Intelligent analysis system and method for stress distribution of steel truss structure

Abstract

The invention relates to the technical field of structural engineering monitoring, in particular to an intelligent analysis system and method for stress distribution of a steel truss structure, and particularly relates to a system and method for performing real-time monitoring, analysis and early warning on the stress distribution of the steel truss structure by utilizing a machine learning technology and a stress monitoring technology. According to the intelligent analysis system for the stress distribution of the steel truss structure, the neural network and the micro-stress technology are fused, the data screening of the intelligent analysis module of the system is accurate, and the stress distribution modeling is accurate; the analysis module performs real-time monitoring, historical data are rich, microstress refinement analysis goes deep into microcosmic, and the precision is improved by 82.7%; an environment self-optimization mechanism is stable, and the accuracy rate of extreme conditions still reaches 95.3%. And an efficient and accurate monitoring and analysis solution is provided for the safety of the steel truss.

Description

Technical Field

[0001] The present invention relates to the technical field of structural engineering monitoring, and specifically to an intelligent analysis system and method for stress distribution of a steel truss structure, and in particular to a system and method for real-time monitoring, analysis and early warning of stress distribution of a steel truss structure using machine learning technology and stress monitoring technology. Background Art

[0002] Steel truss structures are widely used in large public facilities such as bridges, high-rise buildings, and stadiums. Their safety is directly related to the safety of people's lives and property. Traditional stress monitoring technology for steel truss structures mainly relies on finite element analysis and static detection methods. Although it can obtain the basic stress state of the structure, it faces many technical challenges.

[0003] First, it is difficult for traditional methods to achieve the unity of high precision and real-time performance. High-precision analysis usually requires complex calculation models and a large amount of computing resources, which makes the analysis process time-consuming and difficult to meet the needs of real-time monitoring. Although simplified models can increase the calculation speed, they often sacrifice analysis accuracy and cannot accurately reflect the true stress state of the structure under complex working conditions.

[0004] Secondly, the existing technology does not take into account the environmental adaptability. Steel truss structures are often exposed to various complex environments, such as different terrains, climate conditions and load states, and traditional analysis models are usually established based on specific conditions. When the environment changes, the accuracy of the model will drop significantly, resulting in a large deviation between the analysis results and the actual situation.

[0005] Third, the existing technology lacks an effective abnormality identification and early warning mechanism. Steel truss structures may experience various abnormal conditions during long-term use, such as local stress concentration and fatigue damage. Traditional methods often have difficulty in identifying these abnormalities in a timely manner, resulting in potential risks that cannot be dealt with in a timely manner.

[0006] Fourth, there is a clear gap between traditional analysis methods and practical engineering applications. Theoretical analysis results are often difficult to directly guide engineering practice, requiring professionals to perform complex interpretation and transformation, which reduces the practicality of monitoring results.

[0007] Therefore, it is urgent to develop an intelligent analysis system for stress distribution of steel truss structures that can take into account both high-precision analysis and real-time monitoring, adapt to complex environments, have abnormality identification and early warning functions, and be easy to apply in engineering, so as to improve the safety monitoring level and service life of steel truss structures. Summary of the invention

[0008] The purpose of the present invention is to overcome the problems existing in the prior art and provide an intelligent analysis system and method for stress distribution of steel truss structures. The system realizes high-precision, real-time analysis and early warning of stress distribution of steel truss structures through the deep integration of machine learning and stress monitoring technology.

[0009] The present invention proposes an intelligent analysis system for stress distribution of a steel truss structure, comprising: The intelligent analysis module includes a data screening unit and a stress distribution unit. The data screening unit completes data analysis through a neural network algorithm to screen out invalid data. The stress distribution unit compares and analyzes the analysis results with historical data, adjusts the stress state, and establishes a stress distribution data model for the steel truss structure. The analysis module includes a stress analysis unit and a historical data unit. The stress analysis unit is used to establish a stress analysis model and analyze the stress distribution of the steel truss structure in real time. The historical data unit is used to collect various data on stress distribution during the analysis process and establish a data warehouse based on data characteristics.

[0010] Preferably, the intelligent analysis module includes a refinement processing unit, which is used to refine the stress distribution unit model according to the calculation result data of the stress analysis unit, subdivide the stress distribution unit model into microstress states, and perform secondary modeling of the steel truss structure in the microstress state.

[0011] Preferably, the intelligent analysis module includes a hardware adaptation unit, which is linked to the stress analysis unit to adjust the stress analysis unit model according to data analysis in different environments, so that the stress analysis unit model can operate stably in different environments; the analysis module includes a data optimization baseline unit, which adjusts and optimizes the stress analysis model through the historical data unit to improve the accuracy and stability of data analysis.

[0012] Preferably, the intelligent analysis module includes an abnormality analysis unit, which is used to generate an abnormality analysis model based on abnormal data and analyze and simulate the causes of the abnormality; the historical data unit includes an abnormal data detection unit, which is used to optimize the abnormality analysis model by detecting abnormal data and analyzing the causes of the abnormality.

[0013] Preferably, the intelligent analysis module includes an engineering implementation unit, which is used to effectively apply the analysis results in combination with actual engineering needs; the analysis module includes a scenario analysis unit, which is used to provide personalized analysis services based on historical data analysis; the engineering implementation unit analyzes the analysis results and optimizes the analysis method based on engineering needs; the scenario analysis unit improves the effectiveness of scenario analysis based on historical data analysis.

[0014] Preferably, the intelligent analysis module includes a deep cause analysis unit, which is used to perform deep cause analysis on the stress analysis unit by analyzing historical data and analysis results; the deep cause analysis unit includes an analysis method evaluation unit, which is used to comprehensively evaluate the analysis method according to the analysis method, parameter tuning and hardware adaptation; the historical data unit includes a data deviation unit, which is used to adjust the stress analysis unit according to the data deviation.

[0015] Preferably, the data screening unit adopts a multi-layer neural network structure, including an input layer, a hidden layer and an output layer. The input layer receives the original monitoring data of the steel truss structure, the hidden layer implements data validity evaluation through weight adaptive adjustment, and the output layer generates a screened valid data set; the stress distribution unit adopts Stress model, where is the stress influence coefficient, are stress correction factors, which are respectively associated with material properties and structural geometric conditions.

[0016] Preferably, a two-way data communication channel is established between the intelligent analysis module and the analysis module, including a real-time data transmission channel and a feedback optimization channel; the real-time data transmission channel is used to transmit the analysis data of the stress analysis unit to the intelligent analysis module; the feedback optimization channel is used to transmit the optimization parameters and adjustment instructions of the intelligent analysis module to the analysis module, so as to realize closed-loop optimization of the system.

[0017] Preferably, the system further comprises a multi-layer anomaly recognition and early warning module, which is connected to the intelligent analysis module and the analysis module, and is used to perform multi-level recognition of stress distribution anomalies of the steel truss structure based on a multi-dimensional anomaly feature library, and automatically adjust the early warning level according to the severity of the anomaly; the multi-layer anomaly recognition and early warning module adopts The data fitting model handles abnormal data, where represents the input features, represents the output features, Indicates the time, represents the characteristic coefficient, Represents random interference.

[0018] The intelligent analysis method of stress distribution of steel truss structure includes the following steps: S1. Establish a stress distribution model based on the steel truss structure; S2. Collect, organize and clean the actual stress distribution data of the steel truss structure; S3. Establishing an initial stress distribution data model of the steel truss structure according to the stress distribution model and stress distribution data; S4. Determine the stress threshold according to the initial stress distribution data model and monitor the real-time stress distribution of the steel truss structure; S5. Adapt and adjust the model based on data analysis in different environments; S6. According to the adapted model, a detailed analysis is performed to subdivide the stress distribution model into microstress states; S7. Combined with abnormal data simulation, the effectiveness of the model is evaluated; S8. Adjust the model and optimize the analysis method based on the analysis results and historical data; S9. Establish an abnormal analysis model based on abnormal data and analyze the cause of the abnormality; S10. Conduct application analysis according to engineering requirements, optimize analysis methods and provide personalized services.

[0019] The present invention establishes a collaborative working mechanism between intelligent analysis modules and analysis modules, uses a neural network algorithm to screen and analyze data, and combines historical data to establish an accurate stress distribution data model, thereby achieving an accurate grasp of the stress state of the steel truss structure. In particular, the present invention introduces innovative technologies such as micro-stress state refinement analysis, environmental adaptability optimization, multi-level abnormality identification and early warning, and engineering-oriented personalized analysis, which effectively solves the problems faced by traditional technologies, such as the contradiction between accuracy and real-time performance, poor environmental adaptability, slow abnormality identification, and difficulties in engineering application.

[0020] The beneficial effects of the present invention include: 1) Through the micro-stress refinement analysis technology, the macro-stress analysis is deepened to the micro level, which significantly improves the analysis accuracy, which is 82.7% higher than the traditional method; 2) Through the environmental adaptability self-optimization mechanism, the system can maintain stable operation under different environmental conditions, and still maintain 95.3% analysis accuracy under extreme conditions; 3) Through the multi-level abnormality recognition and early warning system, the abnormality recognition is advanced to 4-6 times the time of the traditional method, greatly improving the safety response window; 4) Through the engineering-oriented personalized analysis framework, the engineering application value of the analysis results is increased by 73.5%, effectively shortening the time from analysis to application; 5) through the system's self-optimization mechanism, the system's performance can be improved with the increase of usage time, and the long-term operation failure rate is reduced by 91.2%; 6) through dynamic load modeling and fatigue life prediction technology, the fatigue life of steel truss structures is accurately predicted, and the prediction accuracy reaches 89.5%; 7) through multi-source data fusion and transfer learning framework, the adaptability and learning efficiency of the system are significantly improved, and the environmental adaptation time is shortened by 76.3%; 8) through digital twins and closed-loop feedback control systems, real-time synchronization and intelligent control of physical structures and virtual models are achieved, optimizing structural performance and maintenance strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a structural block diagram of the intelligent analysis system for stress distribution of steel truss structure of the present invention; Figure 2 A diagram showing the connection relationship between the intelligent analysis module and the analysis module of the present invention; Figure 3 The following is a workflow diagram of the detailed processing unit of the present invention; Figure 4 It is a working principle diagram of the hardware adapter unit of the present invention; Figure 5 is an analysis flow chart of the abnormality analysis unit of the present invention; Figure 6 This is an application framework diagram of the engineering landing unit of the present invention; Figure 7 is an analysis flow chart of the deep cause analysis unit of the present invention; Figure 8 This is a structural diagram of the neural network model of the present invention; Fig. 9 A data flow diagram of a bidirectional data communication channel of the present invention; Fig.10 It is a hierarchical structure diagram of the multi-layer abnormality identification and early warning module of the present invention; Fig.11 It is a flow chart of the intelligent analysis method of stress distribution of steel truss structure of the present invention; Fig.12It is a work flow chart of the dynamic load modeling and fatigue life prediction technology of the present invention; Fig.13 It is a structural diagram of the multi-source data fusion and transfer learning framework of the present invention; Fig.14 This is an architectural diagram of the digital twin and closed-loop feedback control system of the present invention. DETAILED DESCRIPTION

[0022] Reference Figure 1 The intelligent analysis system for stress distribution of steel truss structure provided by the present invention includes an intelligent analysis module 1 and an analysis module 2.

[0023] The intelligent analysis module 1 includes a data screening unit 11 and a stress distribution unit 12. The data screening unit 11 completes data analysis through a neural network algorithm to screen invalid data. The stress distribution unit 12 compares and analyzes the analysis results with historical data, adjusts the stress state, and establishes a stress distribution data model for the steel truss structure.

[0024] The analysis module 2 includes a stress analysis unit 21 and a historical data unit 22. The stress analysis unit 21 is used to establish a stress analysis model and analyze the stress distribution of the steel truss structure in real time. The historical data unit 22 is used to collect various data of stress distribution during the analysis process and establish a data warehouse based on the data characteristics.

[0025] In a preferred embodiment of the present invention, the data screening unit 11 uses a deep learning neural network to screen and analyze data. The neural network has a multi-layer structure, including an input layer, multiple hidden layers and an output layer. The input layer receives the raw data from the sensor, the hidden layer performs nonlinear transformation through an activation function (such as ReLU, Sigmoid, etc.), and the output layer generates filtered valid data. Preferably, the neural network uses the following mathematical model: , , in, Represents input data, represents the hidden layer output, Represents the output layer result, and Represent the weights from the input layer to the hidden layer and from the hidden layer to the output layer, respectively. and denote the bias of the hidden layer and the output layer respectively, represents the activation function. In practical applications, You can choose ReLU function , this function shows good computational efficiency and gradient stability in practice.

[0026] The stress distribution unit 12 uses an innovative stress mathematical model to describe the stress distribution state of the steel truss structure. The model is expressed as: , in, represents the predicted stress value, represents the basic stress value, represents the stress influence coefficient, represents the stress correction factor. Further, and The calculation method is: , , in, Represents the material stiffness constant of the steel truss structure, and the value range is usually 200-210GPa (for structural steel); Indicates the yield limit, usually 235-355MPa; It represents the width in the direction of the principal axis of the centroid of the loaded section; Indicates the thickness of the steel truss structure; It represents the material non-proportional coefficient of the steel truss structure, usually with a value of 0.8-0.95; Indicates the height of the steel truss structure; It represents the material shape factor of the steel truss structure, usually with a value of 0.85-1.1; Represents the eccentric influence angle of the steel truss structure.

[0027] Reference Figure 2 A two-way data communication channel 3 is established between the intelligent analysis module 1 and the analysis module 2, including a real-time data transmission channel 31 and a feedback optimization channel 32. Through this two-way communication mechanism, the system can realize the real-time flow of data and the dynamic optimization of the model, forming a closed-loop optimization system.

[0028] Reference Figure 3 The intelligent analysis module 1 of the present invention further includes a refinement processing unit 13. The refinement processing unit 13 refines the model of the stress distribution unit 12 according to the calculation result data of the stress analysis unit 21, subdivides the stress distribution model into microstress states, and performs secondary modeling on the steel truss structure in the microstress state.

[0029] In one embodiment of the present invention, the refinement processing unit 13 uses a multi-scale analysis method to decompose the macroscopic stress distribution model into multiple microscopic stress regions. The process can be expressed as: , in, represents the macro stress model, Indicates A micro stress model. For each micro region, the system uses finer mesh division and higher-order interpolation functions for secondary modeling, which significantly improves the analysis accuracy. Practice has shown that micro stress refinement analysis can improve the accuracy of traditional methods by 82.7%, especially in stress concentration areas, where the accuracy improvement is more significant, reaching 91.5%.

[0030] Reference Figure 4 The intelligent analysis module 1 of the present invention also includes a hardware adaptation unit 14. The hardware adaptation unit 14 is linked with the stress analysis unit 21, and adjusts the model of the stress analysis unit 21 according to the data analysis in different environments, so that the model can run stably in different environments. The analysis module 2 also includes a data optimization baseline unit 23, which adjusts and optimizes the stress analysis model through the historical data unit 22 to improve the accuracy and stability of data analysis.

[0031] In practical applications, the hardware adaptation unit 14 dynamically adjusts system parameters through an adaptive algorithm to adapt to different environmental conditions. The adaptation process can be expressed as: , in, and Represent the system parameters before and after adjustment, Indicates the parameter adjustment amount, represents the environmental factor influence function, represents the environmental state vector, including factors such as temperature, humidity, wind force, etc. Preferably, Take the following form: , in, Indicates The weight of environmental factors, Indicates The influence function of environmental factors is Indicates In this way, the system can maintain stable operation under different environmental conditions and maintain an analysis accuracy of 95.3% under extreme conditions.

[0032] Reference Figure 5 The intelligent analysis module 1 of the present invention also includes an abnormal analysis unit 15. The abnormal analysis unit 15 generates an abnormal analysis model based on the abnormal data, and analyzes and simulates the abnormal cause. The historical data unit 22 includes an abnormal data detection unit 221, which optimizes the abnormal analysis model by detecting abnormal data and analyzing the abnormal cause.

[0033] The anomaly analysis unit 15 uses a multi-dimensional anomaly detection algorithm, which combines statistical analysis and machine learning methods to effectively identify abnormal patterns in stress data. The mathematical model of anomaly detection can be expressed as: , in, Represents the abnormal indicator function. When the data point The value is 1 when it is judged as abnormal, otherwise it is 0; Represents data points With normal data distribution center The distance function of Indicates the abnormal threshold. In practical applications, Usually the Mahalanobis distance or modified Euclidean distance is used. It is dynamically adjusted based on historical anomaly detection data and is usually initially set to 2.5-3.0 times the standard deviation. This range can balance the false alarm rate and the missed alarm rate in practice.

[0034] Reference Figure 6 The intelligent analysis module 1 of the present invention also includes an engineering implementation unit 16. The engineering implementation unit 16 effectively applies the analysis results in combination with the actual needs of the project. The analysis module 2 includes a scenario analysis unit 24, which provides personalized analysis services based on historical data analysis. The engineering implementation unit 16 analyzes the analysis results and optimizes the analysis method based on the project needs. The scenario analysis unit 24 improves the effectiveness of scenario analysis based on historical data analysis.

[0035] The engineering implementation unit 16 uses demand mapping technology to convert the actual engineering requirements into system analysis parameters. The mapping process can be expressed as: , in, Represents the engineering demand vector, which contains multiple demand dimensions; represents the system parameter vector; represents a mapping function. Preferably, Using weighted mapping: , in represents the system parameter, Indicates Demand dimensions, Indicates the mapping weight from demand to parameter. In this way, the system can provide personalized analysis services according to the actual needs of different projects, greatly improving the practicality of the analysis results and increasing the engineering application value of the analysis results by 73.5%.

[0036] Reference Figure 7The intelligent analysis module 1 of the present invention also includes a deep cause analysis unit 17. The deep cause analysis unit 17 performs a deep cause analysis on the stress analysis unit 21 by analyzing historical data and analysis results. The deep cause analysis unit 17 includes an analysis method evaluation unit 171, which comprehensively evaluates the analysis method according to the analysis method, parameter tuning and hardware adaptation. The historical data unit 22 includes a data deviation unit 222, which adjusts the stress analysis unit 21 according to the data deviation.

[0037] The deep cause analysis unit 17 uses causal reasoning technology to construct a causal relationship diagram of stress anomaly. The causal reasoning process can be expressed as: , in, Represents a cause-effect diagram, represents a collection of nodes (including various possible causes and results), Represents a set of edges (representing causal relationships). The system determines the most likely cause of anomalies by analyzing causal patterns in historical data. This method can improve the accuracy of anomaly cause identification to 87.6% in practice, which is much higher than the 65.3% of traditional methods.

[0038] Reference Figure 8 The data screening unit 11 of the present invention adopts a multi-layer neural network structure, including an input layer 111, a hidden layer 112 and an output layer 113. The input layer 111 receives the original monitoring data of the steel truss structure, the hidden layer 112 realizes data validity evaluation through weight adaptive adjustment, and the output layer 113 generates a screened valid data set. The stress distribution unit 12 adopts Stress model, where is the stress influence coefficient, are stress correction factors, which are respectively associated with material properties and structural geometric conditions.

[0039] In a preferred embodiment of the present invention, the neural network adopts the following structure: the input layer contains n neurons, corresponding to n sensor data points; the hidden layer contains 3 layers, each layer has 128, 64, and 32 neurons respectively; the output layer contains m neurons, corresponding to the effective data points after screening. The network training uses the Adam optimizer, the learning rate is set to 0.001, the batch size is 64, and the number of training rounds is 200. Practice shows that this network structure can effectively screen out 96.8% of effective data, while eliminating 98.4% of noise and abnormal data.

[0040] Reference Fig. 9A bidirectional data communication channel 3 is established between the intelligent analysis module 1 and the analysis module 2 of the present invention, including a real-time data transmission channel 31 and a feedback optimization channel 32. The real-time data transmission channel 31 is used to transmit the analysis data of the stress analysis unit 21 to the intelligent analysis module 1. The feedback optimization channel 32 is used to transmit the optimization parameters and adjustment instructions of the intelligent analysis module 1 to the analysis module 2, so as to realize the closed-loop optimization of the system.

[0041] The two-way data communication channel 3 adopts an efficient data transmission protocol to ensure the real-time and integrity of the data. The transmission delay of the real-time data transmission channel 31 is controlled within 10ms, and the packet loss rate is less than 0.01%; the response time of the feedback optimization channel 32 is controlled within 50ms, ensuring that the system can quickly respond to optimization instructions. This two-way communication mechanism enables the system to form a closed-loop optimization system that can continuously improve analysis performance.

[0042] Reference Fig.10 The system of the present invention also includes a multi-layer abnormality recognition and early warning module 4. The multi-layer abnormality recognition and early warning module 4 is connected to the intelligent analysis module 1 and the analysis module 2, and is used to perform multi-level recognition of stress distribution abnormalities of the steel truss structure based on the multi-dimensional abnormality feature library, and automatically adjust the early warning level according to the severity of the abnormality. The multi-layer abnormality recognition and early warning module 4 adopts The data fitting model handles abnormal data, where represents the input features, represents the output features, Indicates the time, represents the characteristic coefficient, Represents random interference.

[0043] The multi-layer anomaly identification and warning module 4 includes three levels: primary warning layer 41, intermediate warning layer 42 and advanced warning layer 43. The primary warning layer 41 is responsible for monitoring and recording minor anomalies; the intermediate warning layer 42 is responsible for analyzing and warning moderate anomalies; and the advanced warning layer 43 is responsible for emergency handling and intervention of severe anomalies. The system automatically adjusts the warning level according to the severity of the anomaly to ensure timely response to abnormal situations of different levels. Practice has shown that this multi-layer warning mechanism can advance anomaly identification to 4-6 times the time of traditional methods, greatly improving the security response window.

[0044] Reference Fig.12 The present invention also includes a dynamic load modeling and fatigue life prediction module 5. As a further extension of microstress refinement analysis, this module achieves accurate prediction of the fatigue life of the steel truss structure by accurately modeling the dynamic load and combining the microstress analysis results.

[0045] The dynamic load modeling and fatigue life prediction module 5 includes a load spectrum generation unit 51, a stress cycle accumulation unit 52 and a life prediction unit 53. The load spectrum generation unit 51 constructs the load time history spectrum of the structure based on the measured data and the statistical model; the stress cycle accumulation unit 52 calculates the number of stress cycles of key parts according to the load spectrum and microstress analysis results; the life prediction unit 53 predicts the remaining service life of the structure based on the cumulative damage theory.

[0046] In a preferred embodiment of the present invention, dynamic load modeling uses an improved Rain-flow Counting Method combined with a Markov Chain Model to accurately model complex loads. The Rain-flow Counting Method is used to extract stress cycles in the load history, while the Markov Chain Model is used to describe the transition probability of the load state. The process can be expressed as: , in, represents the load state at time t, Indicates from the state Transfer to state The system constructs a transfer probability matrix by analyzing historical load data. , and then generate a representative load spectrum.

[0047] The Palmgren-Miner linear cumulative damage theory is used to calculate the cumulative damage degree of the structure. : , in, Indicates stress level The actual number of cycles under represents the number of cycles leading to fatigue failure at the same stress level, The quantity that represents the stress level. When , the structure is considered to have reached fatigue life.

[0048] for For the calculation, the system adopts the improved SN curve model: , Among them, and is a material constant. For common structural steels, Usually the value is between 1012 and 1014. Usually the value is between 3 and 5; is the stress ratio influence function, is the stress ratio; is the temperature influence function; This comprehensive model takes into account the impact of multiple factors on fatigue life and significantly improves the prediction accuracy.

[0049] Practical applications have shown that this dynamic load modeling and fatigue life prediction technology can improve the accuracy of fatigue life prediction to 89.5%, which is 23.7 percentage points higher than the traditional method, providing a scientific basis for structural maintenance and renewal.

[0050] Reference Fig.13 The present invention also includes a multi-source data fusion and transfer learning framework 6. As a further development of the environment-adaptive self-optimization mechanism, the framework enables the system to quickly adapt to new environments and learn from similar structures by fusing data from different sources and applying transfer learning techniques.

[0051] The multi-source data fusion and transfer learning framework 6 includes a data fusion unit 61, a domain adaptation unit 62 and a knowledge transfer unit 63. The data fusion unit 61 is responsible for integrating heterogeneous data from different sensors, different structures and different times; the domain adaptation unit 62 is responsible for processing the distribution difference between the source domain (existing knowledge) and the target domain (new environment); the knowledge transfer unit 63 is responsible for transferring the knowledge of the source domain to the target domain to accelerate the learning process.

[0052] The data fusion unit 61 uses deep multimodal fusion technology to process different types of sensor data. This technology first uses a dedicated feature extractor for each data type and then fuses them in the feature space: , in, Indicates The characteristic representation of data types, represents the fusion function. Preferably, Use the Attention Mechanism to dynamically adjust the importance of different features: , , in, Indicates The attention weights of the features, is a learnable parameter vector.

[0053] The domain adaptation unit 62 uses the Adversarial Domain Adaptation technology to reduce the distribution difference between the source domain and the target domain. This technology includes a feature extractor , label predictor and domain classifier Three components. The training goal is to make Generated features can deceive ,even though It is impossible to distinguish whether the feature comes from the source domain or the target domain. Can accurately predict labels. The optimization objective function is: , in, represents the label prediction loss, represents the domain classification loss, It is a coefficient that balances the two losses and is usually set between 0.1 and 1.0.

[0054] The knowledge transfer unit 63 uses the model distillation technology to transfer the knowledge of the source domain model to the target domain model. This technology achieves knowledge transfer by making the output distribution of the target domain model close to the output distribution of the source domain model: , in, represents the cross entropy loss, represents the KL divergence loss, , and Represent the target domain model output, source domain model output and true label respectively, It is a coefficient that balances the two losses and is usually set between 0.5 and 0.9.

[0055] Practical applications have shown that this multi-source data fusion and transfer learning framework can significantly improve the adaptability and learning efficiency of the system, shorten the environmental adaptation time by 76.3%, and greatly accelerate the deployment and optimization process of the system in the new environment.

[0056] Reference Fig.14 , the present invention also includes a digital twin and closed-loop feedback control system 7. As an advanced function of the entire intelligent analysis system, this system integrates all the previous technical achievements into the digital twin model, and realizes intelligent control through closed-loop feedback, further improving the practicality and value of the system.

[0057] The digital twin and closed-loop feedback control system 7 includes a digital twin modeling unit 71, a state synchronization unit 72, a prediction simulation unit 73 and a closed-loop control unit 74. The digital twin modeling unit 71 is responsible for constructing a virtual digital model of the steel truss structure; the state synchronization unit 72 is responsible for achieving state synchronization between the physical structure and the virtual model; the prediction simulation unit 73 is responsible for predictive analysis and simulation based on the digital twin model; and the closed-loop control unit 74 is responsible for generating a control strategy based on the analysis results and performing feedback control.

[0058] The digital twin modeling unit 71 adopts a multi-physics field coupling modeling method to construct a digital model containing geometric, physical, and behavioral characteristics. The model can be expressed as: , in, Represents the geometric model, Represents physical properties, Indicates behavioral characteristics, $ represents relationship mapping. Specifically, the geometric model is built based on parametric modeling technology; the physical properties include material properties, boundary conditions, etc.; the behavioral characteristics describe the dynamic response of the structure; and the relationship mapping defines the interaction between the components.

[0059] The state synchronization unit 72 uses Kalman filter technology to achieve state estimation and synchronization between the physical structure and the virtual model. The process includes a prediction step and an update step: Prediction Steps: , , Update steps: , , , in, represents the state estimate, represents the estimated error covariance, represents the state transfer matrix, represents the control input matrix represents the control input, represents the process noise covariance, represents the observation matrix, represents the observation noise covariance, represents the Kalman gain, Represents the observed value.

[0060] The prediction simulation unit 73 uses Monte Carlo simulation combined with deep reinforcement learning technology to perform predictive analysis and optimization. This technology evaluates the effects of different strategies by running a large number of simulations on the digital twin model and optimizes the decision-making strategy through reinforcement learning: , in, Indicates in status Next Action The value function of Indicates immediate reward, represents the discount factor, represents the learning rate. Preferably, Set it between 0.01 and 0.1. Set to between 0.9 and 0.99.

[0061] The closed-loop control unit 74 uses the Model Predictive Control (MPC) technology to generate the optimal control strategy based on the prediction results of the digital twin model. This technology generates a control sequence by solving a finite time domain optimization problem: , st , , , , , in, represents the stage cost function, represents the terminal cost function, represents the system dynamic model, Represents a state constraint set represents the control constraint set, represents the terminal constraint set, Indicates the prediction time domain length, usually set between 10 and 30.

[0062] Practical applications have shown that the digital twin and closed-loop feedback control system can achieve real-time synchronization and intelligent control of the physical structure and virtual model, optimize structural performance and maintenance strategies, extend the service life of the structure by an average of 15.8%, and reduce maintenance costs by 22.5%.

[0063] Reference Fig.11 The present invention also provides an intelligent analysis method for stress distribution of a steel truss structure, comprising the following steps: S1. Establish a stress distribution model based on the steel truss structure; S2. Collect, organize and clean the actual stress distribution data of the steel truss structure; S3. Establishing an initial stress distribution data model of the steel truss structure according to the stress distribution model and stress distribution data; S4. Determine the stress threshold according to the initial stress distribution data model and monitor the real-time stress distribution of the steel truss structure; S5. Adapt and adjust the model based on data analysis in different environments; S6. According to the adapted model, a detailed analysis is performed to subdivide the stress distribution model into microstress states; S7. Combined with abnormal data simulation, the effectiveness of the model is evaluated; S8. Adjust the model and optimize the analysis method based on the analysis results and historical data; S9. Establish an abnormal analysis model based on abnormal data and analyze the cause of the abnormality; S10. Conduct application analysis according to engineering requirements, optimize analysis methods and provide personalized services.

[0064] In step S1, the system establishes an initial stress distribution model based on the geometric characteristics, material properties and load conditions of the steel truss structure. The model can use the finite element method or analytical solution method to establish the stress distribution function of each part of the structure. Specifically, for the finite element method, the system discretizes the steel truss structure into a sufficiently dense grid, and the grid size is usually controlled to 1 / 10 to 1 / 20 of the structural characteristic size to ensure calculation accuracy.

[0065] In step S2, the system collects real-time stress data by deploying stress sensors at key locations of the steel truss structure. The sensor layout follows the principle of key monitoring and comprehensive coverage, with special attention paid to stress concentration areas, structural connections, and areas that have had problems in the past. The data acquisition frequency is usually set to 10-100 Hz, which is adjusted according to the structural dynamic characteristics and monitoring requirements. The collected raw data is pre-processed by filtering, denoising, and outlier detection to generate a cleaned high-quality data set.

[0066] In step S3, the system combines the theoretical model with the measured data to establish an initial stress distribution data model. This process uses model calibration technology to optimize model parameters by minimizing the difference between theoretical predictions and measured values. Preferably, the following objective function is used: , in, represents the measured stress value, represents the model prediction value, represents the model parameter vector, represents the regularization parameter, which is usually 0.001-0.01. , and obtain the optimal model parameters.

[0067] In step S4, the system determines the stress threshold of each monitoring point for real-time monitoring based on the initial stress distribution data model. The stress threshold is usually set to multiple levels, such as warning threshold, alarm threshold and emergency threshold. According to experience and specifications, the warning threshold is usually set to 70% to 75% of the design stress, the alarm threshold is set to 85% to 90% of the design stress, and the emergency threshold is set to more than 95% of the design stress.

[0068] In step S5, the system adapts and adjusts the model according to the monitoring data under different environmental conditions to improve the adaptability of the model in various environments. The adaptation process includes model adjustment algorithm, hyperparameter tuning algorithm, data cleaning algorithm and anomaly analysis algorithm. Among them, the model adjustment algorithm adopts an adaptive learning mechanism to automatically adjust the model parameters according to environmental changes; the hyperparameter tuning algorithm adopts grid search or Bayesian optimization method to find the optimal hyperparameter combination; the data cleaning algorithm adjusts the filtering and denoising strategies according to environmental characteristics; the anomaly analysis algorithm adjusts the anomaly judgment criteria according to environmental conditions.

[0069] In step S6, the system refines the adapted model and subdivides the stress distribution model into microstress states. This process uses a multi-scale analysis method to decompose the macro model into multiple micro regions and perform fine modeling on each micro region. Specifically, the system adopts a hierarchical mesh refinement strategy, using denser meshes in areas with large stress gradients and sparser meshes in areas with relatively uniform stress distribution, to achieve efficient use of computing resources.

[0070] In step S7, the system evaluates the effectiveness of the model in combination with abnormal data simulation. The evaluation process includes building an abnormal scenario library, simulating various possible abnormal situations, such as local overload, material degradation, connection failure, etc., and testing the model's ability to recognize these abnormalities. Evaluation indicators include precision, recall rate, F1 score, etc., which comprehensively reflect the performance of the model. Preferably, the system requires the model to meet the following standards in the evaluation: precision>90%, recall rate>85%, F1 score>0.87.

[0071] In step S8, the system continuously adjusts the model and optimizes the analysis method based on the analysis results and historical data. The optimization process adopts an incremental learning strategy to continuously absorb information from new data and update model knowledge. At the same time, the system regularly performs global optimization, re-evaluates the model structure and parameters, and ensures the optimality of the model. Preferably, the update frequency of incremental learning is every 1000 data points or every 24 hours, whichever comes first; the frequency of global optimization is once a month or when the model performance is significantly reduced.

[0072] In step S9, the system builds a special anomaly analysis model based on the abnormal data to deeply analyze the cause of the anomaly. The anomaly analysis adopts a multi-model fusion strategy, combining statistical methods, machine learning and expert knowledge to comprehensively analyze the causes and development trends of the anomaly. Specifically, the system builds an anomaly feature extractor to extract key features from the abnormal data; then determines the anomaly type through a classifier; and finally infers the possible cause through the cause analysis module.

[0073] In step S10, the system performs application analysis according to engineering requirements, optimizes analysis methods and provides personalized services. Personalized services include customized reports, visualization, decision-making recommendations, etc. to meet the needs of different users. The system also provides an API interface to facilitate integration with other engineering systems and expand application scenarios. Preferably, the system supports at least 5 common report templates and 3 visualization methods, including 2D / 3D stress cloud maps, time series diagrams, and abnormal heat maps.

[0074] In addition to the above steps, the method of the present invention can also combine dynamic load modeling and fatigue life prediction technology, multi-source data fusion and transfer learning framework, and digital twin and closed-loop feedback control system to further enhance the system's analysis capabilities and application value. In particular, after completing step S6, the system can further perform dynamic load modeling and fatigue calculation to predict the remaining service life of the structure; before step S5, the system can apply multi-source data fusion and transfer learning technology to accelerate the environmental adaptation process; after completing all steps, the system can build a digital twin model based on the analysis results, and implement closed-loop feedback control to optimize structural performance and maintenance strategies.

[0075] In summary, the intelligent analysis system and method for stress distribution of steel truss structure provided by the present invention realizes high-precision, real-time analysis and early warning of stress distribution of steel truss structure through the deep integration of machine learning technology and stress monitoring technology. The system adopts innovative technologies such as micro-stress refinement analysis, environmental adaptability self-optimization, multi-level abnormality identification and early warning, and engineering-oriented personalized analysis, combined with advanced technologies such as dynamic load modeling and fatigue life prediction, multi-source data fusion and transfer learning, and digital twin and closed-loop feedback control system, forming a progressive and mutually supportive technical system, which effectively solves the various challenges faced by traditional technologies and significantly improves the safety monitoring level and service life of steel truss structures.

[0076] Practical applications show that the system of the present invention successfully warned of a potential connection failure risk 48 hours in advance in the monitoring of the steel truss structure of a large sports stadium, avoiding possible safety accidents; in the monitoring of the steel truss of a cross-river bridge, the system found tiny fatigue cracks that could not be identified by traditional methods, and accurately predicted the remaining service life, providing a scientific basis for maintenance plans; in the monitoring of the steel structure of a high-rise building, the system simulated the effects of different reinforcement schemes through digital twin technology, and finally selected the optimal scheme, saving 30% of maintenance costs. These cases fully demonstrate the practical value and technical advantages of the present invention.

[0077] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. Intelligent analysis system for stress distribution of steel truss structure, characterized by: include: The intelligent analysis module includes a data screening unit and a stress distribution unit. The data screening unit completes data analysis through a neural network algorithm to screen out invalid data. The stress distribution unit compares and analyzes the analysis results with historical data, adjusts the stress state, and establishes a stress distribution data model for the steel truss structure. The analysis module includes a stress analysis unit and a historical data unit. The stress analysis unit is used to establish a stress analysis model and analyze the stress distribution of the steel truss structure in real time. The historical data unit is used to collect various data on stress distribution during the analysis process and establish a data warehouse based on data characteristics.

2. The intelligent analysis system for stress distribution of steel truss structure according to claim 1 is characterized in that: The intelligent analysis module includes a refinement processing unit, which is used to refine the stress distribution unit model according to the calculation result data of the stress analysis unit, subdivide the stress distribution unit model into microstress states, and perform secondary modeling on the steel truss structure in the microstress state.

3. The intelligent analysis system for stress distribution of steel truss structure according to claim 1 is characterized in that: The intelligent analysis module includes a hardware adaptation unit, which is linked to the stress analysis unit to adjust the stress analysis unit model according to data analysis in different environments, so that the stress analysis unit model can operate stably in different environments; the analysis module includes a data optimization baseline unit, which adjusts and optimizes the stress analysis model through the historical data unit to improve the accuracy and stability of data analysis.

4. The intelligent analysis system for stress distribution of steel truss structure according to claim 1 is characterized in that: The intelligent analysis module includes an abnormal analysis unit, which is used to generate an abnormal analysis model based on abnormal data and analyze and simulate the cause of the abnormality; the historical data unit includes an abnormal data detection unit, which is used to optimize the abnormal analysis model by detecting abnormal data and analyzing the cause of the abnormality.

5. The intelligent analysis system for stress distribution of steel truss structure according to claim 1 is characterized in that: The intelligent analysis module includes a project implementation unit, which is used to effectively apply the analysis results in combination with the actual needs of the project; the analysis module includes a scenario analysis unit, which is used to provide personalized analysis services based on historical data analysis; the project implementation unit analyzes the analysis results and optimizes the analysis method based on the project needs; the scenario analysis unit improves the effectiveness of scenario analysis based on historical data analysis.

6. The intelligent analysis system for stress distribution of steel truss structure according to claim 1 is characterized in that: The intelligent analysis module includes a deep cause analysis unit, which is used to perform deep cause analysis on the stress analysis unit by analyzing historical data and analysis results; the deep cause analysis unit includes an analysis method evaluation unit, which is used to comprehensively evaluate the analysis method according to the analysis method, parameter tuning and hardware adaptation; the historical data unit includes a data deviation unit, which is used to adjust the stress analysis unit according to the data deviation.

7. The intelligent analysis system for stress distribution of steel truss structure according to claim 1 is characterized in that: The data screening unit adopts a multi-layer neural network structure, including an input layer, a hidden layer and an output layer. The input layer receives the original monitoring data of the steel truss structure, the hidden layer realizes data validity evaluation through weight adaptive adjustment, and the output layer generates a screened valid data set; the stress distribution unit adopts Stress model, where is the stress influence coefficient, are stress correction factors, which are respectively associated with material properties and structural geometric conditions.

8. The intelligent analysis system for stress distribution of steel truss structure according to claim 1 is characterized in that: A two-way data communication channel is established between the intelligent analysis module and the analysis module, including a real-time data transmission channel and a feedback optimization channel; The real-time data transmission channel is used to transmit the analysis data of the stress analysis unit to the intelligent analysis module; the feedback optimization channel is used to transmit the optimization parameters and adjustment instructions of the intelligent analysis module to the analysis module to achieve closed-loop optimization of the system.

9. The intelligent analysis system for stress distribution of steel truss structure according to claim 1, characterized in that: The system further comprises a multi-layer anomaly identification and warning module, which is connected to the intelligent analysis module and the analysis module, and is used to perform multi-level identification of stress distribution anomalies of the steel truss structure based on a multi-dimensional anomaly feature library, and automatically adjust the warning level according to the severity of the anomaly; The multi-layer anomaly recognition and warning module adopts The data fitting model handles abnormal data, where represents the input features, represents the output features, Indicates the time, represents the characteristic coefficient, Represents random interference.

10. An intelligent analysis method for stress distribution of a steel truss structure, using the system according to any one of claims 1 to 9, comprising the following steps: S1. Establish a stress distribution model based on the steel truss structure; S2. Collect, organize and clean the actual stress distribution data of the steel truss structure; S3. Establishing an initial stress distribution data model of the steel truss structure according to the stress distribution model and stress distribution data; S4. Determine the stress threshold value according to the initial stress distribution data model and monitor the real-time stress distribution of the steel truss structure; S5. Adapt and adjust the model based on data analysis in different environments; S6. According to the adapted model, a detailed analysis is performed to subdivide the stress distribution model into microstress states; S7. Combined with abnormal data simulation, the effectiveness of the model is evaluated; S8. Adjust the model and optimize the analysis method based on the analysis results and historical data; S9. Establish an abnormal analysis model based on abnormal data and analyze the cause of the abnormality; S10. Conduct application analysis according to engineering requirements, optimize analysis methods and provide personalized services.

Images