UHPC composite beam early warning method, device, equipment and medium based on deep learning

Abstract

This application discloses a deep learning-based early warning method, device, equipment, and medium for UHPC composite beams, relating to the technical field of bridge structure monitoring. The method comprises: obtaining structural information of ultra-high performance concrete composite beams, inputting the structural information into a preset state prediction model for processing, and obtaining target characteristic states and target performance prediction results; performing detection through a multi-level early warning mechanism based on the target characteristic states and target performance prediction results, obtaining detection results, and executing corresponding early warning operations based on the detection results. By using a deep learning model to process the structural information of ultra-high performance concrete (UHPC) composite beams, key characteristic states are automatically identified and performance is predicted. Real-time monitoring and risk warning are achieved through a multi-level early warning mechanism, improving the accuracy of characteristic state identification and the reliability of performance prediction, enabling early risk detection and active early warning, and enhancing structural safety and design efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of bridge structure monitoring, and in particular to a UHPC composite beam early warning method, device, equipment and medium based on deep learning. Background Art

[0002] With the widespread application of ultra-high-performance concrete (UHPC) in bridge engineering, steel-UHPC composite beams have attracted increasing attention due to their excellent mechanical properties and durability. Domestic and international scholars have conducted extensive research on these composite beams, achieving significant progress in material properties, structural measures, computational theory, and experimental research. Through systematic research, the constitutive relationship and mechanical properties of UHPC and steel have been established, and various combinations and connection structures have been developed to improve the overall performance of the structure. Furthermore, a system of design methods based on cross-sectional analysis has been established, and a large number of static and fatigue tests have provided reliable data support for theoretical research.

[0003] Existing research methods mainly rely on classical mechanics theory for analysis, which requires a large number of iterative calculations. The process is complex and time-consuming, and it is difficult to meet the needs of rapid design. For characteristic state identification, current methods mainly rely on preset judgment conditions and engineering experience, lack intelligent means, especially in the prediction of state transitions and failure modes under complex working conditions. In terms of performance prediction, especially in complex working conditions and long-term performance prediction, the accuracy and reliability of existing methods need to be improved. Early warning mechanisms generally have lags, which cannot achieve early risk identification and proactive prevention. More importantly, the large amount of accumulated test data and engineering experience has not been fully utilized, and knowledge inheritance and experience accumulation face bottlenecks. Therefore, an efficient and accurate method is needed to detect and warn UHPC composite beams. Summary of the Invention

[0004] The main purpose of this application is to provide a UHPC composite beam early warning method, device, equipment and medium based on deep learning, aiming to solve the technical problem of how to obtain the status of UHPC composite beams and perform corresponding early warning operations.

[0005] To achieve the above objectives, this application proposes a UHPC composite beam early warning method based on deep learning, which includes:

[0006] Obtain structural information of ultra-high performance concrete composite beams;

[0007] Inputting the structural information into a preset state prediction model for processing to obtain target feature state and target performance prediction results, the preset state prediction model is a hybrid neural network model, the preset state prediction model includes a feature extraction network, a temporal feature network, an attention layer and an output layer, the feature extraction network includes a first convolution block, a second convolution block and a third convolution block, the first convolution block, the second convolution block and the third convolution block each include two one-dimensional convolution layers, a normalization layer, an activation function layer and a maximum pooling layer, the temporal feature network includes a bidirectional long short-term memory network layer and a normalization layer, the output layer includes a feature state classification branch and a performance prediction branch, the feature state classification branch outputs the probability distribution of the feature state through a three-layer fully connected network, and the performance prediction branch outputs the performance index value through a three-layer fully connected network, according to the hybrid neural network model, the feature extraction network performs local feature extraction, the bidirectional long short-term memory network performs temporal feature capture and the attention layer performs key feature weighting to achieve deep learning of the feature state of the ultra-high performance concrete composite beam;

[0008] Performing detection based on the target characteristic state and the target performance prediction result through a multi-level early warning mechanism to obtain a detection result;

[0009] Execute corresponding warning operations according to the detection results.

[0010] In one embodiment, before the step of inputting the configuration information into a preset state prediction model for processing to obtain target characteristic state and target performance prediction results, the following steps are included:

[0011] Obtain structural information sample data and build an initial state prediction model;

[0012] The initial state prediction model is trained based on the construction information sample data to obtain a preset state prediction model.

[0013] In one embodiment, the step of training the initial state prediction model based on the construction information sample data to obtain a preset state prediction model includes:

[0014] Initializing weight and bias parameters of the initial state prediction model;

[0015] Inputting the structural information sample data into the initial state prediction model for calculation to obtain characteristic state classification results and performance prediction results;

[0016] Calculating and weighting the feature state classification result and the performance prediction result according to the loss function to obtain a total error value;

[0017] Obtaining the gradients of the weight and bias parameters by back propagation algorithm calculation;

[0018] The weight and bias parameters are iteratively updated according to the gradient through an optimization algorithm until a maximum number of iterations is reached or the total error value converges to a preset threshold, thereby obtaining a preset state prediction model.

[0019] In one embodiment, before the step of inputting the configuration information into a preset state prediction model for processing to obtain target characteristic state and target performance prediction results, the following steps are included:

[0020] Inputting the structural information into a preset state prediction model for processing to obtain a characteristic state;

[0021] Performing an evaluation based on the feature status to obtain a confidence score;

[0022] When the confidence score exceeds a preset threshold, the feature state is output as the target feature state;

[0023] When the confidence score does not exceed the preset threshold, an early warning signal is issued, and the step of inputting the construction information into a preset state prediction model for processing to obtain a characteristic state is executed until the confidence score exceeds the preset threshold.

[0024] In one embodiment, the step of performing detection based on the target characteristic state and the target performance prediction result through a multi-level early warning mechanism to obtain a detection result includes:

[0025] According to the multi-level warning mechanism, a first threshold, a second threshold, and a third threshold are obtained, wherein the first threshold is less than the second threshold, and the second threshold is less than the third threshold;

[0026] Obtaining a composite beam bending moment according to the target characteristic state and the target performance prediction result;

[0027] The composite beam bending moment is compared with the first threshold, the second threshold, and the third threshold to obtain a detection result.

[0028] In one embodiment, the step of performing a corresponding warning operation according to the detection result includes:

[0029] When the bending moment of the composite beam is less than the first threshold, the detection result is obtained as no risk, and normal monitoring and recording of key parameters are performed;

[0030] When the combined beam bending moment is greater than the first threshold and the combined beam bending moment is less than the second threshold, a detection result is obtained as a first-level warning, and operations of increasing the frequency of obtaining structural information and preparing an emergency plan are performed;

[0031] When the combined beam bending moment is greater than the second threshold and the combined beam bending moment is less than the third threshold, a detection result is obtained as a second-level warning, and an operation of sending a warning message to relevant personnel through real-time monitoring is executed;

[0032] When the bending moment of the composite beam is greater than the third threshold, the detection result is a level 3 warning, and an operation of triggering an emergency response for maintenance is executed.

[0033] In addition, to achieve the above objectives, the present application also proposes a UHPC composite beam early warning device based on deep learning, which includes:

[0034] An acquisition module, used to obtain structural information of ultra-high performance concrete composite beams;

[0035] A processing module is configured to input the structural information into a preset state prediction model for processing to obtain target characteristic state and target performance prediction results, wherein the preset state prediction model is a hybrid neural network model, comprising a feature extraction network, a temporal feature network, an attention layer, and an output layer. The feature extraction network comprises a first convolution block, a second convolution block, and a third convolution block, wherein the first convolution block, the second convolution block, and the third convolution block each comprise two one-dimensional convolution layers, a normalization layer, an activation function layer, and a maximum pooling layer. The temporal feature network comprises a bidirectional long short-term memory network layer and a normalization layer. The output layer comprises a characteristic state classification branch and a performance prediction branch. The characteristic state classification branch outputs a probability distribution of the characteristic state through a three-layer fully connected network, and the performance prediction branch outputs a performance index value through a three-layer fully connected network. According to the hybrid neural network model, local feature extraction is performed by the feature extraction network, temporal feature capture is performed by the bidirectional long short-term memory network, and key feature weighting is performed by the attention layer to achieve deep learning of the characteristic state of the ultra-high performance concrete composite beam.

[0036] A detection module, configured to perform detection based on the target characteristic state and the target performance prediction result through a multi-level early warning mechanism to obtain a detection result;

[0037] An execution module is used to execute corresponding warning operations according to the detection results.

[0038] In addition, to achieve the above-mentioned purpose, the present application also proposes a medium, which is a computer-readable medium and stores a computer program. When the computer program is executed by a processor, the steps of the UHPC composite beam early warning method based on deep learning as described above are implemented.

[0039] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the steps of the UHPC composite beam early warning method based on deep learning as described above.

[0040] This application obtains the structural information of ultra-high performance concrete composite beams, inputs this structural information into a preset state prediction model for processing, and obtains target characteristic states and target performance prediction results. Based on these target characteristic states and target performance prediction results, a multi-level early warning mechanism is used to detect and obtain detection results, and corresponding early warning operations are executed based on the detection results. A deep learning model is used to process the structural information of UHPC composite beams, automatically identifying key characteristic states and predicting performance. Real-time monitoring and risk warnings are achieved through a multi-level early warning mechanism, improving the accuracy of characteristic state identification and the reliability of performance predictions, enabling early risk detection and proactive early warning, and enhancing structural safety and design efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0042] Figure 1 This is a flow chart of the first embodiment of the UHPC composite beam early warning method based on deep learning in this application;

[0043] Figure 2 A cross-sectional structural diagram of a UHPC composite beam according to the first embodiment of the deep learning-based early warning method for UHPC composite beams of this application;

[0044] Figure 3 This is a schematic diagram of the characteristic state recognition results of the first embodiment of the UHPC composite beam early warning method based on deep learning in this application;

[0045] Figure 4 This is a schematic diagram of the performance prediction results of the first embodiment of the UHPC composite beam early warning method based on deep learning in this application;

[0046] Figure 5 This is a flow chart of the second embodiment of the UHPC composite beam early warning method based on deep learning in this application;

[0047] Figure 6 This is a flow chart of the third embodiment of the UHPC composite beam early warning method based on deep learning in this application;

[0048] Figure 7This is a schematic diagram of the module structure of the UHPC composite beam early warning device based on deep learning in an embodiment of the present application;

[0049] Figure 8 Schematic diagram of the equipment structure of the hardware operating environment involved in the UHPC composite beam early warning method based on deep learning in the embodiment of the present application.

[0050] The purpose, features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0051] It should be understood that the specific embodiments described herein are merely used to explain the technical solutions of the present application and are not intended to limit the present application.

[0052] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0053] With the widespread application of ultra-high-performance concrete (UHPC) in bridge engineering, UHPC-steel composite beams have attracted increasing attention due to their excellent mechanical properties and durability. UHPC composite beams not only offer high strength and durability, but also effectively reduce structural deadweight, improving their service life and economic efficiency. Consequently, UHPC composite beams have been widely used in modern bridge engineering, especially in large bridges and special structures. However, the widespread adoption of UHPC composite beams has placed higher demands on real-time monitoring and assessment of their performance. Traditional analysis and monitoring methods primarily rely on classical mechanics theory and empirical formulas, which have limitations when dealing with complex structural behavior and dynamic changes. First, characteristic state identification relies heavily on manual experience and pre-set judgment criteria, lacking automated and intelligent approaches. Second, performance predictions lack accuracy, especially when dealing with complex working conditions; the reliability of prediction results needs to be further improved. Third, traditional calculation methods are complex and time-consuming, failing to meet the demands of rapid design. Finally, early warning mechanisms suffer from lags, making real-time risk identification and proactive prevention difficult.

[0054] Therefore, this application proposes a UHPC composite beam early warning method based on deep learning. The main solution of the embodiment of this application is: by obtaining the structural information of the ultra-high performance concrete composite beam, the structural information is input into the preset state prediction model for processing to obtain the target characteristic state and target performance prediction results; based on the target characteristic state and target performance prediction results, detection is performed through a multi-level early warning mechanism to obtain the detection results, and corresponding early warning operations are executed according to the detection results.

[0055] Based on this, the embodiment of the present application provides a UHPC composite beam early warning method based on deep learning, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the UHPC composite beam early warning method based on deep learning in this application.

[0056] In this embodiment, the UHPC composite beam early warning method based on deep learning includes steps S10 to S40:

[0057] Step S10: Acquire structural information of the ultra-high performance concrete composite beam.

[0058] It should be noted that the structural information of ultra-high performance concrete (UHPC) composite beams covers multiple aspects, including geometric characteristics, material characteristics, and load characteristics. First, in terms of geometric characteristics, the specific parameters of UHPC beams and steel beams are key. Figure 2 The cross-sectional diagram of the UHPC composite beam shows that the UHPC beam has a flange width of 1200mm, a thickness of 120mm, a web width of 120mm, and a height of 480mm. The steel beam has a flange width of 400mm, a thickness of 25mm, a web thickness of 16mm, and a height of 575mm. Detailed reinforcement layout parameters are also included, such as the top longitudinal reinforcement using 12 steel bars with a diameter of 12mm and a cover thickness of 20mm; the bottom longitudinal reinforcement uses two steel bars with a diameter of 16mm and a cover thickness of 20mm.

[0059] Specifically, UHPC possesses extremely high compressive strength (fcu = 145 MPa) and tensile strength (ft = 10.87 MPa), an elastic modulus of 45.7 GPa, and specific design compressive strengths (fcud = 71.05 MPa) and tensile strengths (ftd = 7.61 MPa). Steel has an elastic modulus of 206.0 GPa, a yield strength of 355 MPa, and an ultimate strength of 512 MPa, with a design strength of 286.13 MPa. The performance parameters of rebar are also crucial, with an elastic modulus of 200.0 GPa, a yield strength of 400 MPa, and an ultimate strength of 565 MPa. Load characteristics primarily involve positive and negative bending moment parameters, which are crucial for assessing the structural response under various loading conditions. By normalizing these input features, dimensionality effects can be eliminated and the model training process optimized. For example, length dimensions can be divided by a reference span value, area parameters can be divided by a reference cross-sectional area value, and strength parameters can be divided by a baseline strength.

[0060] Step S20: input the structure information into a preset state prediction model for processing to obtain target feature state and target performance prediction results.

[0061] It should be noted that the acquired structural information including geometric features, material features and load features is standardized, including the normalization of length dimensions, area parameters and strength parameters, to eliminate dimensional effects. The sliding window method is used to extract time series features, and Gaussian noise injection is used to achieve data enhancement to ensure the generalization ability of the model.

[0062] Furthermore, the standardized structural information is input into a preset state deep model. The preset state prediction model is a hybrid neural deep learning model, comprising a feature extraction network, a temporal feature network, an attention layer, and an output layer. The feature extraction network comprises a first convolutional block, a second convolutional block, and a third convolutional block. Each of the first, second, and third convolutional blocks comprises two one-dimensional convolutional layers, a normalization layer, an activation function layer, and a maximum pooling layer. The temporal feature network comprises a bidirectional long short-term memory network layer and a normalization layer. The output layer comprises a feature state classification branch and a performance prediction branch. The feature state classification branch outputs the probability distribution of the feature state through a three-layer fully connected network, and the performance prediction branch outputs the performance index value through a three-layer fully connected network. According to the hybrid neural network model, deep learning of the characteristic state of ultra-high performance concrete composite beams is achieved by extracting local features through the feature extraction network, capturing temporal features through the bidirectional long short-term memory network, and weighting key features through the attention layer.

[0063] Furthermore, the hybrid neural network model is adopted because it can fully leverage the strengths of different neural network architectures to address complex feature extraction and prediction tasks. Specifically, feature extraction networks excel at processing local features and can automatically extract spatial hierarchical structures from data such as images and signals. When processing the geometric and material characteristics of UHPC composite beams, feature extraction networks can effectively capture local mechanical behavior. The bidirectional long short-term memory (BiLSTM) network can capture long-term dependencies in time series data and is particularly suitable for processing the temporal evolution of mechanical properties. When analyzing the temporal evolution of stress and strain, the BiLSTM provides more comprehensive temporal information. The attention mechanism can dynamically highlight the importance of key features, enabling the model to adaptively focus on the most helpful components for prediction. When identifying key characteristic states of UHPC composite beams (such as cracking and yielding), the attention mechanism can improve the model's accuracy and reliability. In this embodiment, local mechanical features are first extracted using three convolutional blocks connected in series. Each convolutional block contains two one-dimensional convolutional layers with 64, 128, and 256 channels, respectively, a kernel size of 3, and a normalization layer and ReLU activation function. Next, the temporal feature network employs a bidirectional LSTM network to capture the temporal evolution of mechanical properties. This bidirectional LSTM network consists of two BiLSTM layers, each with 128 and 64 hidden units, respectively. This effectively identifies dependencies across different timescales and is particularly suitable for analyzing stress and strain development. Furthermore, to highlight the influence of key states, the model incorporates a multi-head self-attention mechanism. Eight attention heads, each with a dimension of 64, are used to calculate feature correlations using a query / key / value matrix, enabling the model to adaptively focus on different stress and strain development scenarios. Finally, at the output layer, the model employs a multi-task learning framework, consisting of a feature state classification branch and a performance prediction branch. The feature state classification branch outputs the probability distribution of each feature state, such as UHPC first cracking and steel beam yielding, through a three-layer fully connected network (256-128-n dimensions, respectively). The performance prediction branch also utilizes a three-layer fully connected network (with linear activation functions) to output specific load-bearing capacity and deformation values. Through this hybrid neural network structure, the advantages of different neural network architectures can be fully utilized to achieve accurate identification of the characteristic state of UHPC composite beams and precise prediction of their performance.

[0064] Furthermore, step S20 also includes: inputting the structural information into a preset state prediction model for processing to obtain characteristic states; evaluating the characteristic states to obtain a confidence score; and outputting the characteristic state as a target characteristic state when the confidence score exceeds a preset threshold. If the confidence score does not exceed the preset threshold, a warning signal is issued, and the step of inputting the structural information into the preset state prediction model for processing to obtain characteristic states is repeated until the confidence score exceeds the preset threshold. Specifically, after obtaining preliminary characteristic states, these states are evaluated and a confidence score is calculated. This score is based on the model's accuracy in identifying various characteristic states and its ability to predict stress and strain development under specific working conditions. When the confidence score exceeds a preset threshold (in this embodiment, the preset threshold is set to 0.95), the model's interpretation of the current structural information is highly reliable, and the output characteristic state is considered the target characteristic state, such as a critical state such as first cracking in UHPC or yielding in steel beams. This mechanism ensures that only rigorously verified states are ultimately adopted, providing a solid foundation for engineering design. However, if the confidence score does not reach the preset threshold, it indicates that the current analysis contains uncertainty or error. In this case, an automatic warning signal is issued, indicating the need for further data analysis or model adjustment. Subsequently, the structural information input and processing steps are re-executed. This includes adjusting data preprocessing methods (such as improving normalization techniques or enhancing the dataset), optimizing weight and bias parameters (such as increasing the number of convolutional layers or adjusting the number of LSTM units), and even improving the model architecture itself (such as introducing a more complex attention mechanism). This cycle continues until the confidence score meets the requirements. This iterative optimization strategy not only improves the robustness and adaptability of the model, but also effectively reduces the risk of misjudgment, ensuring that the target feature state of each output is as close as possible to the actual situation, thereby providing strong protection for structural safety.

[0065] Step S30: Perform detection through a multi-level early warning mechanism based on the target feature state and the target performance prediction result to obtain a detection result.

[0066] It should be noted that when the model outputs the target characteristic state (such as the first cracking of UHPC, yielding of steel beams, etc.) and the target performance prediction results (such as the predicted values ​​of bearing capacity and deformation), this information will be input into the preset multi-level early warning system. The above-mentioned early warning system includes three core modules: state classifier, trend predictor and risk assessor. The state classifier identifies the state of the current structure based on the characteristic points on the M-φ curve; the trend predictor analyzes the changes in the slope of the curve to predict future development trends; and the risk assessor monitors key strain values ​​to assess potential risks. The system sets three levels of early warning thresholds based on the characteristic points on the M-φ curve. Figure 3The characteristic state identification result diagram shown in Figure 1 uses the M-φ curve to display each characteristic state point. Key points such as the UHPC cracking state (εtu=0.00293), the steel beam yielding state (εy=0.001723), the steel beam strengthening state (εst=0.0198) and the ultimate state (εu=0.035) are marked on the curve, and the corresponding strain values ​​are given. Figure 4 The performance prediction results are shown in the figure, which compares the predicted bearing capacity and deformation curves. At the beginning of loading, when the lower end of the UHPC web reaches the elastic limit (point B, φ = 0.00157 1 / m), the bending moment is 4380.5 kN·m. As loading continues, the lower steel flange first reaches the design stress (point I, φ = 0.00185 1 / m, M = 5161.4 kN·m) and then reaches the yield strain (point E, φ = 0.00228 1 / m, M = 6331.0 kN·m). When the upper UHPC edge reaches the design stress (point H, φ = 0.00431 1 / m), the bending moment increases to 8094.6 kN·m. Approaching the ultimate limit state, the lower end of the UHPC web reaches the ultimate strain (point C, φ = 0.00837 1 / m, M = 9031.3 kN·m), and finally the lower steel flange enters the strengthening stage (point F, φ = 0.01881 1 / m, M = 9131.6 kN·m). The model comprehensively considers multiple key factors during the prediction process. By analyzing the strain distribution at each characteristic point, such as the strain at the yield point E of the steel beam, which is -1.020‰ at the upper edge and 0.351‰ at the lower edge, the stress state of the component can be accurately determined. The model also tracks the changes in the position of the neutral axis, finding that it gradually increases from an initial position of 448.6 mm to 147.3 mm from the top edge of the cross-section. This trend reflects the process of stress redistribution in the cross-section. Furthermore, by analyzing the changes in the slope of the M-φ curve, the degradation of the structural stiffness can be clearly reflected.

[0067] Step S40: Execute corresponding warning operations according to the detection results.

[0068] It should be noted that when the bending moment reaches 4380.5 kN·m, the system automatically increases the data sampling frequency and activates the trend analysis module for more detailed monitoring. At this point, the system begins preparing emergency response plans, including but not limited to notifying relevant engineering personnel to strengthen on-site monitoring and prepare emergency supplies. The primary purpose of this phase is to be fully prepared for potential problems and to respond quickly to any emergencies. As loading continues, when the bending moment increases to 5161.4 kN·m, the system further escalates its response measures. At this point, real-time monitoring mode is activated. The system not only continuously tracks changes in key strain values ​​but also calls upon the loading control module to adjust the loading rate or suspend construction activities to avoid further structural stress. Simultaneously, the system sends early warning messages to all relevant personnel, ensuring that everyone is promptly informed of the current situation and can take necessary protective measures. This early intervention can effectively prevent minor issues from escalating into major incidents. When the bending moment reaches 8094.6 kN·m, the system immediately initiates emergency protective measures. These may include ceasing all construction work, evacuating on-site personnel, and implementing temporary reinforcement measures.

[0069] This embodiment provides a deep learning-based early warning method for UHPC composite beams. This method obtains structural information of ultra-high performance concrete composite beams and inputs this information into a preset state prediction model for processing, obtaining target characteristic states and target performance prediction results. Based on these target characteristic states and target performance prediction results, a multi-level early warning mechanism is used to detect and obtain detection results, and corresponding early warning operations are performed based on the detection results. By using a deep learning model to process the structural information of UHPC composite beams, key characteristic states are automatically identified and performance is predicted. Real-time monitoring and risk warnings are implemented through a multi-level early warning mechanism, improving the accuracy of characteristic state identification and the reliability of performance predictions. This enables early risk detection and proactive early warning, thereby enhancing structural safety and design efficiency.

[0070] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above embodiment 1 can be referred to the above introduction and will not be described in detail later. Figure 5 The UHPC composite beam early warning method based on deep learning further includes steps S201 to S202 before step S20:

[0071] Step S201: Acquire structural information sample data and build an initial state prediction model.

[0072] It should be noted that a series of representative sample data are collected, which cover the key geometric features, material properties and load conditions of composite beams. For example, the geometric features include the specific dimensional parameters of UHPC beams and steel beams, such as flange width, thickness, web width and height; the material properties involve the mechanical properties of UHPC, steel and steel bars, such as elastic modulus, compressive strength, yield strength, etc.; the load conditions mainly focus on positive and negative bending moment parameters. After obtaining these sample data, it is crucial to perform standardization. This step includes normalizing all input features to eliminate dimensional differences, extracting time series features using the sliding window method, and enhancing the data set by injecting Gaussian noise to ensure that the model has good generalization ability. In addition, the use of five-fold cross-validation technology can further evaluate the stability and accuracy of the model.

[0073] Furthermore, an initial state prediction model was constructed based on deep learning methods. This model comprises several key stages: first, a feature extraction network consisting of three convolutional blocks, each containing two one-dimensional convolutional layers (with 64, 128, and 256 channels, respectively). These layers, along with normalization layers, ReLU activation functions, and max pooling layers, gradually extract local features from the stress and strain fields. Next, a bidirectional LSTM network is used to capture the temporal evolution of mechanical properties. This design is particularly well-suited for handling time-varying structural responses. To highlight the impact of critical states, a multi-head self-attention mechanism is introduced, enabling the model to adaptively focus on different stress and strain developments. Finally, a multi-task learning framework is employed in the output layer, with separate branches for feature state classification and performance indicator prediction. The classification branch outputs the probability distribution of each feature state through a three-layer fully connected network, while the regression branch predicts specific load-bearing capacity and deformation values.

[0074] Step S202 : training the initial state prediction model based on the construction information sample data to obtain a preset state prediction model.

[0075] It should be noted that training the initial state prediction model involves first initializing its weights and bias parameters. Structural information sample data is then fed into the initial state prediction model for computation, resulting in characteristic state classification results and performance prediction results. These results are then weighted and calculated based on a loss function to obtain a total error. The gradients of the weights and bias parameters are then calculated using a backpropagation algorithm. Based on the gradients, the weights and bias parameters are iteratively updated using an optimization algorithm until a maximum number of iterations is reached or the total error converges to a preset threshold, resulting in the preset state prediction model. Specifically, the model's weights and bias parameters are initialized, typically using a random initialization method or the parameters of a pretrained model as a starting point. Next, the prepared structural information sample data is fed into the initial state prediction model, and forward propagation is used to obtain characteristic state classification results (e.g., first cracking of UHPC, yielding of steel beams, etc.) and performance prediction results (e.g., bearing capacity and deformation). This process utilizes the model's convolutional blocks to extract local mechanical features, the bidirectional LSTM network to capture temporal evolution features, and the multi-head self-attention mechanism to highlight the influence of key features. Subsequently, the feature state classification results and performance prediction results are evaluated and weighted based on a loss function to produce a total error. The loss function used here is typically a combination of cross-entropy loss (for classification tasks) and mean squared error loss (for regression tasks), corresponding to the results of the feature state classification branch and the performance metric prediction branch, respectively. This weighting ensures that the relative importance of the two types of tasks is appropriately accounted for in the calculation of the total error. Next, the gradients of the weights and bias parameters of each model layer are calculated based on the total error using the backpropagation algorithm. This process begins at the output layer and propagates the error forward layer by layer, calculating the contribution of each layer parameter to the total error. Based on the calculated gradients, an optimization algorithm (such as Adam or SGD) is then used to iteratively update the model parameters. The selection and adjustment of the optimization algorithm is crucial for accelerating convergence and improving model performance. The training process continues until a predetermined maximum number of iterations is reached or the total error converges to a preset threshold. During this period, the model gradually learns to more accurately identify feature states and predict structural properties while reducing error. To prevent overfitting, regularization techniques such as Dropout or L2 regularization are usually introduced, and a validation set is used to monitor the performance of the model to ensure its generalization ability.

[0076] Ultimately, the fully trained model becomes a preset state prediction model, which can efficiently and accurately complete feature state identification and performance prediction tasks in practical applications, providing strong technical support for the safety monitoring of UHPC composite beams.

[0077] After obtaining the preset state prediction model, the method further includes: evaluating the preset state prediction model to obtain an evaluation result; if the evaluation result does not meet the detection requirements, retraining the preset state prediction model until the detection requirements are met.

[0078] This embodiment collects structural information of UHPC composite beams, constructs and trains an initial state prediction model, and iterates the optimization algorithm until the error converges to obtain a preset state prediction model. This improves the accuracy of characteristic state identification and the reliability of performance prediction, achieves early risk warning, and enhances structural safety and design efficiency.

[0079] Based on the first embodiment of the present application, in the third embodiment of the present application, the same or similar contents as those in the first embodiment can be referred to the above introduction and will not be described in detail later. Figure 6 The UHPC composite beam early warning method based on deep learning, step S30, further includes steps S301 to S303:

[0080] Step S301: Obtain a first threshold, a second threshold, and a third threshold according to a multi-level warning mechanism.

[0081] It should be noted that the above-mentioned first threshold is smaller than the above-mentioned second threshold, and the above-mentioned second threshold is smaller than the above-mentioned third threshold. In this embodiment, the first threshold corresponds to a bending moment reaching 4380.5 kN·m, the second threshold corresponds to a bending moment reaching 5161.4 kN·m, and the third threshold corresponds to a bending moment reaching 8094.6 kN·m.

[0082] Step S302: Obtain the composite beam bending moment according to the target characteristic state and the target performance prediction result.

[0083] It's important to note that after processing the structural information sample, the deep learning model outputs a series of target characteristic states (e.g., first cracking in UHPC, yielding of steel beams, etc.) and target performance predictions (e.g., load-bearing capacity and deformation). These predictions not only provide information on the structure's response under varying stress and strain conditions but also help identify critical stress stages.

[0084] Specifically, by analyzing the characteristic points on the M-φ curve, the bending moment values ​​of the composite beam under specific operating conditions can be accurately determined. For example, when the lower end of the UHPC web reaches the elastic limit, the corresponding bending moment value is 4380.5 kN·m, marking the beginning of the structure's nonlinear operating range. As loading continues, when the lower steel flange reaches the design stress, the bending moment increases to 5161.4 kN·m, indicating that the structure is bearing a greater load. Furthermore, when the lower steel flange reaches the yield strain, the bending moment value rises to 6331.0 kN·m, a key indicator that the structure is approaching its ultimate load-bearing capacity. Finally, when the upper UHPC edge reaches the design stress, the bending moment value reaches as high as 8094.6 kN·m, and the maximum bending moment at the ultimate strain is 9031.3 kN·m.

[0085] Step S303 : Compare the composite beam bending moment with the first threshold, the second threshold, and the third threshold to obtain a detection result.

[0086] It should be noted that the bending moment value predicted by the deep learning model is compared and analyzed with the threshold corresponding to the set key feature point to obtain the detection result.

[0087] Furthermore, step S302 also includes: when the bending moment of the composite beam is less than a first threshold, the detection result is determined to be risk-free, and normal monitoring and recording of key parameters are performed. Specifically, when the bending moment of the UHPC composite beam is less than 4380.5 kN·m, the detection result indicates that the structure is currently in a risk-free state. This means that the structure is still within its elastic operating range and has not yet shown any signs of potential failure or damage. In this case, the system will continue to perform normal monitoring and record key parameters to ensure continuous tracking of the structural health.

[0088] When the composite beam bending moment exceeds the first threshold and is less than the second threshold, a Level 1 warning is detected, and the system initiates actions to increase the frequency of structural information acquisition and prepare emergency response plans. Specifically, when the bending moment of the UHPC composite beam exceeds 4380.5 kN·m and is less than 5161.4 kN·m, the system detects a Level 1 warning. This indicates that the structure has entered a nonlinear operating range. While not yet a dangerous state, potential risk signals are present, requiring closer attention and response measures. In this case, the system initiates a series of actions to increase the frequency of structural information acquisition and initiate emergency response plan preparation. First, the data sampling frequency is significantly increased to more accurately capture any subtle changes in the structural response. This includes monitoring not only key parameters such as bending moment, strain, and curvature, but may also be expanded to other relevant factors such as temperature changes and humidity effects, ensuring a comprehensive understanding of the current structural condition. Simultaneously, the system activates a trend analysis module, which uses a deep learning model to predict future short-term structural behavior trends. This module compares and analyzes historical data with real-time data to identify possible development patterns and potential risk points. Based on these analysis results, engineers can develop more detailed contingency plans, covering everything from temporary reinforcement measures to emergency evacuation plans, to ensure a rapid response should the situation deteriorate. Furthermore, at the Level 1 alert stage, the system notifies all relevant parties, including project managers, on-site engineers, and safety supervisors, ensuring everyone is aware of the current situation and prepared accordingly.

[0089] When the composite beam bending moment exceeds the second threshold and is less than the third threshold, a Level 2 warning is detected, and real-time monitoring is initiated to send an alert to relevant personnel. Specifically, when the bending moment of the UHPC composite beam exceeds 5161.4 kN·m and is less than 8094.6 kN·m, the system detects a Level 2 warning. This indicates that the structure is experiencing significant loads, approaching its design limits, presenting a significant risk signal and necessitating more stringent monitoring and response measures. In Level 2, the system activates real-time monitoring mode, collecting data on key parameters such as bending moment, strain, and curvature at a higher frequency and performing real-time analysis. This high-density data collection helps capture even the slightest changes, ensuring timely identification of potential issues. Simultaneously, the system sends alerts to all relevant personnel, including project managers, on-site engineers, and safety supervisors, ensuring everyone is fully informed of the current situation and prepared to respond.

[0090] When the bending moment of the composite beam exceeds the third threshold, the detection result is a level 3 warning, and an emergency response is triggered for maintenance. Specifically, when the bending moment value of the UHPC composite beam exceeds 8094.6 kN·m, the system detection result is a level 3 warning, indicating that the structure has approached or reached its ultimate bearing capacity and there is a serious risk. In this case, the emergency response mechanism is immediately triggered to ensure the safety of personnel and prevent further deterioration of structural damage. The system automatically performs a series of emergency maintenance operations. All ongoing construction activities will be immediately suspended, and on-site personnel will be quickly evacuated to a safe area. At the same time, the pre-established emergency plan will be activated, including but not limited to the implementation of temporary reinforcement measures, the installation of support structures, and necessary local repair work to stabilize the structural state and avoid catastrophic damage.

[0091] This embodiment sets thresholds based on a multi-level early warning mechanism, predicts the bending moment of the composite beam through a deep learning model and compares it with the threshold to achieve hierarchical early warning, improves the accuracy and response speed of structural safety monitoring, and realizes early risk identification and intervention.

[0092] This application also provides a UHPC composite beam early warning device based on deep learning, please refer to Figure 7 , the device comprises:

[0093] The acquisition module 10 is used to obtain the structural information of the ultra-high performance concrete composite beam.

[0094] The processing module 20 is used to input the structural information into the preset state prediction model for processing to obtain the target characteristic state and target performance prediction results. The preset state prediction model is a hybrid neural network model. The preset state prediction model includes a feature extraction network, a temporal feature network, an attention layer and an output layer. The feature extraction network includes a first convolution block, a second convolution block and a third convolution block. The first convolution block, the second convolution block and the third convolution block each contain two one-dimensional convolution layers, a normalization layer, an activation function layer and a maximum pooling layer. The temporal feature network includes a bidirectional long short-term memory network layer and a normalization layer. The output layer includes a feature state classification branch and a performance prediction branch. The feature state classification branch outputs the probability distribution of the feature state through a three-layer fully connected network. The performance prediction branch outputs the performance index value through a three-layer fully connected network. According to the hybrid neural network model, local feature extraction is performed through the feature extraction network, temporal feature capture is performed through the bidirectional long short-term memory network, and the attention layer is weighted in key features to achieve deep learning of the characteristic state of the ultra-high performance concrete composite beam.

[0095] The detection module 30 is used to perform detection through a multi-level early warning mechanism based on the target feature state and the target performance prediction result to obtain the detection result.

[0096] The execution module 40 is used to execute corresponding warning operations according to the detection results.

[0097] The deep learning-based UHPC composite beam early warning device provided in this application utilizes the deep learning-based UHPC composite beam early warning method described in the aforementioned embodiment to address the technical problem of obtaining the status of a UHPC composite beam and executing corresponding early warning operations. Compared to the prior art, the beneficial effects of the deep learning-based UHPC composite beam early warning device provided in this application are the same as those of the deep learning-based UHPC composite beam early warning method described in the aforementioned embodiment. Other technical features of the deep learning-based UHPC composite beam early warning device are the same as those disclosed in the aforementioned embodiment and are not further elaborated here.

[0098] In one embodiment, the processing module 20 is further configured to obtain sample data of construction information and construct an initial state prediction model; and train the initial state prediction model based on the sample data of construction information to obtain a preset state prediction model.

[0099] In one embodiment, the processing module 20 is also used to initialize the weights and bias parameters of the initial state prediction model; input the construction information sample data into the initial state prediction model for calculation to obtain the feature state classification results and performance prediction results; calculate and weight the feature state classification results and performance prediction results according to the loss function to obtain the total error value; calculate through the back propagation algorithm to obtain the gradient of the weight and bias parameters; iteratively update the weight and bias parameters through the optimization algorithm according to the gradient until the maximum number of iterations is reached or the total error value converges to the preset threshold, thereby obtaining the preset state prediction model.

[0100] In one embodiment, the processing module 20 is further used to input the construction information into a preset state prediction model for processing to obtain a characteristic state; perform an evaluation based on the characteristic state to obtain a confidence score; when the confidence score exceeds a preset threshold, output the characteristic state as a target characteristic state; when the confidence score does not exceed the preset threshold, issue a warning signal, and execute the step of inputting the construction information into the preset state prediction model for processing to obtain the characteristic state until the confidence score exceeds the preset threshold.

[0101] In one embodiment, the detection module 30 is further used to obtain a first threshold, a second threshold, and a third threshold according to the multi-level warning mechanism, wherein the first threshold is less than the second threshold, and the second threshold is less than the third threshold; obtain the bending moment of the composite beam according to the target characteristic state and the target performance prediction results; and compare the bending moment of the composite beam with the first threshold, the second threshold, and the third threshold to obtain a detection result.

[0102] In one embodiment, the execution module 40 is also used to, when the combined beam bending moment is less than the first threshold, obtain a detection result of no risk, and perform normal monitoring and record key parameters; when the combined beam bending moment is greater than the first threshold and the combined beam bending moment is less than the second threshold, obtain a detection result of a first-level warning, and perform operations of increasing the frequency of obtaining structural information and preparing emergency plans; when the combined beam bending moment is greater than the second threshold and the combined beam bending moment is less than the third threshold, obtain a detection result of a second-level warning, and perform operations of sending warning information to relevant personnel through real-time monitoring; when the combined beam bending moment is greater than the third threshold, obtain a detection result of a third-level warning, and perform operations of triggering an emergency response for maintenance.

[0103] The present application provides a UHPC composite beam early warning device based on deep learning. The UHPC composite beam early warning device based on deep learning includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the UHPC composite beam early warning method based on deep learning in the above-mentioned embodiment 1.

[0104] Reference below Figure 8 , which shows a schematic structural diagram of a deep learning-based UHPC composite beam early warning device suitable for implementing an embodiment of the present application. The deep learning-based UHPC composite beam early warning device in the embodiment of the present application can include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), and in-vehicle terminals (e.g., in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. Figure 8 The deep learning-based UHPC composite beam early warning device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0105] like Figure 8The deep learning-based UHPC composite beam early warning device may include a processing device 1001 (e.g., a central processing unit, graphics processing unit, etc.), which can perform various appropriate actions and processes based on programs stored in ROM (Read Only Memory) 1002 or programs loaded from storage device 1003 into RAM (Random Access Memory) 1004. RAM 1004 also stores various programs and data required for the operation of the deep learning-based UHPC composite beam early warning device. Processing device 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 1003 including, for example, a magnetic tape, hard disk, etc.; and communication devices 1009. Communication devices 1009 can allow the deep learning-based UHPC composite beam early warning device to communicate wirelessly or wired with other devices to exchange data. While the figure shows a deep learning-based UHPC composite beam early warning device with various systems, it should be understood that implementation or presence of all the illustrated systems is not required. More or fewer systems may alternatively be implemented or present.

[0106] In particular, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes program code for executing the method described in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are performed.

[0107] The deep learning-based UHPC composite beam early warning device provided in this application utilizes the deep learning-based UHPC composite beam early warning method described in the aforementioned embodiment to address the technical problem of obtaining the status of a UHPC composite beam and executing corresponding early warning operations. Compared to the prior art, the deep learning-based UHPC composite beam early warning device provided in this application achieves the same beneficial effects as the deep learning-based UHPC composite beam early warning method described in the aforementioned embodiment. Other technical features of this deep learning-based UHPC composite beam early warning device are the same as those disclosed in the aforementioned embodiment and are not further elaborated upon here.

[0108] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0109] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0110] The present application provides a computer-readable medium having computer-readable program instructions (i.e., a computer program) stored thereon for performing calculations, and obtaining machine-readable program instructions for executing the UHPC composite beam early warning method based on deep learning in the above-mentioned embodiment.

[0111] The computer-readable medium provided in this application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination thereof. More specific examples of computer-readable media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. Calculations are performed in this embodiment to obtain a machine-readable medium that may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0112] The above-mentioned computer-readable medium may be included in the UHPC composite beam early warning device based on deep learning; or it may exist independently without being assembled into the UHPC composite beam early warning device based on deep learning.

[0113] The computer-readable medium carries one or more programs. When executed by the deep learning-based UHPC composite beam early warning device, the one or more programs enable the deep learning-based UHPC composite beam early warning device to write computer program code for performing the operations of the present application in one or more programming languages, or a combination thereof. These programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0114] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, program segment or a part of code, and the module, program segment or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented by a dedicated hardware-based system that performs the specified function or operation, or can be implemented by a combination of dedicated hardware and computer instructions.

[0115] The modules described in the embodiments of the present application may be implemented in software or hardware, wherein the name of a module does not necessarily limit the unit itself.

[0116] The computer-readable medium provided in this application stores computer-readable program instructions (i.e., a computer program) for executing the aforementioned deep learning-based early warning method for UHPC composite beams. This computer-readable medium addresses the technical problem of obtaining the status of UHPC composite beams and executing corresponding early warning operations. Compared to the prior art, the beneficial effects of the computer-readable medium provided in this application are similar to those of the deep learning-based early warning method for UHPC composite beams provided in the aforementioned embodiments, and are not further elaborated here.

[0117] The present application also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the above-mentioned UHPC composite beam early warning method based on deep learning.

[0118] The computer program product provided in this application solves the technical problem of how to obtain the status of UHPC composite beams and execute corresponding early warning operations. Compared with the existing technology, the beneficial effects of the computer program product provided in this application are the same as those of the deep learning-based early warning method for UHPC composite beams provided in the above-mentioned embodiment, and will not be elaborated here.

[0119] The above description is only part of the embodiments of the present application and does not limit the patent scope of the present application. All equivalent structural transformations made by using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect application in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A UHPC composite beam early warning method based on deep learning, characterized in that: The method comprises: Obtain structural information of ultra-high performance concrete composite beams; Inputting the structural information into a preset state prediction model for processing to obtain target feature state and target performance prediction results, the preset state prediction model is a hybrid neural network model, the preset state prediction model includes a feature extraction network, a temporal feature network, an attention layer and an output layer, the feature extraction network includes a first convolution block, a second convolution block and a third convolution block, the first convolution block, the second convolution block and the third convolution block each include two one-dimensional convolution layers, a normalization layer, an activation function layer and a maximum pooling layer, the temporal feature network includes a bidirectional long short-term memory network layer and a normalization layer, the output layer includes a feature state classification branch and a performance prediction branch, the feature state classification branch outputs the probability distribution of the feature state through a three-layer fully connected network, and the performance prediction branch outputs the performance index value through a three-layer fully connected network, according to the hybrid neural network model, the feature extraction network performs local feature extraction, the bidirectional long short-term memory network performs temporal feature capture and the attention layer performs key feature weighting to achieve deep learning of the feature state of the ultra-high performance concrete composite beam; Based on the target characteristic state and the target performance prediction result, a multi-level early warning mechanism is used to detect and obtain the detection results. The characteristic state points are displayed using an M-φ curve. The cracking state of the ultra-high performance concrete composite beam, the yielding state of the steel beam, the strengthening state of the steel beam and the ultimate state are marked on the curve, and the corresponding strain values ​​are given; Execute corresponding warning operations according to the detection results; The step of performing detection based on the target characteristic state and the target performance prediction result through a multi-level early warning mechanism to obtain a detection result includes: According to the multi-level warning mechanism, a first threshold, a second threshold, and a third threshold are obtained, wherein the first threshold is less than the second threshold, and the second threshold is less than the third threshold; Obtaining a composite beam bending moment according to the target characteristic state and the target performance prediction result; Comparing the composite beam bending moment with the first threshold, the second threshold, and the third threshold to obtain a detection result; After the step of comparing the composite beam bending moment with the first threshold, the second threshold, and the third threshold to obtain a detection result, the method further includes: When the bending moment of the composite beam is less than the first threshold, the detection result is obtained as no risk, and normal monitoring and recording of key parameters are performed; When the combined beam bending moment is greater than the first threshold and the combined beam bending moment is less than the second threshold, a detection result is obtained as a first-level warning, and operations of increasing the frequency of obtaining structural information and preparing an emergency plan are performed; When the combined beam bending moment is greater than the second threshold and the combined beam bending moment is less than the third threshold, a detection result is obtained as a second-level warning, and an operation of sending a warning message to relevant personnel through real-time monitoring is executed; When the bending moment of the composite beam is greater than the third threshold, the detection result is a level 3 warning, and an operation of triggering an emergency response for maintenance is executed.

2. The method according to claim 1, wherein Before the step of inputting the construction information into a preset state prediction model for processing to obtain target characteristic state and target performance prediction results, the method includes: Obtain structural information sample data and build an initial state prediction model; The initial state prediction model is trained based on the construction information sample data to obtain a preset state prediction model.

3. The method according to claim 2, wherein The step of training the initial state prediction model based on the construction information sample data to obtain a preset state prediction model includes: Initializing weight and bias parameters of the initial state prediction model; Inputting the structural information sample data into the initial state prediction model for calculation to obtain characteristic state classification results and performance prediction results; Calculating and weighting the feature state classification result and the performance prediction result according to the loss function to obtain a total error value; Obtaining the gradients of the weight and bias parameters by back propagation algorithm calculation; The weight and bias parameters are iteratively updated according to the gradient through an optimization algorithm until a maximum number of iterations is reached or the total error value converges to a preset threshold, thereby obtaining a preset state prediction model.

4. The method according to claim 1, wherein Before the step of inputting the construction information into a preset state prediction model for processing to obtain target characteristic state and target performance prediction results, the method includes: Inputting the structural information into a preset state prediction model for processing to obtain a characteristic state; Performing an evaluation based on the feature status to obtain a confidence score; When the confidence score exceeds a preset threshold, the feature state is output as the target feature state; When the confidence score does not exceed the preset threshold, an early warning signal is issued, and the step of inputting the construction information into a preset state prediction model for processing to obtain a characteristic state is executed until the confidence score exceeds the preset threshold.

5. A UHPC composite beam early warning device based on deep learning, characterized in that: The device comprises: An acquisition module, used to obtain structural information of ultra-high performance concrete composite beams; A processing module is configured to input the structural information into a preset state prediction model for processing to obtain target characteristic state and target performance prediction results, wherein the preset state prediction model is a hybrid neural network model, comprising a feature extraction network, a temporal feature network, an attention layer, and an output layer. The feature extraction network comprises a first convolution block, a second convolution block, and a third convolution block, wherein the first convolution block, the second convolution block, and the third convolution block each comprise two one-dimensional convolution layers, a normalization layer, an activation function layer, and a maximum pooling layer. The temporal feature network comprises a bidirectional long short-term memory network layer and a normalization layer. The output layer comprises a characteristic state classification branch and a performance prediction branch. The characteristic state classification branch outputs a probability distribution of the characteristic state through a three-layer fully connected network, and the performance prediction branch outputs a performance index value through a three-layer fully connected network. According to the hybrid neural network model, local feature extraction is performed by the feature extraction network, temporal feature capture is performed by the bidirectional long short-term memory network, and key feature weighting is performed by the attention layer to achieve deep learning of the characteristic state of the ultra-high performance concrete composite beam. a detection module configured to perform detection based on the target characteristic state and the target performance prediction result through a multi-level early warning mechanism to obtain a detection result, use an M-φ curve to display characteristic state points, mark the cracking state of the ultra-high performance concrete composite beam, the yielding state of the steel beam, the strengthening state of the steel beam, and the ultimate state on the curve, and provide corresponding strain values; and further configured to obtain a first threshold value, a second threshold value, and a third threshold value according to the multi-level early warning mechanism, wherein the first threshold value is less than the second threshold value, and the second threshold value is less than the third threshold value; obtain a composite beam bending moment based on the target characteristic state and the target performance prediction result; and compare the composite beam bending moment with the first threshold value, the second threshold value, and the third threshold value to obtain a detection result; An execution module is used to perform corresponding early warning operations according to the detection results; it is also used to, when the combined beam bending moment is less than the first threshold, obtain a detection result of no risk, and perform normal monitoring and record key parameters; when the combined beam bending moment is greater than the first threshold and the combined beam bending moment is less than the second threshold, obtain a detection result of a first-level early warning, and perform operations of increasing the frequency of obtaining structural information and preparing emergency plans; when the combined beam bending moment is greater than the second threshold and the combined beam bending moment is less than the third threshold, obtain a detection result of a second-level early warning, and perform operations of sending early warning information to relevant personnel through real-time monitoring; when the combined beam bending moment is greater than the third threshold, obtain a detection result of a third-level early warning, and perform operations of triggering an emergency response for maintenance.

6. A UHPC composite beam early warning device based on deep learning, characterized in that: The device includes: a memory, a processor, and a deep learning-based UHPC composite beam early warning program stored in the memory and running on the processor, wherein the deep learning-based UHPC composite beam early warning program is configured to implement the steps of the deep learning-based UHPC composite beam early warning method according to any one of claims 1 to 4.

7. A medium, characterized in that The medium stores a UHPC composite beam early warning program based on deep learning. When the UHPC composite beam early warning program based on deep learning is executed by the processor, the steps of the UHPC composite beam early warning method based on deep learning according to any one of claims 1 to 4 are implemented.

Images