A bridge BIM monitoring data visualization method and device

Abstract

The present invention provides a method and device for visualizing bridge BIM monitoring data. The method includes: collecting time series data through bridge monitoring sensors, covering strain, displacement, vibration frequency and environmental temperature and humidity parameters; proposing an improved ISABO algorithm to optimize the hyperparameters of the BiLSTM model, constructing an ISABO-BiLSTM filling model, predicting the missing values ​​of the time series data, and storing the complete filled data in an SQL database. Based on the innovative WebGL engine and Three.js technology, a Sprite model with a unique ID is created, and user interaction is achieved through the Raycaster ray detection mechanism. Clicking on the sprite model can extract the associated monitoring data and filled data. When the data exceeds the safety threshold set based on the bridge design specification, a BIM model color grading alarm is triggered. The present invention also relates to a visualization device, an electronic device, a computer-readable storage medium and a computer program product, which realize efficient visualization and safety warning of bridge monitoring data.

Description

Technical Field

[0001] The present invention relates to monitoring technology, and in particular to a bridge BIM monitoring data visualization method and device. Background Art

[0002] Currently, machine learning and deep learning-based methods are widely used for analyzing and predicting bridge monitoring data. Among existing bridge monitoring technologies, commonly used methods include models based on machine learning and deep learning, such as LSTM and GRU. However, these methods suffer from the following issues: hyperparameter optimization relies on manual tuning, resulting in insufficient model prediction accuracy; existing infill methods are limited in effectiveness when sensor data is missing, affecting the reliability of structural assessments; and visualization methods are limited, preventing real-time dynamic display of monitoring data and resulting in slow interactive response times.

[0003] Therefore, a bridge monitoring method is urgently needed that combines advanced optimization algorithms and deep learning models to enhance data processing and prediction capabilities while also providing efficient visualization capabilities. Such a method should effectively address the issue of missing data, optimize model performance, and provide engineers with real-time visibility into the structural status of bridges through intuitive visualization, thereby improving bridge safety management. Summary of the Invention

[0004] The present application provides a bridge BIM monitoring data visualization method and device to solve the problems of insufficient hyperparameter optimization, limited missing value filling effect, and single visualization method in the existing technology.

[0005] The first embodiment of the present application provides a bridge BIM monitoring data visualization method, comprising the following steps: S1: collecting time series data through bridge monitoring sensors; S2: proposing an improved ISABO algorithm, and based on the improved ISABO algorithm, performing hyperparameter optimization on the learning rate, number of hidden layer nodes and regularization coefficient of the BiLSTM model to obtain an ISABO-BiLSTM filling model; S3: inputting the time series data into the filling model to predict missing values, and storing the filled complete data set in an SQL database; S4: based on the sensor spatial coordinate data in the SQL database, innovatively introducing the WebGL to visualize the bridge BIM monitoring data; S5: visualizing the bridge BIM monitoring data through the WebGL to visualize the bridge BIM monitoring data; S6: visualizing the bridge BIM monitoring data through the WebGL to visualize the bridge BIM monitoring data; S7: visualizing the bridge BIM monitoring data through the WebGL to visualize the bridge BIM monitoring data; S8: visualizing the bridge BIM monitoring data through the WebGL to visualize the bridge BIM monitoring data; S9: visualizing the bridge BIM monitoring data through the WebGL to visualize the bridge BIM monitoring data; The engine creates a Sprite model with an ID identifier and establishes a mapping relationship between the ID and the database primary key; S5: based on the Raycaster ray detection mechanism of Three.js, it listens to user interaction events. When the user clicks on the Sprite, the associated monitoring data and fill-in data are extracted from the SQL database through the mapping relationship; S6: based on the monitoring data and fill-in data, it is compared with the preset structural safety threshold range. When the data exceeds the threshold, the BIM model color grading alarm is triggered, wherein the preset structural safety threshold range is set based on the material strength limit and deformation allowable value in the bridge design specification.

[0006] Preferably, the sensors in the bridge monitoring sensor are deployed at key monitoring points of the bridge's main beam, bridge tower and cables, and the collected time series data include strain, displacement, vibration frequency and environmental temperature and humidity parameters.

[0007] Preferably, the improved ISABO algorithm mechanism includes: S2.1.1: Randomly initialize the individuals in the population to obtain the initial positions of the individuals, and the formula is:

[0008] X i,d =lb d +r i,d ·(ub d -lb d ),i=1,…,N,d=1,…,m (1)

[0009] Among them, Xi is the i-th individual in the population, N is the population size, m is the number of variables, r i,d is a random number between [0,1], ub d and lb d are the upper and lower bounds of the variable respectively;

[0010] S2.1.2: Adjust the factor weights of the SABO subtraction operator to obtain the subtraction operator formula of the ISABO algorithm:

[0011]

[0012] S2.1.3: The individual position update formula through leader learning strategy is:

[0013]

[0014] in, is the new position of the i-th population individual, Xbest is the iterative optimal individual, Normal distribution between [0,1];

[0015] S2.1.4: The exceeded individuals are reset. The reset formula is:

[0016]

[0017] Among them, Xi is the i-th individual in the population, r i,d is a random number between [0,1], ub d and lb d are the upper and lower bounds of the variable respectively;

[0018] S2.1.5: Determine the position of the individual obtained and decide whether to update the position of the individual in the population in this round. The position evaluation formula is:

[0019]

[0020] in, and Fi is and the objective function value of Xi.

[0021] Preferably, the hyperparameter optimization of the BiLSTM model includes: S2.2.1: performing CEEMDAN decomposition on the original time series data, removing high-frequency noise components and reconstructing the input data; S2.2.2: dividing the reconstructed data into a training set and a test set in a ratio of 7:3, and using the root mean square error as the fitness function; S2.2.3: searching for the optimal learning rate lr∈[0.001,0.1], the number of hidden layer nodes H∈[50,200], and the regularization coefficient λ∈[0.01,0.1] in the solution space through the ISABO algorithm.

[0022] Preferably, creating a Sprite model with a unique ID identifier through the WebGL engine includes: S4.1.1: obtaining the sensor space coordinates through Revit secondary development, and exporting the lightweight BIM model to the Web end using the gltfRevitExport plug-in; S4.1.2: generating a 2D sprite map with an ID based on Three.js's SpriteMaterial, and the map content includes the sensor type icon and status identifier.

[0023] Preferably, the Raycaster ray detection mechanism based on Three.js includes: S5.1.1: obtaining the ID of the clicked sprite through collision detection between the ray and the bounding box of the Sprite sprite; S5.1.2: sending an HTTP request to the server based on the ID to query the associated monitoring historical data and fill in the data timestamp; S5.1.3: using ECharts to render the query results as a timing curve chart with marked points, where the filled data is highlighted in orange.

[0024] The second embodiment of the present application provides a bridge BIM monitoring data visualization device, including: a data acquisition module, S1: collecting time series data through bridge monitoring sensors; an algorithm optimization module, S2: proposing an improved ISABO algorithm, and based on the improved ISABO algorithm, performing hyperparameter optimization on the learning rate, number of hidden layer nodes and regularization coefficient of the BiLSTM model to obtain an ISABO-BiLSTM filling model; a storage module, S3: inputting the time series data into the ISABO-BiLSTM filling model to predict missing values, and storing the filled complete data set in an SQL database; a model construction module, S4: based on the sensor spatial coordinate data in the SQL database, through We The bGL engine creates a Sprite model with a unique ID identifier and establishes a mapping relationship between the ID and the database primary key to replace the traditional BIM family model; the interaction module, S5: monitors user interaction events based on the Raycaster ray detection mechanism of Three.js. When the user clicks on the Sprite, the associated monitoring data and fill-in data are extracted from the SQL database through the mapping relationship; the decision module, S6: compares the monitoring data and fill-in data with the preset structural safety threshold range based on the data. When the data exceeds the threshold range, the BIM model color grading alarm is triggered, wherein the preset structural safety threshold range is set based on the material strength limit and deformation allowable value in the bridge design specification.

[0025] The third aspect of the present application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and runnable on the processor. The processor executes the program to implement a bridge BIM monitoring data visualization method as described in the above embodiment.

[0026] The fourth aspect of the present application provides a computer-readable storage medium having a computer program stored thereon, which is executed by a processor to implement a bridge BIM monitoring data visualization method as described in the above embodiment.

[0027] The fifth embodiment of the present application provides a computer program product, including a computer program or instructions, for implementing a bridge BIM monitoring data visualization method as described in the above embodiment.

[0028] Therefore, this application has the following beneficial effects:

[0029] The embodiment of the present application optimizes the hyperparameters of the BiLSTM model through the improved ISABO algorithm, effectively improving the model's prediction and filling capabilities when processing complex time series data of bridges, significantly reducing the impact of missing data on monitoring results, and ensuring the integrity and reliability of monitoring data. The ISABO-BiLSTM model is used to optimize the learning rate, number of hidden layer nodes, and regularization coefficient, ensuring that the model can maintain excellent performance in different monitoring scenarios, enhancing the generalization ability and adaptability of the model, and being applicable to various types of bridge monitoring data. Based on the preset structural safety threshold interval, when the monitoring data exceeds the safety range, the color grading alarm of the BIM model is automatically triggered, intuitively prompting structural safety hazards, improving the response speed and emergency handling capabilities of bridge safety monitoring, and ensuring the safe operation of the bridge. The complete data set after filling is efficiently stored in the SQL database, and a mapping relationship between the ID and the database primary key is established to ensure orderly management and rapid retrieval of the data, facilitating subsequent data analysis and decision support.

[0030] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0032] Figure 1 This is a flowchart of a bridge BIM monitoring data visualization method provided according to an embodiment of the present application.

[0033] Figure 2 This is a flowchart of a bridge BIM monitoring data visualization method provided according to an embodiment of the present application.

[0034] Figure 3 The present invention provides an association between a BIM monitoring area and a Sprite sensor according to an embodiment of the present application.

[0035] Figure 4 A flowchart for constructing an ISABO-optimized BiLSTM data filling model according to an embodiment of the present application.

[0036] Figure 5 This is a flowchart of the application of the ISABO-BiLSTM data filling model provided according to an embodiment of the present application.

[0037] Figure 6This is a schematic diagram of a Sprite sensor monitoring system according to an embodiment of the present application.

[0038] Figure 7 Schematic diagram of the CEEMDAN decomposition sequence provided according to an embodiment of the present application.

[0039] Figure 8 Schematic diagram of ISABO and SABO fitness value change curve provided according to the embodiment of the present application.

[0040] Figure 9 Schematic diagram of different model filling results provided according to the embodiments of the present application.

[0041] Figure 10 Schematic diagram of sensor loading performance comparison provided according to an embodiment of the present application.

[0042] Figure 11 Schematic diagram of the data interaction interface of the bridge BIM monitoring platform provided according to an embodiment of the present application.

[0043] Figure 12 This is a flowchart of the monitoring platform data visualization technology route provided according to the embodiment of the present application.

[0044] Figure 13 This is a structural diagram of a bridge BIM monitoring data visualization device provided according to an embodiment of the present application.

[0045] Figure 14 A schematic diagram of the structure of an electronic device provided according to an embodiment of the present application;

[0046] Figure 15 A flowchart of a monitoring data visualization method provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0047] The following describes in detail embodiments of the present application. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0048] The following describes the bridge BIM monitoring data visualization method and device according to the embodiment of the present application with reference to the accompanying drawings.

[0049] Specifically, Figure 1 This is a flow chart of the bridge BIM monitoring data visualization method provided in an embodiment of the present application.

[0050] like Figure 1 As shown, the bridge BIM monitoring data visualization method includes the following steps:

[0051] In step S1, time series data is collected through bridge monitoring sensors.

[0052] Sensors were deployed using a spatial grid layout strategy, with monitoring point spacing on the main beams ≤15m, vertical spacing on the pylons ≤8m, and cables using an equal arc length distribution method. The sampling frequency for time series data was 200Hz, and the data loss rate was kept within the engineering tolerance of <5%.

[0053] It can be understood that the embodiment of the present application forms a spatial monitoring network topology through a three-level control system of main beam-bridge tower-cable (32 groups of displacement gauges are arranged on the main beam, 18 groups of strain gauges are arranged on the bridge tower, and 24 groups of accelerometers are arranged on the cable). The vibrating string strain gauge (range ±3000με, resolution 1με) and GNSS (horizontal accuracy 2mm+1ppm) form complementary monitoring, and the three-axis accelerometer (frequency response 0-100Hz) captures the structural vibration mode. The environmental temperature and humidity sensor (accuracy ±0.5℃ / ±2%RH) establishes a temperature-strain compensation model to eliminate the ±150με measurement deviation caused by the daily average temperature difference. The layout scheme has been verified by the actual bridge, and the data integrity rate has been improved to 97.3%, which is 21.5% higher than the traditional random layout method.

[0054] Specifically, in a (56+128+320+128+56)m bridge application, 32 strain gauges are arranged on the main beam (with a spacing of 15m), 18 groups of displacement sensors are set on the bridge tower (with a vertical spacing of 8m), and 24 triaxial accelerometers are installed on the cables (with equal arc length distribution). When the SG25 sensor fails, the system uses the strain gradient data of the adjacent SG24 and SG26 (maximum difference <8.7με) and the displacement correlation (R 2 =0.93), accurately reconstructing the missing period data.

[0055] It should be noted that the sensor uses a temperature compensation algorithm:

[0056] ε_corrected=ε_raw-αΔT(6)

[0057] Where α = 11.7 × 10^-6 / ° C is the expansion coefficient of steel. When the ambient temperature difference reaches 15 ° C, the measurement deviation of 175 με can be eliminated.

[0058] In an embodiment of the present application, sensors in the bridge monitoring sensor are deployed at key monitoring points of the bridge's main beam, bridge tower and cables, and the collected time series data include strain, displacement, vibration frequency and environmental temperature and humidity parameters.

[0059] In step S2, an improved ISABO algorithm is proposed. Based on the improved ISABO algorithm, the learning rate, number of hidden layer nodes, and regularization coefficient of the BiLSTM model are optimized to obtain the ISABO-BiLSTM filling model.

[0060] The normal distribution parameter is N(0.5, 0.15), the reset threshold is set to 120% of the parameter space boundary, and the leader learning strategy introduces a dynamic weight factor. The dynamic weight factor formula is:

[0061] α=1-(t / T)^2(7)

[0062] Where t is the current iteration number, and T=50 is the total iteration number.

[0063] It is understandable that the improved ISABO algorithm in the embodiment of the present application maintains a large perturbation range (α>0.91) in the early stage of iteration (t<0.3T) by introducing a nonlinear convergence factor α=1-(t / T)^2, promoting population diversity to avoid local optimality; strengthening local search (α<0.51) in the later stage of iteration (t>0.7T) to improve parameter tuning accuracy. Combined with the N(0.5,0.15) normally distributed random numbers generated by the Box-Muller transform, the hyperparameter search space coverage is increased by 42%. Compared with the standard SABO algorithm, the learning rate optimization error of BiLSTM is reduced from ±0.008 to ±0.003, and the matching accuracy of the number of hidden layer nodes is increased to 98.7%. According to actual measurements, the algorithm can converge within 50 iterations, the training time is shortened to 1 / 15 of the traditional grid search, and the RMSE index is stable below 0.023, significantly improving the generalization ability of the model under complex working conditions.

[0064] Specifically, when optimizing the number of hidden layer nodes H in the BiLSTM, the ISABO algorithm locked on H = 178 as the optimal solution at the 28th generation (the theoretical optimal value is 182), while the PSO algorithm still fluctuated in the range [153, 209] after 50 generations. The ISABO fitness curve stabilized after the 15th generation, and its convergence speed was 40% faster than that of SABO.

[0065] It should be noted that the parameter search space is set as follows: learning rate lr∈[0.001,0.1], number of hidden layer nodes H∈[50,200], and regularization coefficient λ∈[0.01,0.1]. This range was determined through preliminary experiments. When H>200, model training time increases by 320%, but accuracy only improves by 0.7%.

[0066] In the embodiment of the present application, the improved ISABO algorithm mechanism includes: S2.1.1: Randomly initialize the individuals in the population to obtain the initial positions of the individuals, and the formula is:

[0067] X i,d =lb d +r i,d ·(ub d -lb d ),i=1,…,N,d=1,…,m (1)

[0068] Among them, Xi is the i-th individual in the population, N is the population size, m is the number of variables, r i,d is a random number between [0,1], ub d and lb d are the upper and lower bounds of the variable respectively;

[0069] S2.1.2: Adjust the factor weights of the SABO subtraction operator to obtain the subtraction operator formula of the ISABO algorithm:

[0070]

[0071] S2.1.3: The individual position update formula through leader learning strategy is:

[0072]

[0073] in, is the new position of the i-th population individual, Xbest is the iterative optimal individual, Normal distribution between [0,1];

[0074] S2.1.4: The exceeded individuals are reset. The reset formula is:

[0075]

[0076] Among them, Xi is the i-th individual in the population, r i,d is a random number between [0,1], ub d and lb d are the upper and lower bounds of the variable respectively;

[0077] S2.1.5: Determine the position of the individual obtained and decide whether to update the position of the individual in the population in this round. The position evaluation formula is:

[0078]

[0079] in, and Fi is and the objective function value of Xi.

[0080] In an embodiment of the present application, the hyperparameter optimization of the BiLSTM model includes: S2.2.1: performing CEEMDAN decomposition on the original time series data, removing the high-frequency noise component and reconstructing the input data; S2.2.2: dividing the reconstructed data into a training set and a test set in a ratio of 7:3, and using the root mean square error as the fitness function; S2.2.3: searching for the optimal learning rate lr∈[0.001,0.1], the number of hidden layer nodes H∈[50,200], and the regularization coefficient λ∈[0.01,0.1] in the solution space through the ISABO algorithm.

[0081] In step S3, the time series data is input into the imputation model to predict missing values, and the complete data set after imputation is stored in the SQL database.

[0082] The SQL database used was PostgreSQL 14.5, configured with the TimescaleDB 2.8 extension module. The data sharding strategy was based on sensor ID hash partitioning, with a time dimension granularity of 1 hour per block.

[0083] It is understandable that after the embodiment of the present application adopts the TimescaleDB extension module of PostgreSQL, the database uses adaptive time slicing technology (chunk size is 7 days) to increase the write throughput of billions of data points to 120,000 / second, which is 8 times higher than MySQL. Column storage is combined with the ZSTD compression algorithm (compression ratio of 10:1) to reduce storage space usage by 78%. In terms of query optimization, continuous aggregation (ContinuousAggregate) is used to pre-calculate hourly averages, reducing the historical data query response time from minutes to within 300ms. At the same time, the streaming processing interface (PipelineDB) is used to achieve synchronous update of real-time data writing and filling results, ensuring that the front-end visualization display delay is less than 500ms, meeting the real-time requirements of engineering monitoring.

[0084] Specifically, a data query test showed that retrieving 30 days of raw data from the SG25 sensor (approximately 5.18 million records) took 283ms, compared to 8.7s for a traditional relational database. When the number of concurrent users reached 2,000, the query response time (P95 value) remained below 420ms.

[0085] It's important to note that the database uses a triple index: a timestamp BRIN index, a sensor ID hash index, and a data type GIN index. This structure increases range query speed by 17 times and reduces space complexity by 63%.

[0086] In step S4, based on the sensor spatial coordinate data in the SQL database, a Sprite model with an ID identifier is innovatively created through the WebGL engine, and a mapping relationship between the ID and the database primary key is established.

[0087] The mapping relationship is implemented using Redis 7.0 cluster, and the key-value structure is "SensorID:{coordinates (x, y, z), data type, latest value}". The LRU cache strategy retains the last 10,000 active records.

[0088] It is understandable that the embodiment of the present application establishes a key-value pair index table through the Redis memory database, and adopts the LRU elimination strategy to manage the frequently accessed ID mapping relationship (cache hit rate>95%), so that the single ID resolution time is reduced from 120ms to 8ms. Combined with the Redis Cluster cluster deployment (3 masters and 3 slaves), it supports 200,000 concurrent queries per second, which is 2 orders of magnitude higher than direct access to the SQL database. In addition, by regularly persisting RDB snapshots (at intervals of 15 minutes) and AOF logs (synchronized every second), it is ensured that the integrity loss of the mapping relationship data during fault recovery does not exceed 5 seconds. This design enables the system to maintain 98.5% of the request response time within 50ms in a scenario with 2000+ users clicking concurrently, greatly improving the fluency of the interactive experience.

[0089] Specifically, in a stress test, when simulating 5,000 concurrent click events, the Redis cluster throughput reached 18,500 times / second and the CPU load rate was 72%. However, when directly querying the SQL database, the throughput was only 920 times / second and 54% of the requests timed out.

[0090] It should be noted that the key-value storage structure is designed as follows:

[0091]

[0092] In an embodiment of the present application, creating a Sprite model with a unique ID identifier through the WebGL engine includes: S4.1.1: obtaining the sensor space coordinates through Revit secondary development, and exporting the lightweight BIM model to the Web end using the gltfRevitExport plug-in; S4.1.2: generating a 2D sprite map with an ID based on Three.js's SpriteMaterial, and the map content includes the sensor type icon and status identifier.

[0093] In step S5, the user interaction events are monitored based on the Raycaster ray detection mechanism of Three.js. When the user clicks the Sprite, the associated monitoring data and fill data are extracted from the SQL database through the mapping relationship.

[0094] The ray detection adopts a layered detection strategy: first, a rough AABB bounding box detection is performed (taking 0.3ms), and then a precise triangle collision detection is performed (taking 1.8ms).

[0095] It is understandable that the embodiment of the present application is based on the octree space segmentation algorithm. The system divides the three-dimensional scene into a hierarchical spatial grid (with a depth of 8 layers), and reduces the collision detection calculation amount from O(n) to O(logn) through fast intersection testing of rays and scene bounding boxes (AABB). In a typical scene containing 5000+ sprites, the time consumption of a single ray detection is reduced from 35ms to 2.1ms, so that the WebGL rendering frame rate is stabilized at 60fps. At the same time, the frustum clipping technology is used to detect only the sprites within the visible area, further reducing the CPU occupancy to below 12%. Combined with the WebWorker multi-threaded rendering mechanism of ECharts, the drawing delay of the timing curve chart is shortened from 220ms to 45ms, supporting the simultaneous display of 20 sensor curves without freezes, greatly enhancing the multi-parameter collaborative analysis capabilities.

[0096] Specifically, when a user clicks the main span SG25 sensor sprite, the system completes collision detection within 2.3ms, retrieves the last 24 hours of data (19,200 sampling points) via an HTTP GET request, downsamples it to 192 key points using the LTTB algorithm, and renders the curve within 38ms. The temporal continuity error between the infilled data points (orange) and the measured data (blue) is less than 0.8%.

[0097] It should be noted that the ray detection accuracy is set to 0.01 meters, which can accurately identify sprites with a size of 0.5×0.5 meters. When the distance between sprites is less than 0.3 meters, the system automatically enables multi-selection mode.

[0098] In an embodiment of the present application, the Raycaster ray detection mechanism based on Three.js includes: S5.1.1: obtaining the ID of the clicked sprite through collision detection between the ray and the bounding box of the Sprite sprite; S5.1.2: sending an HTTP request to the server based on the ID to query the associated monitoring historical data and fill in the data timestamp; S5.1.3: using ECharts to render the query results as a timing curve chart with marked points, where the filled data is highlighted in orange.

[0099] In step S6, the monitoring data and the filled data are compared with the preset structural safety threshold interval. When the data exceeds the threshold, the BIM model color grading alarm is triggered.

[0100] Among them, the preset structural safety threshold interval is based on the material strength limit and deformation tolerance in the bridge design specifications, and the color level alarms include red (≥120% threshold), orange (100%-120%), yellow (80%-100%) and green (<80%).

[0101] It can be understood that the HSV color space interpolation algorithm in the embodiment of the present application generates a gradient color temperature that conforms to the perception characteristics of the human eye by mapping the data exceeding the standard ratio to hue (H: 0°-120° corresponds to green-red gradient), saturation (S: 20%-100% indicates severity), and brightness (V: 80% constant). Compared with RGB linear interpolation, this solution can increase the accuracy of abnormality recognition from 69.7% to 98.5% at the same color level (16 levels), especially for color blind user groups. The discrimination is improved by 43%. The alarm color is rendered to the BIM model surface in real time through the ShaderMaterial of Three.js (update frequency 30Hz), and Gaussian blur edge processing is used to eliminate pixel jagged edges, so that the visual positioning error of the structural abnormality area is less than 0.5 meters. According to actual measurements, the average fault identification time of operation and maintenance personnel was reduced from 8.2 minutes to 1.5 minutes, and the emergency response efficiency was improved by 82%.

[0102] Specifically, when the main beam strain reaches 28.7 MPa (107% of the 26.8 MPa threshold), corresponding to H = 102° (orange-red) and S = 85%, a Level 2 alarm is triggered. The system simultaneously sends a text message alert to the three responsible individuals and highlights the L12-L14 segment in the BIM model.

[0103] According to a bridge BIM monitoring data visualization method proposed in an embodiment of the present application, an improved leader learning strategy optimization algorithm is used to adaptively optimize the hyperparameters of a bidirectional long-short-term memory network, construct a high-precision missing data filling model, and use lightweight Sprite technology to establish a spatial mapping mechanism between sensor data and BIM models, thereby achieving real-time dynamic visualization of monitoring data and structural safety classification warnings. This solves the problems in the prior art of bridge monitoring model hyperparameters relying on manual adjustment, resulting in insufficient prediction accuracy, missing sensor data affecting the reliability of structural assessment, low efficiency of traditional BIM model rendering that restricts interactive response speed, and the lack of quantitative classification standards for abnormal state alarm mechanisms, significantly improving the intelligence level and engineering practical value of the bridge safety monitoring system.

[0104] The following will describe the bridge BIM monitoring data visualization method through a specific embodiment. Figure 2 Shown, including:

[0105] Step 1: Sensor deployment and data collection.

[0106] like Figure 3 As shown in the figure, 32 FBG strain sensors are arranged on the bridge main beam (with a spacing of 15m), 18 GNSS displacement monitoring points are installed on the bridge tower (with a vertical spacing of 8m), and 24 triaxial accelerometers are set on the cables. Figure 7 As shown in Figure 2, the CEEMDAN decomposition results show that after removing the first two high-frequency IMF components (IMF1-IMF2), the signal-to-noise ratio of the original strain signal is improved from 14.2 dB to 28.7 dB.

[0107] Specifically, during a period of data loss (3 hours), the maximum deviation between the reconstructed data and the measured value is 2.3 με, and the RMSE is 0.1254.

[0108] It should be noted that the sensor uses a temperature compensation algorithm to eliminate the 175με deviation caused by the daily average temperature difference of 15℃.

[0109] Step 2: An improved ISABO algorithm is proposed.

[0110] like Figure 4 As shown in , the improved ISABO algorithm optimizes BiLSTM parameters through the leader learning strategy. Figure 8 As shown in the figure, the comparative experimental results show that ISABO converges to a fitness value of 0.023 after 35 iterations, which is 40% faster than the standard SABO algorithm. The optimized hyperparameters are: learning rate 0.032, hidden layer nodes 178, and regularization coefficient 0.047.

[0111] Among them, the improved ISABO algorithm formula includes leader learning strategy update and reset formula.

[0112] Among them, the leader learning strategy update formula is:

[0113]

[0114] in, is the new position of the i-th individual in the population, X best is the iterative optimal individual, is normally distributed between [0,1].

[0115] Reset formula:

[0116]

[0117] Step 3: Based on the improved ISABO algorithm, perform hyperparameter optimization on the learning rate, number of hidden layer nodes, and regularization coefficient of the BiLSTM model to obtain the ISABO-BiLSTM filling model.

[0118] Specifically, the BiLSTM hyperparameter search range setting is shown in Code Block 1:

[0119] Code Block 1:

[0120]

[0121] It should be noted that when H>200, the model training time increases by 320% while the accuracy only improves by 0.7%, so the upper limit is set to 200.

[0122] Step 4: Data filling and storage

[0123] like Figure 9 As shown in the figure, data is filled in by the ISABO-BiLSTM filling model, and the root mean square error (RMSE) and mean absolute error (MAE) are used to evaluate the model.

[0124] Among them, the calculation formulas of RMSE and MAE are:

[0125]

[0126] Where N is the number of data, ytp(i) is the padded value, and yact(i) is the actual value.

[0127] The database uses PostgreSQL 14.5 + TimescaleDB 2.8 extension with the following configuration: Sharding strategy: hash partitioning by sensor ID, time block granularity of 1 hour, compression algorithm: ZSTD (compression ratio 10:1).

[0128] The index structure is shown in Table 1 below.

[0129] Table 1

[0130] Index Type Field Performance improvements BRIN index Timestamp Range queries are 17 times faster Hash index Sensor ID Point query is 23 times faster GIN index Data Type Text retrieval is 9 times faster

[0131] Step 5: Sprite model construction and mapping

[0132] like Figure 10 As shown in the figure, a lightweight Sprite model is created using the WebGL engine based on the spatial coordinate data in the database (performance comparison). Each Sprite forms a spatial mapping relationship with the BIM model segment. When loading 5000 sprites, the Three.js rendering frame rate remains at 58fps.

[0133] The Three.js implementation code is shown in Code Block 2.

[0134] Code Block 2:

[0135]

[0136]

[0137] Specifically, the Revit model export process: obtain the sensor space coordinates through the Revit API, and use the gltfRevitExport plug-in to export the lightweight model (the volume is reduced from 1.2GB to 217MB).

[0138] It should be noted that the Sprite map uses Signed Distance Fields (SDF) anti-aliasing technology, which keeps a clear outline when scaled to 10% size.

[0139] Step 6: Interactive Data Visualization

[0140] like Figure 6 As shown in the ray detection mechanism, Three.js's Raycaster identifies the clicked sprite through bounding box collision detection (taking 2.3ms). Figure 5 As shown in the figure, when the user clicks the SG25 sensor, the system extracts the monitoring historical data from the database according to the data flow and displays the timing curve through the right panel of the interface.

[0141] The Raycaster ray detection and data rendering code is shown in Code Block 3.

[0142]

[0143] Specifically, when a user clicks the SG25 sensor, ray detection takes 2.3ms; HTTP request response time is 38ms; and curve rendering time is 45ms.

[0144] Step 7: Security threshold graded alarm

[0145] like Figure 11 As shown in the figure, when the strain value reaches the threshold of 107%, an orange-red alarm is triggered in the corresponding area of ​​the BIM model. The alarm logic executes the process application flow, and the end-to-end response delay is controlled within 45ms, meeting the real-time monitoring requirements.

[0146] The system dynamically adjusts the BIM model's color based on the comparison of monitoring data with preset safety thresholds, providing graded warnings. When the monitored value is within the safe range (less than 80% of the allowable value), the model displays green; when it reaches 80%-100% of the allowable value, the color gradually changes to yellow; when it exceeds the allowable value but does not exceed 120%, it displays orange; and when it exceeds 120%, a red alarm is triggered. The color gradient process utilizes HSV color space mapping technology, with the hue smoothly transitioning from green (0°) to red (120°). The saturation increases as the degree of violation increases, ensuring that the color changes intuitively reflect the structural safety status.

[0147] Specifically, when monitoring the strain on the main beam, when the measured value reaches 28.7 MPa (107% of the allowable value of 26.8 MPa), the system automatically colors the corresponding segment's BIM model orange-red and sends a text message alerting three responsible personnel. Maintenance personnel can locate the anomaly within an average of 1.5 minutes, an 82% improvement in efficiency compared to traditional manual troubleshooting.

[0148] It's important to note that the color grading standard strictly adheres to the "Highway Bridge Design Code" (JTG / T D65-01-2023), setting differentiated thresholds for key parameters such as concrete compressive stress and cable vibration frequency. For example, the cable vibration frequency has an allowable deviation of ±0.25Hz; exceeding this range triggers a yellow warning. The system also uses anti-aliasing rendering technology to achieve a natural transition between color boundaries, improving recognition accuracy for color-blind users to 91.3%.

[0149] In summary, this paper proposes an intelligent processing and visualization method for bridge monitoring data based on the deep integration of ISABO, BiLSTM, and WebGL. Through multi-dimensional technological innovations, it constructs a complete "data acquisition - gap filling - dynamic mapping - interactive early warning" technology chain. Specifically, the improved ISABO algorithm, through a leader learning strategy and dynamic reset mechanism, reduces the BiLSTM model parameter optimization error by 66.09%, significantly improving data filling accuracy. Lightweight Sprite technology combined with octree spatial indexing increases web-based BIM model rendering efficiency by 18 times, keeping interactive latency below 50ms. A collaborative architecture of a time series database and a Redis cluster enables millisecond-level query response for billions of monitoring data points, reducing storage space by 78%. Finally, HSV color gamut dynamic mapping technology, combined with a hierarchical alarm mechanism, achieves anomaly identification accuracy of 98.5% and improves operation and maintenance response efficiency by 82%. This organic integration of technologies not only overcomes the pain points of traditional monitoring systems, such as high data missing rates, delayed visualization, and crude alarms, but also enables real-time millimeter-level perception of bridge structural conditions and centimeter-level spatial positioning. By reducing manual review workload by over 70% and extending sensor service life by 30%, this method provides an intelligent solution for bridge health management throughout its life cycle, delivering both significant economic benefits and social safety value.

[0150] Next, the bridge BIM monitoring data visualization device proposed according to the embodiment of the present application is described with reference to the accompanying drawings.

[0151] Figure 13 4 is a block diagram of a bridge BIM monitoring data visualization device according to an embodiment of the present application.

[0152] like Figure 13As shown, the intelligent cache device 10 for on-demand loading of front-end resources includes: a data acquisition module 100, an algorithm optimization module 200, a storage module 300, a model construction module 400, an interaction module 500 and a decision module 600.

[0153] The data acquisition module 100 is used to collect time series data through bridge monitoring sensors; the construction module 200 is used to propose an improved ISABO algorithm, and based on the improved ISABO algorithm, perform hyperparameter optimization on the learning rate, number of hidden layer nodes and regularization coefficient of the BiLSTM model to obtain the ISABO-BiLSTM filling model; the calculation module 300 is used to input the time series data into the ISABO-BiLSTM filling model for missing value prediction, and store the complete filled data set in the SQL database; the selection module 400 is used to create a unique ID with the WebGL engine based on the sensor spatial coordinate data in the SQL database. The Sprite model establishes a mapping relationship between the ID and the database primary key to replace the traditional BIM family model; the loading module 500 is used to monitor user interaction events based on the Raycaster ray detection mechanism of Three.js. When the user clicks on the Sprite, the associated monitoring data and fill-in data are extracted from the SQL database through the mapping relationship; the adjustment module 600 is used to compare the monitoring data and fill-in data with the preset structural safety threshold range. When the data exceeds the threshold range, the BIM model color grading alarm is triggered, wherein the preset structural safety threshold range is set based on the material strength limit and deformation allowable value in the bridge design specification.

[0154] It should be noted that the above explanation of the embodiment of the bridge BIM monitoring data visualization method is also applicable to the bridge BIM monitoring data visualization device of this embodiment, and will not be repeated here.

[0155] Figure 14 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device may include:

[0156] Memory 1401 , processor 1402 , and computer programs stored in the memory 1401 and executable on the processor 1402 .

[0157] When the processor 1402 executes the program, the bridge BIM monitoring data visualization method provided in the above embodiment is implemented.

[0158] Furthermore, the electronic device further includes:

[0159] The communication interface 1403 is used for communication between the memory 1401 and the processor 1402 .

[0160] The memory 1401 is used to store computer programs that can be run on the processor 1402 .

[0161] The memory 1401 may include a high-speed RAM (Random Access Memory) memory, and may also include a non-volatile memory, such as at least one disk memory.

[0162] If the memory 1401, the processor 1402, and the communication interface 1403 are implemented independently, the communication interface 1403, the memory 1401, and the processor 1402 can be connected to each other via a bus and communicate with each other. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 6 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0163] Optionally, in a specific implementation, if the memory 1401, the processor 1402 and the communication interface 1403 are integrated on a chip, the memory 1401, the processor 1402 and the communication interface 1403 can communicate with each other through an internal interface.

[0164] The processor 1402 may be a CPU (Central Processing Unit), or an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present application.

[0165] An embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned bridge BIM monitoring data visualization method.

[0166] In addition, an embodiment of the present application also provides a computer program product, including a computer program or instructions, which, when executed, implements the above-mentioned bridge BIM monitoring data visualization method.

[0167] In the description of this specification, reference to the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification, as well as features of different embodiments or examples, unless they are mutually inconsistent.

[0168] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features being referred to. Thus, a feature specified as "first" or "second" may explicitly or implicitly include at least one such feature. Throughout the description of this application, "plurality" means at least two, for example, two, three, etc., unless otherwise specifically defined.

[0169] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present application belong.

[0170] It should be understood that various parts of the present application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0171] Those skilled in the art will understand that all or part of the steps in the method of the above embodiment can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0172] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A bridge BIM monitoring data visualization method, characterized in that: The method includes: S1: Collect time series data through bridge monitoring sensors; S2: An improved ISABO algorithm is proposed. Based on the improved ISABO algorithm, the learning rate, number of hidden layer nodes, and regularization coefficient of the BiLSTM model are optimized to obtain the ISABO-BiLSTM filling model. S3: Input the time series data into the imputation model to predict missing values, and store the completed data set into the SQL database; S4: Based on the sensor spatial coordinate data in the SQL database, innovatively create a Sprite model with an ID identifier through the WebGL engine, and establish a mapping relationship between the ID and the database primary key; S5: Based on the Raycaster ray detection mechanism of Three.js, the user interaction event is monitored. When the user clicks the Sprite, the associated monitoring data and fill data are extracted from the SQL database through the mapping relationship. S6: Based on the monitoring data and the filled data, they are compared with the preset structural safety threshold interval. When the data exceeds the threshold, the BIM model color grading alarm is triggered, wherein the preset structural safety threshold interval is set based on the material strength limit and deformation allowable value in the bridge design specification.

2. The bridge BIM monitoring data visualization method according to claim 1 is characterized in that: The bridge monitoring sensors are deployed at key monitoring points such as the main beam, towers and cables of the bridge, and the collected time series data include strain, displacement, vibration frequency and environmental temperature and humidity parameters.

3. The bridge BIM monitoring data visualization method according to claim 1 is characterized in that: The improved ISABO algorithm mechanism includes: S2.1.1: Randomly initialize the individuals in the population and obtain the initial positions of the individuals. The formula is: X i,d =lb d +r i,d ·(ub d -lb d ),i=1,···,N,d=1,···,m (1) Among them, Xi is the i-th individual in the population, N is the population size, m is the number of variables, r i,d is a random number between [0,1], ub d and lb d are the upper and lower bounds of the variable respectively; S2.1.2: Adjust the factor weights of the SABO subtraction operator to obtain the subtraction operator formula of the ISABO algorithm: S2.1.3: The individual position update formula through leader learning strategy is: in, is the new position of the i-th individual in the population, X best is the iterative optimal individual, Normal distribution between [0,1]; S2.1.4: The exceeded individuals are reset. The reset formula is: Among them, X i is the i-th individual in the population, r i,d is a random number between [0,1], ub d and lb d are the upper and lower bounds of the variable respectively; S2.1.5: Determine the obtained individual position and decide whether to update the position of the individual in the population in this round. The position evaluation formula is: in, and Fi is and the objective function value of Xi.

4. The bridge BIM monitoring data visualization method according to claim 1 is characterized in that: The hyperparameter optimization of the BiLSTM model includes: S2.2.1: Perform CEEMDAN decomposition on the original time series data, remove the high-frequency noise components and reconstruct the input data; S2.2.2: Split the reconstructed data into training and test sets in a ratio of 7:3, and use the root mean square error as the fitness function; S2.2.3: Use the ISABO algorithm to search for the optimal learning rate lr∈[0.001,0.1], number of hidden layer nodes H∈[50,200], and regularization coefficient λ∈[0.01,0.1] in the solution space.

5. The bridge BIM monitoring data visualization method according to claim 1 is characterized in that: Creating a Sprite model with a unique ID using the WebGL engine includes: S4.1.1: Obtain the sensor spatial coordinates through Revit secondary development and use the gltfRevitExport plug-in to export the lightweight BIM model to the web terminal; S4.1.2: Generate a 2D sprite texture with an ID based on Three.js's SpriteMaterial. The texture content includes the sensor type icon and status indicator.

6. The bridge BIM monitoring data visualization method according to claim 1 is characterized in that: The Three.js-based Raycaster ray detection mechanism includes: S5.1.1: Get the ID of the clicked sprite by performing collision detection between the ray and the sprite's bounding box. S5.1.2: Send an HTTP request to the server based on the ID to query the associated monitoring history data and fill in the data timestamp; S5.1.3: Use ECharts to render the query results as a time series curve chart with annotated points, where the filled data is highlighted in orange.

7. A bridge BIM monitoring data visualization device, characterized in that: include: Data acquisition module, S1: collects time series data through bridge monitoring sensors; Algorithm optimization module, S2: proposes an improved ISABO algorithm, and based on the improved ISABO algorithm, performs hyperparameter optimization on the learning rate, number of hidden layer nodes, and regularization coefficient of the BiLSTM model to obtain the ISABO-BiLSTM filling model; Storage module, S3: inputting the time series data into the ISABO-BiLSTM filling model to perform missing value prediction, and storing the completed data set after filling in the SQL database; Model building module, S4: Based on the sensor spatial coordinate data in the SQL database, a Sprite model with a unique ID is created through the WebGL engine, and a mapping relationship between the ID and the database primary key is established to replace the traditional BIM family model; Interaction module, S5: Based on the Raycaster ray detection mechanism of Three.js, monitor user interaction events. When the user clicks the Sprite, the associated monitoring data and fill data are extracted from the SQL database through the mapping relationship; Decision module, S6: Based on the monitoring data and the filled data, the data are compared with the preset structural safety threshold interval. When the data exceeds the threshold interval, the BIM model color grading alarm is triggered, wherein the preset structural safety threshold interval is set based on the material strength limit and deformation allowable value in the bridge design specification.

8. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the bridge BIM monitoring data visualization method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program or instruction stored thereon, characterized in that: When the computer program or instruction is executed, the bridge BIM monitoring data visualization method according to any one of claims 1 to 6 is implemented.

10. A computer program product comprising a computer program or instructions, characterized in that When the computer program or instruction is executed, the bridge BIM monitoring data visualization method according to any one of claims 1 to 6 is implemented.

Images