Strain detection method, device, electronic equipment and storage medium for steel structure
By constructing and using the steel structure strain detection model, the steel structure strain detection is automatically carried out, and the problems of low detection efficiency and low accuracy in the existing technology are solved, and efficient and accurate detection results are achieved.
Patent Information
- Application Number
- CN202411891276.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-20
- Publication Date
- 2025-04-08
AI Technical Summary
In the prior art, the detection efficiency of steel structure strain detection is low and the diagnostic accuracy cannot be guaranteed, mainly because it relies on manual experience for analysis, low efficiency and the results depend on the level of the operator.
Construct a steel structure strain detection model, obtain the model through strain detection sample data training, collect the strain change data of the target object, and input it into the model to generate strain detection results, including normal or abnormal steel structure of the target object.
Through automated strain detection methods, the detection efficiency and accuracy are improved, the dependence on manual diagnosis is reduced, and the reliability of the detection results is ensured.
Smart Images
Figure CN119782884A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of steel structure detection, and in particular to a strain detection method, device, electronic equipment and storage medium for steel structures. Background Art
[0002] Steel structure is a structure composed of steel sections, steel trusses and other components connected by welds, bolts or rivets. It is widely used in various engineering fields due to its high strength, good ductility and easy construction. It is the main material of mechanical equipment and building structures. During the long-term use of steel structure, phenomena such as structural deformation and cracking will occur, which seriously threatens production safety. Strain detection, as one of the means of steel structure detection, has been recognized as important without damaging or affecting the performance of the inspected steel structure.
[0003] In the existing technology, the process of steel structure strain detection is to collect and analyze the strain model, and then generate a signal spectrum, which is assisted by experienced engineers to perform fault diagnosis. Although this process is effective, it relies on manual experience to perform analysis, which is inefficient. In addition, the correctness of the test results depends largely on the level of the operator, and the accuracy of the test results cannot be guaranteed. Summary of the invention
[0004] In view of this, the present application provides a steel structure strain detection method, device, electronic device and storage medium to solve the problems of low detection efficiency and inability to guarantee diagnostic accuracy in the prior art of steel structure strain detection.
[0005] To achieve the above objectives, this application provides the following technical solutions:
[0006] The first aspect of the present application discloses a strain detection method for a steel structure, comprising:
[0007] Constructing a steel structure strain detection model; wherein the steel structure strain detection model is pre-trained by strain detection sample data;
[0008] Collecting strain change data of a target object; wherein the target object is a steel structure object to be detected;
[0009] The change data is input into the steel structure strain detection model to obtain the strain detection result of the target object; wherein the strain detection includes the normal steel structure of the target object and the abnormal steel structure of the target object.
[0010] Optionally, in the above method, the step of constructing a steel structure strain detection model comprises:
[0011] Constructing an initial model for steel structure strain detection based on a gated recurrent unit; wherein the initial model for steel structure strain detection includes an input layer, a convolutional layer, a gated recurrent unit layer, and a fully connected layer;
[0012] Acquire strain detection sample data and actual strain detection results corresponding to the strain detection sample data;
[0013] Inputting the strain detection sample data into the steel structure strain detection initial model for calculation to obtain the strain detection result corresponding to the strain detection sample data;
[0014] Determining whether a strain detection result corresponding to the strain detection sample data is consistent with an actual strain detection result corresponding to the strain detection sample data;
[0015] If the strain detection result corresponding to the strain detection sample data is consistent with the actual strain detection result corresponding to the strain detection sample data, the construction of the steel structure strain detection model is completed;
[0016] If the strain detection result corresponding to the strain detection sample data is inconsistent with the actual strain detection result corresponding to the strain detection sample data, a loss function is calculated, and the parameters of the initial model of steel structure strain detection are adjusted based on the loss function until the strain detection result corresponding to the strain detection sample data is consistent with the actual strain detection result corresponding to the strain detection sample data or the preset number of training times is reached, and the construction of the steel structure strain detection model is completed.
[0017] Optionally, the above method, after constructing the initial model for steel structure strain detection based on the gated cyclic unit, further comprises:
[0018] The weight parameters and bias parameters of the initial model for steel structure strain detection are optimized by using the Adam optimizer.
[0019] Optionally, the above method, after constructing the steel structure strain detection model, further comprises:
[0020] The steel structure strain detection model is subjected to transfer learning processing to obtain a transfer model.
[0021] The second aspect of the present application discloses a strain detection device for a steel structure, comprising:
[0022] A construction unit is used to construct a steel structure strain detection model; wherein the steel structure strain detection model is pre-trained by strain detection sample data;
[0023] A collection unit, used for collecting strain change data of a target object; wherein the target object is a steel structure object to be detected;
[0024] The detection unit is used to input the change data into the steel structure strain detection model to obtain the strain detection result of the target object; wherein the strain detection includes the normal steel structure of the target object and the abnormal steel structure of the target object.
[0025] Optionally, the construction unit of the above-mentioned device includes:
[0026] A first construction subunit is used to construct an initial model for steel structure strain detection based on a gated cyclic unit; wherein the initial model for steel structure strain detection includes an input layer, a convolutional layer, a gated cyclic unit layer, and a fully connected layer;
[0027] An acquisition subunit, used to acquire strain detection sample data and actual strain detection results corresponding to the strain detection sample data;
[0028] A calculation subunit, used for inputting the strain detection sample data into the steel structure strain detection initial model for calculation, and obtaining the strain detection result corresponding to the strain detection sample data;
[0029] A judging subunit, used to judge whether the strain detection result corresponding to the strain detection sample data is consistent with the actual strain detection result corresponding to the strain detection sample data;
[0030] A second construction subunit is used to complete the construction of the steel structure strain detection model if the strain detection result corresponding to the strain detection sample data is consistent with the actual strain detection result corresponding to the strain detection sample data;
[0031] The adjustment subunit is used to calculate the loss function if the strain detection result corresponding to the strain detection sample data is inconsistent with the actual strain detection result corresponding to the strain detection sample data, and adjust the parameters of the initial model of the steel structure strain detection based on the loss function, until the strain detection result corresponding to the strain detection sample data is consistent with the actual strain detection result corresponding to the strain detection sample data or the preset number of training times is reached, thereby completing the construction of the steel structure strain detection model.
[0032] Optionally, the above device further includes:
[0033] The optimization unit is used to optimize the weight parameters and bias parameters of the steel structure strain detection initial model through the Adam optimizer.
[0034] Optionally, the above device further includes:
[0035] The processing unit is used to perform transfer learning processing on the steel structure strain detection model to obtain a transfer model.
[0036] The third aspect of the present application discloses an electronic device, comprising:
[0037] one or more processors;
[0038] a storage device having one or more programs stored thereon;
[0039] When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of the first aspects of the present invention.
[0040] A fourth aspect of the present application discloses a computer storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the method described in any one of the first aspects of the present invention is implemented.
[0041] It can be seen from the above technical scheme that the present application provides a method for strain detection of steel structures, which first constructs a steel structure strain detection model; wherein the steel structure strain detection model is pre-trained by strain detection sample data. Then, the strain change data of the target object is collected; wherein, the target object is a steel structure object to be detected. Finally, the change data is input into the steel structure strain detection model to obtain the strain detection result of the target object; wherein, the strain detection includes the normal steel structure of the target object and the abnormal steel structure of the target object. It can be seen that, by using the method of the present application, by constructing a steel structure strain detection model, the steel structure strain detection can be automatically performed, without relying on manual diagnosis and analysis, thereby improving the efficiency and accuracy of the detection. It solves the problem of low detection efficiency and inability to guarantee diagnostic accuracy of steel structure strain detection in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.
[0043] Figure 1 A flow chart of a strain detection method for a steel structure disclosed in an embodiment of the present application;
[0044] Figure 2 A flowchart of an implementation of the training process of the risk assessment model disclosed in the embodiments of the present application;
[0045] Figure 3 A schematic diagram of a strain detection device for a steel structure disclosed in an embodiment of the present application;
[0046] Figure 4 A schematic diagram of an electronic device disclosed in an embodiment of the present application. DETAILED DESCRIPTION
[0047] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0048] In this application, the terms "comprises", "comprising" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element.
[0049] Furthermore, in this document, relational terms such as first and second, etc. are used only to distinguish one entity or operation from another entity or operation, but do not necessarily require or imply any such actual relationship or order between these entities or operations.
[0050] As can be seen from the background technology, in the prior art, the process of steel structure strain detection is to collect and analyze the strain model, and then generate a signal spectrum, assisted by experienced engineers to perform fault diagnosis. Although this process is effective, it requires manual experience to perform analysis, which is inefficient, and the accuracy of the detection results depends largely on the level of the operator, and the accuracy of the detection results cannot be guaranteed.
[0051] In view of this, the present application provides a steel structure strain detection method, device, electronic device and storage medium to solve the problems of low detection efficiency and inability to guarantee diagnostic accuracy in the prior art of steel structure strain detection.
[0052] The present application embodiment provides a method for detecting strain of a steel structure, such as Figure 1 As shown, specifically including:
[0053] S101, constructing a steel structure strain detection model; wherein the steel structure strain detection model is pre-trained with strain detection sample data.
[0054] It should be noted that first, strain detection sample data is obtained, and a steel structure strain detection model is obtained by training the strain detection sample data. The steel structure strain detection model is a deep learning model based on GRU (Gated Recurrent Unit).
[0055] Optionally, in another embodiment of the present application, an implementation of step S101 is as follows: Figure 2 As shown, specifically including:
[0056] S201. Constructing an initial model for steel structure strain detection based on a gated recurrent unit; wherein the initial model for steel structure strain detection includes an input layer, a convolutional layer, a gated recurrent unit layer, and a fully connected layer.
[0057] It should be noted that, first, the gated recurrent unit constructs the initial model of steel structure strain detection; wherein, the initial model of steel structure strain detection includes an input layer, a convolutional layer, a gated recurrent unit layer and a fully connected layer. The first is the input layer, which performs an input on the data. Then there are two one-dimensional convolutional layers, each consisting of 16 convolution kernels of size 4, a step size of 1, and a linear activation function. After the two convolutional layers extract the features of the data, it passes through two GRU layers, and finally two fully connected layers. The first fully connected layer integrates all the features of the upper layer, and finally outputs the decomposition result through the fully connected layer of N=1.
[0058] It should also be noted that the GRU (Gated Recurrent Unit) has two main gates: the Reset Gate and the Update Gate.
[0059] The main function of the reset gate is to determine the extent to which the state information of the previous moment should be forgotten. It is adjusted by a sigmoid activation function, and the output value is between 0 and 1.
[0060] The main function of the update gate is to determine the influence of the state information at the previous moment on the current moment, that is, to control how much memory of the previous moment should be retained in the hidden state at the current moment.
[0061] The relevant formula is as follows:
[0062] Update gate formula:
[0063] Reset gate formula:
[0064] Current memory content:
[0065] Current time final memory:
[0066] Among them, z tis the update gate, r t is the reset gate, x t represents the input at time t, h t Represents the hidden state at time t, and W and U both represent weight matrices.
[0067] Optionally, in another embodiment of the present application, after executing step S201, the following may also be included:
[0068] The weight parameters and bias parameters of the initial model for steel structure strain detection are optimized by using the Adam optimizer.
[0069] It should be noted that Adam is a gradient-based optimization algorithm that updates parameter values by calculating parameter gradients. Adam introduces a momentum term to solve the oscillation problem of gradient descent in non-convex functions. Adam also introduces the calculation of the root mean to solve the problem of excessive smoothing of gradient descent in the sequence. At the same time, Adam also introduces the information of the second-order derivative to solve the slow convergence problem of gradient descent in large-scale gradient updates. Therefore, the weight parameters and bias parameters of the initial model of steel structure strain detection are optimized by the Adam optimizer.
[0070] S202: Acquire strain detection sample data and actual strain detection results corresponding to the strain detection sample data.
[0071] It should be noted that first, the strain detection sample data and the actual strain detection result corresponding to the strain detection sample data are obtained.
[0072] S203: Input the strain detection sample data into an initial model for steel structure strain detection to perform calculations, and obtain strain detection results corresponding to the strain detection sample data.
[0073] It should be noted that the strain detection sample data is input into the steel structure strain detection initial model for operation to obtain the strain detection results corresponding to the strain detection sample data, and the strain detection results indicate that the steel structure is normal and the steel structure is abnormal.
[0074] S204: Determine whether the strain detection result corresponding to the strain detection sample data is consistent with the actual strain detection result corresponding to the strain detection sample data.
[0075] It should be noted that after the strain detection results corresponding to the strain detection sample data are calculated in the initial model of steel structure strain detection, the strain detection results corresponding to the calculated strain detection sample data are compared with the actual strain detection results corresponding to the strain detection sample data, so as to determine whether the strain detection results corresponding to the strain detection sample data calculated in the initial model of steel structure strain detection are accurate.
[0076] S205: If the strain detection result corresponding to the strain detection sample data is consistent with the actual strain detection result corresponding to the strain detection sample data, the construction of the steel structure strain detection model is completed.
[0077] It should be noted that if the strain detection result corresponding to the strain detection sample data is consistent with the actual strain detection result corresponding to the strain detection sample data, it means that the model operation is accurate, and the construction of the steel structure strain detection model is completed.
[0078] S206. If the strain detection result corresponding to the strain detection sample data is inconsistent with the actual strain detection result corresponding to the strain detection sample data, a loss function is calculated, and the parameters of the initial model of the steel structure strain detection are adjusted based on the loss function until the strain detection result corresponding to the strain detection sample data is consistent with the actual strain detection result corresponding to the strain detection sample data or a preset number of training times is reached, and the construction of the steel structure strain detection model is completed.
[0079] It should be noted that if the strain detection result corresponding to the strain detection sample data is inconsistent with the actual strain detection result corresponding to the strain detection sample data, the loss function is calculated, and the parameters of the initial model of the steel structure strain detection are adjusted based on the loss function. Among them, the loss function is an important learning criterion, which can be used to measure the performance of a network and is often related to the optimization amount. The mean square error loss function is used to calculate the mean square error between them. The calculation formula is shown in the following formula:
[0080]
[0081] Where N is the length of the input sequence, is the predicted value of the model, is the true value;
[0082] By carefully adjusting the parameters and functions, an effective network training process can be achieved, that is, the parameters of the model are continuously adjusted according to the pre-set number of training times until a certain threshold is reached. In this way, even if the expected number of training times is not reached, the model can stop running when the threshold is met.
[0083] Until the strain detection result corresponding to the strain detection sample data is consistent with the actual strain detection result corresponding to the strain detection sample data, or a preset number of training times (for example, 3 times) is reached, if the loss function is still not significantly improved, the construction of the steel structure strain detection model is completed.
[0084] Optionally, in another embodiment of the present application, after executing step S101, the following may also be included:
[0085] The steel structure strain detection model is subjected to transfer learning processing to obtain a transfer model.
[0086] It should be noted that if a migration model needs to be constructed, the steel structure strain detection model can be subjected to migration learning processing to obtain a migration model, as follows:
[0087] (1) Dataset preparation: Create two datasets. One is the source domain dataset, which is used to train the source domain model, i.e., the steel structure strain detection model, and the other is the target domain dataset, which is used to train the migration model.
[0088] (2) Loading data and models: Loading pre-trained network models and target domain datasets;
[0089] (3) Source domain model construction: construct source domain dataset and train steel structure strain detection model;
[0090] (4) Migration model construction: Based on the above source domain model, except for the fully connected layer, the remaining network layers are set as migration layers. Then new network layers are added to adapt to the data of the target domain;
[0091] (5) Hyperparameter adjustment: Fine-tune the model by reducing the initial learning rate of the source domain model to slow down its learning rate while increasing the learning rate of the fully connected layer;
[0092] (6) Testing: Use unlabeled sample data from the target domain to test the above model. Through multiple tests, evaluate the accuracy of the model on the target domain data to obtain the recognition result.
[0093] S102, collecting strain change data of a target object; wherein the target object is a steel structure object to be detected.
[0094] It should be noted that the strain change data of the steel structure object to be detected is collected by the strain rosette, numbered according to the arrangement of the strain rosettes, and combined with the thickness data of the steel structure object to extract the strain detection data sequence.
[0095] S103, inputting the change data into the steel structure strain detection model to obtain the strain detection result of the target object; wherein the strain detection includes the steel structure of the target object being normal and the steel structure of the target object being abnormal.
[0096] It should be noted that after the strain change data of the target object is collected, the change data is input into the steel structure strain detection model to obtain the strain detection result of the target object; wherein the strain detection includes the normal steel structure of the target object and the abnormal steel structure of the target object. In this way, the steel structure strain detection of the target object can be completed.
[0097] A method for strain detection of steel structures provided in an embodiment of the present application first constructs a steel structure strain detection model; wherein the steel structure strain detection model is pre-trained with strain detection sample data. Then, the strain change data of the target object is collected; wherein, the target object is a steel structure object to be detected. Finally, the change data is input into the steel structure strain detection model to obtain the strain detection result of the target object; wherein, the strain detection includes the normal steel structure of the target object and the abnormal steel structure of the target object. It can be seen that, by using the method of the present application, by constructing a steel structure strain detection model, the steel structure strain detection can be automatically performed without relying on manual diagnosis and analysis, thereby improving the efficiency and accuracy of the detection. The problem of low detection efficiency and inability to guarantee diagnostic accuracy of steel structure strain detection in the prior art is solved.
[0098] Another embodiment of the present application also discloses a strain detection device for a steel structure, such as Figure 3 As shown, specifically including:
[0099] The construction unit 301 is used to construct a steel structure strain detection model; wherein the steel structure strain detection model is pre-trained by strain detection sample data.
[0100] The acquisition unit 302 is used to acquire strain change data of a target object; wherein the target object is a steel structure object to be detected.
[0101] The detection unit 303 is used to input the change data into the steel structure strain detection model to obtain the strain detection result of the target object; wherein the strain detection includes the steel structure of the target object being normal and the steel structure of the target object being abnormal.
[0102] In this embodiment, the specific execution process of the construction unit 301, the collection unit 302 and the detection unit 303 can be found in the corresponding Figure 1 The content of the method embodiment will not be repeated here.
[0103] The present application provides a strain detection device for a steel structure. First, a construction unit 301 constructs a steel structure strain detection model; wherein the steel structure strain detection model is obtained in advance through training of strain detection sample data. Then, the acquisition unit 302 acquires the strain change data of the target object; wherein, the target object is a steel structure object to be detected. Finally, the detection unit 303 inputs the change data into the steel structure strain detection model to obtain the strain detection result of the target object; wherein, the strain detection includes the normal steel structure of the target object and the abnormal steel structure of the target object. It can be seen that, by using the method of the present application, by constructing a steel structure strain detection model, the steel structure strain detection can be automatically performed without relying on manual diagnosis and analysis, thereby improving the efficiency and accuracy of the detection. The problem of low detection efficiency and inability to guarantee diagnostic accuracy of steel structure strain detection in the prior art is solved.
[0104] Optionally, in another embodiment of the present application, an implementation of the construction unit 301 includes:
[0105] The first construction subunit is used to construct an initial model for steel structure strain detection based on a gated cyclic unit; wherein the initial model for steel structure strain detection includes an input layer, a convolutional layer, a gated cyclic unit layer and a fully connected layer.
[0106] The acquisition subunit is used to acquire strain detection sample data and actual strain detection results corresponding to the strain detection sample data.
[0107] The operation subunit is used to input the strain detection sample data into the steel structure strain detection initial model for operation to obtain the strain detection result corresponding to the strain detection sample data.
[0108] The judging subunit is used to judge whether the strain detection result corresponding to the strain detection sample data is consistent with the actual strain detection result corresponding to the strain detection sample data.
[0109] The second construction subunit is used to complete the construction of the steel structure strain detection model if the strain detection result corresponding to the strain detection sample data is consistent with the actual strain detection result corresponding to the strain detection sample data.
[0110] The adjustment subunit is used to calculate the loss function if the strain detection result corresponding to the strain detection sample data is inconsistent with the actual strain detection result corresponding to the strain detection sample data, and adjust the parameters of the initial model of the steel structure strain detection based on the loss function, until the strain detection result corresponding to the strain detection sample data is consistent with the actual strain detection result corresponding to the strain detection sample data or the preset number of training times is reached, thereby completing the construction of the steel structure strain detection model.
[0111] In this embodiment, the specific execution process of the first construction subunit, the acquisition subunit, the operation subunit, the judgment subunit, the second construction subunit and the adjustment subunit can be referred to in the corresponding Figure 2 The content of the method embodiment will not be repeated here.
[0112] Optionally, in another embodiment of the present application, the strain detection device for the steel structure may further include:
[0113] The optimization unit is used to optimize the weight parameters and bias parameters of the steel structure strain detection initial model through the Adam optimizer.
[0114] In this embodiment, the specific execution process of the optimization unit can refer to the corresponding method embodiment content mentioned above, which will not be repeated here.
[0115] Optionally, in another embodiment of the present application, the strain detection device for the steel structure may further include:
[0116] The processing unit is used to perform transfer learning processing on the steel structure strain detection model to obtain a transfer model.
[0117] In this embodiment, the specific execution process of the processing unit can refer to the corresponding method embodiment content mentioned above, which will not be repeated here.
[0118] Another embodiment of the present application further provides an electronic device, such as Figure 4 As shown, specifically including:
[0119] One or more processors 401 .
[0120] The storage device 402 stores one or more programs.
[0121] When one or more programs are executed by one or more processors 401 , the one or more processors 401 implement any one of the methods in the above embodiments.
[0122] Another embodiment of the present application further provides a computer storage medium on which a computer program is stored, wherein when the computer program is executed by a processor, any one of the methods in the above embodiments is implemented.
[0123] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can refer to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system or system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiment. The system and system embodiments described above are merely schematic, 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 creative work.
[0124] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may 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.
[0125] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A strain detection method for a steel structure, characterized in that: include: Constructing a steel structure strain detection model; wherein the steel structure strain detection model is pre-trained by strain detection sample data; Collecting strain change data of a target object; wherein the target object is a steel structure object to be detected; The strain change data is input into the steel structure strain detection model to obtain the strain detection result of the target object; wherein the strain detection result includes the normal steel structure of the target object and the abnormal steel structure of the target object.
2. The method according to claim 1, characterized in that: The construction of the steel structure strain detection model includes: Constructing an initial model for steel structure strain detection based on a gated recurrent unit; wherein the initial model for steel structure strain detection includes an input layer, a convolutional layer, a gated recurrent unit layer, and a fully connected layer; Acquire strain detection sample data and actual strain detection results corresponding to the strain detection sample data; Inputting the strain detection sample data into the steel structure strain detection initial model for calculation to obtain the strain detection result corresponding to the strain detection sample data; Determining whether a strain detection result corresponding to the strain detection sample data is consistent with an actual strain detection result corresponding to the strain detection sample data; If the strain detection result corresponding to the strain detection sample data is consistent with the actual strain detection result corresponding to the strain detection sample data, the construction of the steel structure strain detection model is completed; If the strain detection result corresponding to the strain detection sample data is inconsistent with the actual strain detection result corresponding to the strain detection sample data, a loss function is calculated, and the parameters of the initial model of steel structure strain detection are adjusted based on the loss function until the strain detection result corresponding to the strain detection sample data is consistent with the actual strain detection result corresponding to the strain detection sample data or the preset number of training times is reached, and the construction of the steel structure strain detection model is completed.
3. The method according to claim 2, characterized in that After the initial model of steel structure strain detection is constructed based on the gated cyclic unit, the method further includes: The weight parameters and bias parameters of the initial model for steel structure strain detection are optimized by using the Adam optimizer.
4. The method according to claim 1, characterized in that After the steel structure strain detection model is constructed, the following steps are also included: The steel structure strain detection model is subjected to transfer learning processing to obtain a transfer model.
5. A strain detection device for a steel structure, characterized in that: include: A construction unit is used to construct a steel structure strain detection model; wherein the steel structure strain detection model is pre-trained by strain detection sample data; A collection unit, used for collecting strain change data of a target object; wherein the target object is a steel structure object to be detected; The detection unit is used to input the strain change data into the steel structure strain detection model to obtain the strain detection result of the target object; wherein the strain detection result includes the normal steel structure of the target object and the abnormal steel structure of the target object.
6. The device according to claim 5, characterized in that The construction unit comprises: A first construction subunit is used to construct an initial model for steel structure strain detection based on a gated cyclic unit; wherein the initial model for steel structure strain detection includes an input layer, a convolutional layer, a gated cyclic unit layer, and a fully connected layer; An acquisition subunit, used to acquire strain detection sample data and actual strain detection results corresponding to the strain detection sample data; A calculation subunit, used for inputting the strain detection sample data into the steel structure strain detection initial model for calculation, and obtaining the strain detection result corresponding to the strain detection sample data; A judging subunit, used to judge whether the strain detection result corresponding to the strain detection sample data is consistent with the actual strain detection result corresponding to the strain detection sample data; A second construction subunit is used to complete the construction of the steel structure strain detection model if the strain detection result corresponding to the strain detection sample data is consistent with the actual strain detection result corresponding to the strain detection sample data; The adjustment subunit is used to calculate the loss function if the strain detection result corresponding to the strain detection sample data is inconsistent with the actual strain detection result corresponding to the strain detection sample data, and adjust the parameters of the initial model of the steel structure strain detection based on the loss function, until the strain detection result corresponding to the strain detection sample data is consistent with the actual strain detection result corresponding to the strain detection sample data or the preset number of training times is reached, thereby completing the construction of the steel structure strain detection model.
7. The device according to claim 6, characterized in that Also includes: The optimization unit is used to optimize the weight parameters and bias parameters of the steel structure strain detection initial model through the Adam optimizer.
8. The device according to claim 6, characterized in that Also includes: The processing unit is used to perform transfer learning processing on the steel structure strain detection model to obtain a transfer model.
9. An electronic device, characterized in that: include: one or more processors; a storage device having one or more programs stored thereon; When the one or more programs are executed by the one or more processors, the one or more processors are enabled to implement the method according to any one of claims 1 to 4.
10. A computer storage medium, characterized in that: A computer program is stored thereon, wherein when the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.