A method for early warning of building structure vibration caused by rail transit operation

Abstract

The application provides a kind of track traffic operation causes building structure vibration to exceed standard early warning calculation method, it is related to track traffic environmental vibration prediction technical field, including the following steps: collecting data;Processing data and extracting its features;Establish vibration exceeding standard evaluation model;Start early warning mechanism and according to early warning level, propose corresponding control measures;Optimize model to improve early warning accuracy and response speed.The application combines real-time monitoring data, historical vibration data and machine learning algorithm, to realize the real-time vibration state evaluation and exceeding standard risk early warning of building structure, the purpose is to effectively avoid the vibration damage of building structure caused by track traffic operation, and protect the vibration comfort of residents.This method has wide application range and can be applied to various building structures with potential vibration risks caused by track traffic.

Description

Technical Field

[0001] This invention belongs to the field of rail transit environmental vibration prediction technology, specifically relating to a calculation method for early warning of excessive vibration of building structures caused by rail transit operation. Background Technology

[0002] With the growth of urban populations and the continuous increase in transportation demand, rail transit plays an increasingly important role in modern society. The operation of rail transit systems such as subways, light rail, and high-speed rail not only significantly reduces air pollution and greenhouse gas emissions but also effectively alleviates road traffic pressure, making commuting between cities more convenient. However, this also brings with it the problem of building structural vibration caused by rail transit.

[0003] Vibrations caused by rail transit propagate outwards in the form of waves. After energy dissipation due to the coupling between the soil and the building foundation, the vibrations further propagate into the building structure. Excessive vibrations in building structures can lead to a series of adverse effects, such as harm to human health.

[0004] In existing technologies, some scholars have proposed simplified prediction models for the vibration impact of urban rail transit lines and provided prediction calculation formulas. In addition, other scholars have summarized targeted solutions based on the vibration sources present in urban rail transit. However, these vibration exceedance early warning methods all have limitations, insufficient applicability, and fail to effectively integrate real-time on-site monitoring data with historical data. Summary of the Invention

[0005] This invention provides a method for calculating early warning of excessive vibration in building structures caused by rail transit operation. This method combines real-time monitoring data, historical vibration data, and machine learning algorithms to achieve real-time vibration status assessment and early warning of excessive vibration risks in building structures. The aim is to effectively avoid the vibration impact on building structures caused by rail transit operation and ensure residents' vibration comfort.

[0006] The objective of this invention is achieved through the following technical solution.

[0007] S1: Data collection: Real-time vibration data is collected by setting up a multi-point vibration sensor system at key locations of the building structure and along the rail transit line, and the data of special events and historical vibration data during rail transit operation are integrated.

[0008] S2: Process the data and extract features. Preprocess the collected vibration data, including denoising and normalization, and use time series analysis and spectrum analysis methods to extract vibration features, including frequency, amplitude and duration.

[0009] S3: Establish a vibration over-standard evaluation model, based on the extracted vibration data features, combined with the characteristics of the building structure, integrate vibration features, including frequency, amplitude, duration, power spectral density, establish evaluation models based on support vector machine and neural network, output over-standard risk evaluation results;

[0010] S4: Start the early warning mechanism, when the risk evaluation result output by the vibration over-standard evaluation model exceeds the early warning threshold, start the corresponding control measures according to the early warning level, including adjusting the rail transit running speed, strengthening the structure reinforcement;

[0011] S5: Optimize the model, improve the early warning accuracy and response speed through retraining, integrate practical feedback and cyclic iteration.

[0012] In step S1 of the present application, the collected data includes real-time vibration data and historical vibration data, and the collection process includes the following steps:

[0013] Step S11: Select and deploy vibration data collection equipment vibration sensor, the vibration sensor should have enough measurement range and resolution to ensure the accuracy and reliability of the data; high sensitivity accelerometer and displacement sensor should be selected as the vibration sensor to accurately capture the vibration of the building structure in the rail transit operation;

[0014] The vibration sensor should be deployed in the key load-bearing parts of the building structure (such as beams, columns, foundations, etc.), as well as the areas where the vibration impact is expected to be the greatest; at the same time, vibration sensors should be deployed on the ground and tunnel walls along the rail transit line to monitor the rail transit operation status;

[0015] Step S12: Build a real-time data acquisition system; use high-performance data collectors to ensure that the data of multiple vibration sensors can be collected synchronously and accurately; use modern wireless communication technologies (such as Wi-Fi, Bluetooth, 5G, etc.) to realize real-time remote transmission of vibration data, reduce interference and cost caused by wiring; build a cloud or local data center, use an efficient database management system to store a large amount of real-time monitoring data, support high-concurrency data access and query;

[0016] Step S13: Integrate historical data and real-time data; collect historical vibration data including past monitoring data, rail transit operation logs, records of past vibration over-standard events, etc.; develop data integration tools and algorithms to integrate real-time monitoring data with rich historical data, providing comprehensive data support for subsequent feature extraction and model training;

[0017] Step S14: control data quality; pre-processing data includes data cleaning (remove outliers and noise), data normalization and other steps to ensure the quality of data meets the analysis requirements; regularly check and maintain the data acquisition and storage system, and set up a data review mechanism to ensure the accuracy and integrity of the data;

[0018] In step S2 of the present application, a method for data processing and feature extraction is provided, which specifically includes the following steps:

[0019] Step S21: data preprocessing; remove outliers caused by sensor failure, environmental interference and other reasons to improve data quality; synchronize data through timestamp to ensure that all data are analyzed under the same time frame; use filtering techniques (such as low-pass filter, median filter, etc.) to remove noise in the data and improve the signal-to-noise ratio; scale the data according to the proportion so that it falls within a small specific interval, such as [0, 1], which helps subsequent algorithm processing and comparison;

[0020] Step S22: extract data features; analyze data time domain including peak value, mean and standard value, variance, skewness and kurtosis; perform frequency domain analysis through Fourier transform and power spectral density; analyze waveforms such as wavelet transform;

[0021] Step S23: select data features; use feature selection techniques to filter out the most useful features for establishing a vibration over-limit early warning model to reduce model complexity and improve early warning accuracy;

[0022] Step S24: build data set; build the data after preprocessing and feature extraction into a data set suitable for machine learning training, which usually includes a feature set and a label (such as whether it is over-limit).

[0023] In step S3 of the present application, a method for establishing a vibration over-limit evaluation model is provided, which specifically includes the following steps:

[0024] Step S31: model establishment basis; evaluation standard formulation, according to the building codes, vibration safety standards of the country or region, and the characteristics of the building structure (such as material, purpose, structure type, etc.), formulate vibration safety threshold and evaluation standard; integrate vibration features, including frequency, amplitude, duration, power spectral density, etc.;

[0025] Step S32: select and train model; divide data, divide the data set into training set, validation set, and test set to evaluate model performance; train and optimize model, use training set data to train model and optimize model parameters through validation set to obtain the best prediction performance;

[0026] Step S33: evaluation and verification of the model; the performance of the model is evaluated by multiple indicators such as accuracy, precision, recall, F1 score, etc. to ensure the accuracy and reliability of the model in predicting building structure vibration exceeding the standard; the stability and generalization ability of the model are verified by methods such as K-fold cross-validation to avoid overfitting of the model;

[0027] Step S34: deployment of the model and continuous learning; implementation of deployment, deployment of the trained model to the actual vibration monitoring system to realize real-time vibration exceeding the standard early warning; continuous learning, through regular use of newly collected data to update and optimize the model, maintain the adaptability and accuracy of the model to cope with possible environmental and condition changes.

[0028] Step S4 of the present application provides a method for starting an early warning mechanism and proposing corresponding control measures according to the warning level, which specifically includes the following steps:

[0029] Step S41: design early warning strategy; immediate warning, when the vibration is about to exceed the predetermined safety threshold, the system immediately generates a warning signal, and notifies the relevant managers and users through email, SMS or application; classified warning, according to the severity of vibration exceeding the standard, design a multi-level warning system (such as yellow, orange, red warning), the higher the level of warning, the closer the vibration to or exceeds the safety threshold;

[0030] Step S42: real-time monitoring and data updating; deploy real-time data monitoring system, continuously collect building structure vibration data and track operation status; data updating, the warning and control decision module needs to be able to receive the latest monitoring data and perform immediate data updating and re-evaluation on the model to ensure the accuracy and timeliness of the warning information;

[0031] Step S43: execute control decision; automatic control, for some cases that can reduce the impact of vibration through technical means, such as adjusting the operation mode and speed of rail transit, the system can automatically issue control instructions; manual intervention, for cases that require professional assessment, the system will provide detailed warning information and analysis report to support further intervention decision-making by human;

[0032] Step S44: develop emergency plan; establish plan library, according to different levels of vibration exceeding the standard risk and possible impact, develop a series of emergency plans, including emergency evacuation, temporary reinforcement of structure, temporary interruption of rail transit operation, etc.; implement the plan, once the warning system triggers a specific level of warning, the corresponding level of emergency plan will be started until the vibration risk is effectively controlled;

[0033] Step S45: notification and feedback; timely notification to ensure that all relevant parties, including building managers, residents, rail transit management agencies, etc., can timely receive early warning information and the execution of emergency plans; feedback collection, establish a feedback collection mechanism to allow all relevant parties to feedback the effect of early warning response for subsequent analysis and model optimization.

[0034] Step S5 of the present application provides a method for optimizing the model to improve the accuracy of early warning and response speed, specifically comprising the following steps:

[0035] Step S51: continuously collect and analyze data; accumulate data, continuously collect new vibration data and related rail transit operation data, enrich the data set; performance analysis, periodically analyze the prediction performance of the current model on the newly collected data, and identify the deficiencies or biases of the model in prediction performance;

[0036] Step S52: update and optimize features; re-evaluate features, re-evaluate the contribution of each feature to the prediction performance of the model based on the results of new data analysis, eliminate ineffective or less influential features, and introduce new beneficial features; feature engineering, optimize the feature set through feature engineering methods (such as feature construction, feature selection, feature conversion, etc.) to improve the prediction accuracy and generalization ability of the model;

[0037] Step S53: retrain and fine-tune the model; train the model, use the updated data set and feature set to retrain and fine-tune the model, and find the optimal model parameters and structure; cross-validation, comprehensively evaluate the performance of the optimized model to ensure that the model has good prediction effect on different data subsets;

[0038] Step S54: integrate practical feedback; collect feedback, build a feedback collection mechanism to obtain various feedback information obtained from early warning practice, including the accuracy of early warning, the execution effect of emergency plans, etc.; analyze feedback, identify problems and potential improvement space of the early warning system in actual application;

[0039] Step S55: update the knowledge base and decision logic; update the knowledge base, integrate practical feedback and the latest research achievements into the knowledge base, update the judgment logic and early warning strategy of vibration exceeding; adjust the decision logic, adjust the decision logic in the control decision module according to the new knowledge base content, to ensure the accuracy and adaptability of early warning and control decision;

[0040] Step S56: cycle iteration and continuous optimization; iteration, form a continuous optimization iteration process with the above steps to ensure that the model and system can continuously adapt to changes in external environment and internal operation; performance monitoring, set up indicators to monitor the performance of the model and system, timely discover new optimization needs or potential problems, and maintain the highest operating efficiency of the system.

[0041] From the above technical solution, the advantages and positive effects of the present application are:

[0042] The present application provides a kind of building structure vibration caused by rail transit operation exceeds standard early warning calculation method.The innovative point of this calculation method is to combine real-time monitoring data and historical data with machine learning for in-depth analysis, and a comprehensive evaluation and early warning system is proposed, which provides an advanced and practical solution for the problem of building structure vibration exceeding standard caused by rail transit. BRIEF DESCRIPTION OF DRAWINGS

[0043] Figure 1 The step block diagram of the building structure vibration caused by rail transit operation exceeds standard early warning calculation method provided by the embodiment of the present application is shown;

[0044] Figure 2 The flowchart of the specific implementation method of step S1 data collection module provided by the embodiment of the present application is shown;

[0045] Figure 3 The flowchart of the specific implementation method of step S2 data processing and feature extraction provided by the embodiment of the present application is shown;

[0046] Figure 4 The flowchart of the specific implementation method of step S3 establishing vibration exceeding standard evaluation model provided by the embodiment of the present application is shown;

[0047] Figure 5 The flowchart of the specific implementation method of step S4 starting early warning mechanism and according to early warning level, proposing corresponding control measures provided by the embodiment of the present application is shown;

[0048] Figure 6 The flowchart of the specific implementation method of step S5 optimization model provided by the embodiment of the present application is shown. DETAILED DESCRIPTION

[0049] In order to make the invention purpose, features and advantages of the present application more obvious and easy to understand, the technical solutions in the embodiments of the present application will be described clearly and completely in conjunction with the drawings in the embodiments of the present application.

[0050] Obviously, the embodiments described below are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor belong to the scope of protection of the present application.

[0051] As Figure 1As shown, the present application provides a track transportation operation induced building structure vibration exceeding standard early warning calculation method, comprising the following steps:

[0052] Step S1: Collect data, set up a multi-point vibration sensor system at key positions of the building structure and along the track transportation operation line, collect real-time vibration data, and integrate special event data and historical vibration data during track transportation operation;

[0053] Step S2: Process data and extract features, pre-process the collected vibration data, including denoising, normalization, and use time series analysis and frequency spectrum analysis methods to extract vibration features, including frequency, amplitude, and duration;

[0054] Step S3: Establish a vibration exceeding standard evaluation model, based on the extracted vibration data features, integrate vibration features including frequency, amplitude, duration, and power spectral density, establish an evaluation model based on support vector machine and neural network, and output the exceeding standard risk evaluation result;

[0055] Step S4: Start the early warning mechanism, when the risk evaluation result output by the vibration exceeding standard evaluation model exceeds the early warning threshold, start the corresponding control measures according to the early warning level, including adjusting the track transportation operation speed and strengthening the structure reinforcement;

[0056] Step S5: Optimize the model, improve the early warning accuracy and response speed through retraining, integrate practical feedback and cyclic iteration.

[0057] According to the above embodiment, preferably, the data collected in step S1 includes real-time vibration data and historical vibration data; a multi-point vibration sensor system needs to be set up, distributed at key positions of the building structure and along the track transportation operation line, to collect vibration data in real time; historical vibration data is collected, including special event data (such as high-speed passing, emergency braking, etc.) during track transportation operation.

[0058] Figure 2 A flowchart of a specific implementation method for collecting data provided by the embodiment of the present application.

[0059] The method comprises the following steps:

[0060] Step S11, select and deploy vibration sensors for data collection equipment. High-sensitivity accelerometers and displacement sensors are selected for vibration sensors to accurately capture the vibration of the building structure during track transportation operation. In addition, the sensors should have sufficient measurement range and resolution to ensure the accuracy and reliability of the data. Vibration sensors should be deployed at key load-bearing parts of the building structure, such as beams, columns, foundations, etc., and the areas with the maximum expected vibration impact. To monitor the track transportation operation state, vibration sensors should also be deployed on the ground and track structure along the track transportation line.

[0061] Step S12, build a real-time data acquisition system. The data acquisition device is selected to be a high-performance data acquisition device to ensure that the data of multiple sensors can be synchronized and accurately collected. In terms of real-time vibration data transmission, modern wireless communication technologies (such as Wi-Fi, Bluetooth, 5G, etc.) should be used to reduce interference and costs caused by wiring. The storage and management of data should consider establishing a cloud or local data center, using an efficient database management system to store a large amount of real-time monitoring data, and supporting high-concurrency data access and query.

[0062] Step S13, integrate historical data and real-time data. Past monitoring data, rail transit operation logs, and records of past vibration exceedance events can be obtained from relevant management departments or collected through historical documents, research reports, etc. After obtaining historical data and real-time data, develop data integration tools and algorithms to fuse the two types of data, providing comprehensive data support for subsequent feature extraction and model training.

[0063] Step S14, control data quality. Preprocess the data, including data cleaning (remove outliers and noise), data normalization, etc. to ensure the quality of the data meets the analysis requirements. Regularly check and maintain the data acquisition and storage system, and establish a data review mechanism to ensure the accuracy and integrity of the data.

[0064] Through the above steps, valuable vibration information can be efficiently and accurately collected, providing a solid data foundation for the early warning of building structure vibration exceedance caused by rail transit operation.

[0065] Further, step S2 preprocesses the collected vibration data and uses time series analysis, spectral analysis, etc. to extract vibration features, including but not limited to frequency, amplitude, duration, etc.

[0066] Figure 3 A specific implementation method for processing data and extracting features provided by an embodiment of the present application is shown in the flowchart.

[0067] The method includes the following steps:

[0068] Step S21, preprocess the data. Clean the data to remove outliers caused by sensor failure, environmental interference, etc. Synchronize the data by timestamp to ensure that all data is analyzed under the same time frame. To improve the signal-to-noise ratio of the signal, filtering techniques (such as low-pass filters, median filters, etc.) should be used to remove noise in the data. Finally, scale the data according to the proportion so that it falls within a small specific interval, such as [0, 1], which helps subsequent algorithm processing and comparison.

[0069] Step S22, feature extraction. The methods used for feature extraction include time domain analysis, frequency domain analysis and waveform analysis, etc. Time domain analysis can extract features such as peak value, mean value and standard deviation, variance and skewness and kurtosis. For frequency domain analysis, Fourier transform is used to convert time series into frequency domain signals, identify the main frequency components in the vibration data, reflect the natural frequency of the structure and possible resonance problems; power spectral density (PSD) is used to provide power information of signal frequency distribution, which is used to identify the energy distribution at a specific frequency. When performing waveform analysis, wavelet transform is used for feature extraction of non-stationary signals, which can analyze vibration signals in time-frequency domain and identify possible nonlinear and transient vibration characteristics.

[0070] Step S23, feature selection. After extracting a large number of features, feature selection techniques (such as statistical-based feature selection, machine learning-based feature selection algorithms such as random forest) are used to screen out the most useful features for establishing a vibration over-standard early warning model, so as to reduce the model complexity and improve the early warning accuracy.

[0071] Step S24, constructing a data set. The pre-processed and feature-extracted data is used to construct a data set suitable for machine learning training, which usually contains a feature set and a label (such as whether it is over-standard).

[0072] Further, step S3 establishes vibration over-standard evaluation criteria according to the vibration reduction design requirements of building structures and the actual vibration bearing capacity. A vibration over-standard evaluation model based on machine learning algorithms such as support vector machine (SVM) and neural network is developed, which inputs vibration features and outputs over-standard risk evaluation results.

[0073] Figure 4 The flowchart of a specific implementation method for establishing a vibration over-standard evaluation model provided by the embodiment of the present application.

[0074] The method comprises the following steps:

[0075] Step S31, model establishment basis. According to the building specifications, vibration safety standards of the country or region, and the characteristics of the building structure (such as material, purpose, structure type, etc.), vibration safety threshold and evaluation criteria are formulated. Based on the data processing and feature extraction of the previous step, features highly related to vibration over-standard risk are selected as model inputs.

[0076] Step S32, select and train the model. According to different vibration data, select the corresponding model. For small sample, high dimensional feature classification problem, select support vector machine (SVM), which can effectively process nonlinear problems. Neural network (especially deep learning model) has strong feature extraction and nonlinear mapping ability, so it is suitable for processing complex vibration data mode. And random forest is an ensemble learning method, which is suitable for processing high-dimensional data, and has good accuracy and anti-overfitting ability. After determining the model, the data set is divided into training set, validation set and test set to evaluate the performance of the model. Finally, the model is trained using the training set data, and the model parameters are optimized through the validation set to obtain the best prediction performance.

[0077] Step S33, evaluate and validate the model. In order to ensure the accuracy and reliability of the model in predicting the vibration of the building structure, consider using accuracy, precision, recall, F1 score and other indicators to evaluate the performance of the model. K-fold cross-validation and other methods are also used to verify the stability and generalization ability of the model to avoid overfitting problem of the model.

[0078] Step S34, deploy the model and continuous learning. The trained model is applied to the actual vibration monitoring system, and the model is updated and optimized regularly by using newly collected data to maintain the adaptability and accuracy of the model to cope with possible environmental and condition changes.

[0079] Further, step S4 is the warning and control decision module of the present application. When the risk assessment result output by the vibration exceeding threshold evaluation model exceeds the warning threshold, the warning mechanism is started, and the relevant management personnel and residents are informed in time through various ways such as short message, application notification, etc. According to the warning level, the corresponding control measures are put forward, such as adjusting the running speed of rail transit, strengthening structure reinforcement, etc.

[0080] Figure 5 The flowchart of a specific implementation method provided by the embodiment of the present application for starting a warning mechanism and proposing corresponding control measures according to the warning level is shown in the figure.

[0081] The method comprises the following steps:

[0082] Step S41, set up the warning strategy. When the vibration is about to exceed the predetermined safety threshold, the system immediately generates a warning signal, and notifies the relevant management personnel and users through email, short message or application program. According to the severity of the vibration exceeding the threshold, a multi-level warning system (such as yellow, orange, red warning) is designed, and the higher the level of warning, the closer the vibration is to or exceeds the safety threshold.

[0083] Step S42, real-time monitoring and data updating. Deploy a real-time data monitoring system to continuously collect vibration data of the building structure and operation status of the rail transit. Use the latest monitoring data to perform instant data updating and re-evaluation of the model, ensuring the accuracy and timeliness of the early warning information.

[0084] Step S43, execute control decisions. For some cases where the vibration impact can be reduced through technical means, such as adjusting the operation mode, speed, etc. of the rail transit, the system can automatically issue control instructions. For cases that require professional assessment, the system will provide detailed early warning information and analysis reports to support further intervention decisions by human.

[0085] Step S44, develop emergency plans. According to different levels of vibration exceeding risk and possible impact, develop a series of emergency plans, including emergency evacuation, temporary reinforcement of the structure, temporary interruption of rail transit operation, etc. Once a specific level of early warning is triggered, the corresponding level of plan is activated.

[0086] Step S45, notification and feedback. Ensure that relevant parties, such as building managers, residents, rail transit management agencies, etc. can receive early warning information and execution status of emergency plans in a timely manner. Establish a feedback collection mechanism for subsequent analysis and optimization of the model.

[0087] Further, step S5 is the optimization module of the present application, which provides a method for improving the accuracy of early warning and response speed by providing an optimization model.

[0088] Figure 6 The flowchart of the specific implementation method of the optimization model provided by the embodiment of the present application.

[0089] The method comprises the following steps:

[0090] Step S51, continuously collect and analyze data. Continuously collect new vibration data and related rail transit operation data, regularly analyze the prediction performance of the current model on the newly collected data, and identify the deficiencies or biases of the model in prediction performance.

[0091] Step S52, update and optimize features. According to the results of new data analysis, re-evaluate the contribution of each feature to the prediction performance of the model, eliminate ineffective or less influential features, and introduce new beneficial features. Optimize the feature set through feature engineering methods (such as feature construction, feature selection, feature transformation, etc.) to improve the prediction accuracy and generalization ability of the model.

[0092] Step S53, retraining and fine-tuning the model. Using the updated data set and feature set, retrain and fine-tune the model to find the optimal model parameters and structure. Through cross-validation and other methods, comprehensively evaluate the performance of the optimized model to ensure that the model has good prediction effect on different data subsets.

[0093] Step S54, integrate practice feedback. Through feedback collection, obtain feedback information about the accuracy of early warning and the effect of emergency plan execution from early warning practice. Analyze these feedback information to identify the problems and potential improvement space of the early warning system in actual application.

[0094] Step S55, update the knowledge base and decision logic. Update the knowledge base with practice feedback and the latest research results, update the judgment logic and early warning strategy of vibration exceeding. After updating, the decision logic in the control decision module may be adjusted to ensure the accuracy and adaptability of early warning and control decision.

[0095] Step S56, cycle iteration and continuous optimization. Form a continuous optimization iteration process with the above steps to ensure that the model and system can continuously adapt to changes in the external environment and internal operation. To maintain the highest operating efficiency of the system, indicators should be set to monitor the performance of the model and system, and new optimization needs or potential problems should be found in time.

[0096] Therefore, the calculation method of the present application provides an advanced and practical solution. The application range of this method is wide, and it can be applied to various types of building structures with potential vibration risks due to rail transit.

[0097] The above examples are only used to illustrate the technical solutions of the present application, but not to limit it; although the present application has been described in detail with reference to the foregoing examples, those skilled in the art should understand that they can modify the technical solutions recorded in the foregoing examples, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for calculating a pre-warning of excessive vibration of a building structure caused by rail transit operation, characterized in that, The method comprises the following steps: S1: Collecting data, collecting real-time vibration data by setting up a multi-point vibration sensor system at key positions of the building structure and along the track transportation operation line, and integrating special event data and historical vibration data during track transportation operation; S2: Processing data and extracting features, pre-processing the collected vibration data, including denoising, normalization, and using time series analysis and spectral analysis methods to extract vibration features, including frequency, amplitude, and duration; S3: Establishing a vibration over-standard evaluation model, based on the extracted vibration data features, combining the characteristics of the building structure, integrating vibration features including frequency, amplitude, duration, and power spectral density, establishing an evaluation model based on support vector machine and neural network, and outputting over-standard risk evaluation results; S4: Starting the early warning mechanism, when the risk evaluation results output by the vibration over-standard evaluation model exceed the warning threshold, starting the corresponding control measures according to the warning level, including adjusting the track transportation operation speed and strengthening the structure reinforcement; S5: Optimizing the model, improving the warning accuracy and response speed through retraining, integrating practical feedback, and cyclic iteration.

2. The early warning calculation method for building structure vibration exceeding the standard caused by rail transit operation according to claim 1, characterized in that, In step S1: The special event data includes high-speed passing and emergency braking event data; Real-time data and historical data are integrated by time stamp alignment.

3. The early warning calculation method for building structure vibration exceeding the standard caused by rail transit operation according to claim 1, characterized in that, In step S3, the vibration over-standard evaluation standard is combined with the characteristics of the building structure.

4. The early warning calculation method for building structure vibration exceeding the standard caused by rail transit operation according to claim 1, characterized in that, In step S4, the early warning mechanism is started, and relevant management personnel and residents are informed in a timely manner, including the following steps: Step S41: Designing a multi-level warning strategy, including yellow, orange, and red warning levels; Step S42: Real-time monitoring and data updating; Step S45: Notification and feedback.

Images