in the matrix corresponds to the position of the bundle electrode in an n×n matrix:

[0044] S w =[s 1,1 1 ,s 1,2 1 ,s 1,3 1 ,…,s 1,n 1 ,…,s k,1 1 ,s k,2 1 ,s k,31 ,…,s k,n 1 ,…,s n,1 1 ,s n,2 1 ,s n,3 1 ,…,s n,n 1 ]

[0045] like Figure 3 As shown, experimental system 2 is used to measure the current density distribution on the surface of the working electrode of a weld joint under stress loading under natural corrosion conditions. It includes: an electrochemical workstation 2-3, a working electrode 2-10 for the weld joint, an auxiliary electrode 2-11, a reference electrode 2-12, a scanning vibration electrode 2-6, a three-dimensional motion control device 2-5, a corrosion electrolytic cell 2-9, an electrometer 2-4, a stress loading device 2-8, a tension sensor 2-7, and a host computer 2-1. The working electrode material is the same as the pipeline steel used in experimental system 1. The working electrode 2-10 is formed by submerged arc welding of two sections of the same base material. Three electrodes are used during the welding process, with a welding current of 600-700A and a welding voltage of 35-40V. Through holes are machined at both ends of the working electrode 2-10 for installing the stress loading device. The cross-sectional area of ​​the working electrode involved in corrosion is the same as the cross-sectional area of ​​the wire beam electrode probe in experimental system 1. The remaining area is insulated by silicone rubber encapsulation. The scanning area of ​​the scanning vibration electrode is the remaining corrosion area after the silicone rubber encapsulation, as shown in Figure 1. Figure 4 As shown in Figure 2 , the stress loading device 2-8 is bolted and nutted to each end of the working electrode, and the tensile stress load on both ends of the working electrode is measured by the tension sensor 2-7. The scanning vibration electrode 2-6 and the three-dimensional motion control device 2-5 are both implemented using the M370 system. The scanning vibration electrode 2-6 used in Experimental System 2 has a 10μm platinum-plated tip, a vibration amplitude of 30μm, and a vibration frequency of 300Hz.

[0046] Using the experimental system 2, the stress load F m The current density distribution matrix of the working electrode probe under the condition of , as the current density distribution matrix sequence I of the weld joint under stress d (F m ), which is:

[0047] I d (F m )=[I 1,1 2 (F m ,s 1,1 2 ),I1,2 2 (F m ,s 1,2 2 ),I 1,3 2 (F m ,s 1,3 2 ),…,I 1,n 2 (F m ,s 1,n 2 );…,I k,1 2 (F m , s k,1 2 ),I k,2 2 (F m ,s k,2 2 ),I k,3 2 (F m ,s k,3 2 ),…,I k,n 2 (F m ,s k,n 2 );…,I 2,1 2 (F m ,s n,1 2 ),I 2,2 2 (F m ,s n,2 2 ), I 2,3 2 (F m ,s n,3 2 ),…,I 2,n 2 (F m ,s n,n 2 )], m is any number from 1 to n;

[0048] n×n matrix S of the working electrode probe current density test position d The current density test position n×n matrix S on the wire beam electrode probe w The same, that is:

[0049] S w =[s 1,1 2 ,s 1,22 ,s 1,3 2 ,…,s 1,n 2 ;…,s k,1 2 ,s k,2 2 ,s k,3 2 ,…,s k,n 2 ;…,s n,1 2 ,s n,2 2 ,s n,3 2 ,…,s n,n 2 ]

[0050] S d =S w

[0051] The electrolyte solution composition and concentration, basic electrochemical test parameters, ambient temperature and humidity used in experimental system 1 and experimental system 2 must be kept the same.

[0052] The present invention also proposes a method for compensating the current density stress of the wire bundle electrode of a pipeline steel welding joint, such as Figure 5 This stress compensation method is applicable to various corrosion types and electrolyte compositions. The specific implementation steps are as follows:

[0053] Step 1: Place experimental system 1 and experimental system 2 in the same electrolyte solution concentration, ambient temperature and humidity, and use experimental system 1 to measure the current density distribution matrix I of wire bundle electrode probes 1-6 w =[I 1,1 1 (s 1,1 1 ),I 1,2 1 (s 1,2 1 ),I 1,3 1 (s 1,3 1 ),…,I 1,n 1 (s 1,n 1 );…,I k,1 1 (s k,1 1 ),I k,2 1 (s k,2 1), I k,3 1 (s k,3 1 ),…,I k,n 1 (s k,n 1 );…,I n,1 1 (s n,1 1 ),I n,2 1 (s n,2 1 ),I n,3 1 (s n,3 1 ),…,I n,n 1 (s n,n 1 )] and the position matrix S of the wire beam electrode w =[s 1,1 1 ,s 1,2 1 ,s 1,3 1 ,…,s 1,n 1 ;…,s k,1 1 ,s k,2 1 ,s k,3 1 ,…,s k,n 1 ;…,s n,1 1 ,s n,2 1 ,s n,3 1 ,…, s n,n 1 ]; change the loading stress [F1, F2, F3, ..., F m ,…,F n ], using the experimental system 2 to measure the current density distribution matrix sequence of the working electrode 2-10 of the welding joint {I d (F1),I d (F2),…,I d (F m ),…, I d (F n )}, k is any number from 1 to n, and m is any number from 1 to n.

[0054] Step 2: Construct the current density stress compensation data set. The data set includes two parts: input data set and output data set. The input data set includes: the position information of the current density measurement points of the wire beam electrode probes 1-6 and the working electrodes 2-10 of the weld joint [s 1,1 1 ,s 1,2 1 ,s 1,3 1 ,…,s 1,n 1 ;…,s k,1 1 ,s k,2 1 ,s k,3 1 ,…,s k,n 1 ;…, s n,1 1 ,s n,2 1 ,s n,3 1 ,…,s n,n 1 ]、Loading stress on working electrode 2-10 of welding joint [F1, F2, F3, ..., F m ,…, F n ] and the current density distribution information of the welding joint wire beam electrode probe 1-6 under no stress I w =[I 1,1 1 (s 1,1 1 ), I 1,2 1 (s 1,2 1 ),I 1,3 1 (s 1,3 1 ),…,I 1,n 1 (s 1,n 1 );…,I k,1 1 (s k,1 1 ),I k,2 1 (s k,2 1 ),I k,3 1 (s k,3 1 ),…,I k,n 1 (s k,n 1);…, I n,1 1 (s n,1 1 ),I n,2 1 (s n,2 1 ),I n,3 1 (s n,3 1 ),…,I n,n 1 (s n,n 1 )], the output data set is the current density distribution information of the working electrode 2-10 of the welding joint {I d (F1),I d (F2),…,I d (F m ),…,I d (F n )}.

[0055] Step 3: Position the current density measurement points with the wire beam electrode probes 1-6 and the working electrodes 2-10 of the weld joint. k,n 1 , working electrode loading stress F m The current density I of the wire bundle electrode probe 1-6 without stress k,n 1 (s k,n 1 ) as the input variable, and the current density under the working electrode stress as the output variable I k,n 2 (F m , s k,n 2 ), establish an artificial neural network. The process of building and training an artificial neural network is as follows Figure 6 shown.

[0056] Step 4: Based on the data set established in step 2, the input data set and the output data set are randomly divided into three parts: training set, test set and validation set in a ratio of 70%:15%:15%, and the training set, test set and validation set are normalized.

[0057] Step 5: Train, verify, and test the artificial neural network to meet the following conditions to complete the establishment of the artificial neural network model f(·):

[0058]

[0059] Among them, I k，n 2 (Fm , S k，n 2 ) is the current density test value at a certain position of the welding joint under stress, is the predicted value of current density at a certain position of the welding joint, and ε is the preset convergence accuracy of artificial neural network training.

[0060] During the neural network training process, in order to avoid the shortcomings of the traditional neural network training learning process, such as slow convergence speed, difficulty in obtaining ideal robustness, and unsatisfactory network performance, the quantum particle swarm algorithm is used to optimize the neural network weight and threshold update process. The optimization goal of the quantum particle swarm algorithm is expressed in the form of a vector:

[0061] [ω ih (1,1),...,ω ih (3,x),θ 1,1 ,...,θ 1,x ,ω ho (1,1),...,ω ho (x,1),θ 2,1 ]

[0062] Among them, ω ih (1,1),…,ω ih (3,1),…,ω ih (1,x),…,ω ih (3,x) is the input weight between the input layer and the hidden layer, ω ho (1,1),ω ho (2,1),…,ω ho (x,1) is the output weight between the hidden layer and the output layer, θ 1,1 ,θ 1,2 ,…, θ 1,x is the hidden layer neuron threshold, θ 2,1 is the output layer neuron threshold. The neural network weight and threshold optimization update process based on quantum particle swarm algorithm is as follows: Figure 7 As shown, the optimization iteration end condition is determined by presetting the number of iterations. When the following conditions are met, the prediction accuracy of the artificial neural network meets the requirements:

[0063]

[0064] Among them, I k，n 2 (F m , Sk ，n 2 ) is the current density test value at a certain position of the welding joint under stress, is the predicted value of the current density at a certain position of the weld joint, and ε is the preset convergence accuracy of the artificial neural network training. After the artificial neural network training accuracy is met, the artificial neural network model f(·) is established based on the optimized thresholds of each neural network node and the weights between nodes.

[0065] Step 6: Based on the artificial neural network model f(·) obtained after training, a current density stress compensation model for pipeline steel weld joints under stress is established. The output of the current density stress compensation model for the wire beam electrode of the weld joint is:

[0066] I k,n 2 =f(I k,n 1 ,F m ,s k,n )

[0067] Step 7: Measure the wire beam electrode current density distribution matrix of the weld joint under stress-free conditions, and substitute the current density measurement value and its corresponding position and loading stress amplitude into the stress compensation model to calculate the current density of a certain area of ​​the weld joint under the corresponding loading stress. This allows the wire beam electrode probe to indirectly measure the surface current density distribution of the pipeline steel weld joint when natural corrosion occurs under stress loading.

[0068] In addition, those skilled in the art should recognize that the above embodiments are merely intended to illustrate the present application and are not intended to limit the present application. As long as they are within the spirit of the present application, appropriate changes and modifications to the above embodiments are within the scope of disclosure of the present application.

Claims

1. A method for compensating current density stress of wire bundle electrodes in pipeline steel welding joints, characterized in that: Based on experimental system 1 and experimental system 2, current density stress compensation of wire beam electrode in pipeline steel welding joint is realized: The experimental system 1 is used to measure the current density distribution on the surface of the wire beam electrode probe of the weld joint under natural corrosion conditions without stress, and includes: a multi-channel electrochemical workstation (1-2), a wire beam electrode probe (1-6), an auxiliary electrode (1-7), a reference electrode (1-8), a wire beam electrode test module (1-3), a corrosion electrolytic cell (1-8), a multi-channel high-speed switching device (1-4), and a host computer (1-1); Experimental system 2 is used to measure the current density distribution on the surface of the working electrode of the weld joint when a stress load is applied under a natural corrosion state, and includes: an electrochemical workstation (2-3), a working electrode of the weld joint (2-10), an auxiliary electrode (2-11), a reference electrode (2-12), a scanning vibration electrode (2-6), a three-dimensional motion control device (2-5), a corrosion electrolytic cell (2-9), an electrometer (2-4), a stress loading device (2-8), a tension sensor (2-7) and a host computer (2-1). The stress loading device (2-8) is loaded on both ends of the working electrode (2-10), and the tensile stress load on the working electrode (2-10) is measured by the tension sensor (2-7). The following steps are involved: Step 1: Place experimental system 1 and experimental system 2 in the same electrolyte solution concentration, ambient temperature and humidity, and measure the current density distribution matrix I of the wire bundle electrode of experimental system 1. W =[I 1,1 1 (s 1,1 1 ),I 1,2 1 (s 1,2 1 ),I 1,3 1 (s 1,3 1 ),…,I 1,n 1 (s 1,n 1 );…,I k,1 1 (s k,1 1 ),I k,2 1 (s k,2 1 ),I k,3 1 (s k,3 1 ),…,I k,n 1 (s k,n 1 );…,I n,1 1 (s n,1 1 ),I n,2 1 (s n,2 1 ),I n,3 1 (s n,3 1 ),…,I n,n 1 (s n,n 1 )] and the position matrix S of the wire beam electrode W =[s 1,1 1 ,s 1,2 1 ,s 1,3 1 ,…,s 1,n 1 ;…,s k,1 1 ,s k,2 1 ,s k,3 1 ,…,s k,n 1 ;…,s n,1 1 ,s n,2 1 ,s n,3 1 ,…,s n,n 1 ]; change the loading stress [F1,F2,F3,…,F m ,…,F n ], measuring the working electrode current density distribution matrix sequence of experimental system 2 {I d (F1),I d (F2),…,I d (F m ),…,I d (F n )}, the position matrix of the working electrode is the same as that of the wire beam electrode, k is any number from 1 to n, and m is any number from 1 to n; Step 2: Construct the current density stress compensation data set, which includes the input data set and the output data set. The input data set includes: the position information of the wire bundle electrode array and the current density measurement point of the working electrode [s 1,1 1 ,s 1,2 1 ,s 1,3 1 ,…,s 1,n 1 ;…,s k,1 1 ,s k,2 1 ,s k,3 1 ,…,s k,n 1 ;…,s n,1 1 ,s n,2 1 ,s n,3 1 ,…,s n,n 1 ]、Loading stress of working electrode of welding joint [F1,F2,F3,…,F m ,…,F n ], and the current density distribution information of the wire bundle electrode array of the welded joint under stress-free conditions I W =[I 1,1 1 (s 1,1 1 ),I 1,2 1 (s 1,2 1 ),I 1,3 1 (s 1,3 1 ),…,I 1,n 1 (s 1,n 1 );…,I k,1 1 (s k,1 1 ),I k,2 1 (s k,2 1 ),I k,3 1 (s k,3 1 ),…,I k,n 1 (s k,n 1 );…,I n,1 1 (s n,1 1 ),I n,2 1 (s n,2 1 ),I n,3 1 (s n,3 1 ),…,I n,n 1 (s n,n 1 )], the output data set is the current density distribution information of the working electrode of the welding joint {I d (F1),I d (F2),…,I d (F m ),…,I d (F n )}; Step 3: Position the current density measurement point of the wire bundle electrode array and the working electrode Working electrode loading stress F m and the current density of the wire bundle electrode array without stress As the input variable, the current density under the working electrode stress is used as the output variable Building artificial neural networks; Step 4: Based on the dataset created in step 2, divide the input dataset and output dataset into three parts: training set, test set and validation set in a ratio of 70%:15%:15%, and normalize the training set, test set and validation set; Step 5: Train, verify, and test the artificial neural network to meet the following conditions to complete the establishment of the artificial neural network model f(·): in, is the current density test value at a certain position of the welding joint under stress, is the predicted value of current density at a certain position of the welding joint, and ε is the preset convergence accuracy of artificial neural network training; Step 6: Based on the artificial neural network model f(·), a current density stress compensation model for pipeline steel weld joints under stress is established. The output of the current density stress compensation model for the wire beam electrode of the weld joint is: Step 7: Measure the current density of the wire bundle electrode of the weld joint under stress-free conditions, and substitute the current density measurement value and its corresponding position and loading stress amplitude into the stress compensation model to obtain the current density of a certain area of ​​the weld joint under the corresponding loading stress.

2. A method for compensating current density stress of a wire bundle electrode in a pipeline steel welding joint according to claim 1, characterized in that: In step 5, the quantum particle swarm algorithm is used to optimize the neural network weight and threshold update process. The optimization goal of the quantum particle swarm algorithm is expressed in the form of a vector: [oh ih (1,1),…,ω ih (3,x),θ 1,1 ,…,θ 1,x ,oh ho (1,1),…,ω ho (x,1),θ 2,1 ] Among them, ω ih (1,1),…,ω ih (3,1),…,ω ih (1,x),…,ω ih (3,x) is the input weight between the input layer and the hidden layer, ω ho (1,1),ω ho (2,1),…,ω ho (x,1) is the output weight between the hidden layer and the output layer, θ 1,1 ,θ 1,2 ,…,θ 1,x is the hidden layer neuron threshold, θ 2,1 is the output layer neuron threshold.

3. The experimental system according to claim 1, wherein: The wire bundle electrode probe (1-6) passes through 200 cross-sectional areas of 1mm 2 The array of wire beam electrodes is encapsulated in epoxy resin. The spacing between each wire beam electrode is 1mm. The wire beam electrode is made of metal in the welding zone, base material zone and heat affected zone respectively. The ratio of wire beam electrode in welding zone, base material zone and heat affected zone is 30%:40%:30%. The reference electrode is a saturated silver chloride reference electrode and is equipped with a Luggin capillary. The auxiliary electrode has a working area of ​​2cm 2 of platinum electrodes.

4. The experimental system according to claim 3, characterized in that The working electrode (2-10) is the same as the wire beam electrode base material, and the cross-sectional area involved in corrosion is the same as the cross-sectional area of ​​the wire beam electrode probe (1-6) in experimental system 1. The remaining part is insulated by silicone rubber packaging. The scanning area of ​​the scanning vibration electrode is the remaining corrosion area after silicone rubber packaging.

Images