Foundation pit monitoring system based on BIM

Abstract

The invention provides a BIM-based foundation pit monitoring system, and relates to the technical field of foundation pit monitoring. The invention discloses a foundation pit monitoring system based on BIM. Comprising a data acquisition module, a data preprocessing module, a multi-modal data fusion module, an edge calculation module, a real-time data stream processing module, a data transmission and communication module, a real-time data storage module, a real-time analysis and prediction module, an early warning and alarm module and a BIM model integration and visualization module. A decision support and optimization module; and a system management and monitoring module. Preliminary data processing and analysis are carried out on edge equipment close to a data source through an edge computing module, only processed key data and analysis results are transmitted to a central control system, data transmission quantity is reduced, dependence on network bandwidth is reduced, data transmission delay is reduced through local processing, and data transmission efficiency is improved. The multi-modal data fusion module solves the problems of data islands and intellectualization, and improves the accuracy of early warning and prediction by comprehensively analyzing various data sources.

Description

Technical Field

[0001] The present invention relates to the technical field of foundation pit monitoring, and in particular to a foundation pit monitoring system based on BIM. Background Art

[0002] With the acceleration of urbanization and the continuous advancement of infrastructure construction, foundation pit projects have become increasingly common in various construction and transportation projects. Due to their complex geological conditions, surrounding environment and uncertainties in the construction process, foundation pit projects face many safety risks and technical challenges. In order to effectively respond to these challenges, a foundation pit monitoring system based on building information modeling has emerged and combined with a variety of advanced technologies to achieve comprehensive, real-time and intelligent monitoring of the foundation pit construction process.

[0003] There are still some problems in the use of existing BIM-based foundation pit monitoring systems. Traditional foundation pit monitoring systems usually rely on a centralized data processing architecture, and data is transmitted from sensors to a central server for processing and analysis. This approach is prone to data transmission delays and low processing efficiency when faced with large amounts of real-time data. In addition, existing systems often have certain delays when processing and analyzing data, and cannot achieve true real-time monitoring and early warning. Therefore, those skilled in the art provide a BIM-based foundation pit monitoring system to solve the problems raised in the above background technology. Summary of the Invention

[0004] (1) Technical problems solved

[0005] In response to the shortcomings of the existing technology, the present invention provides a BIM-based foundation pit monitoring system, which solves the problem that traditional foundation pit monitoring systems usually rely on a centralized data processing architecture, in which data is transmitted from sensors to a central server for processing and analysis. This approach is prone to data transmission delays and low processing efficiency when faced with large amounts of real-time data. In addition, the existing system often has a certain delay when processing and analyzing data, making it impossible to achieve true real-time monitoring and early warning.

[0006] (2) Technical solution

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: A BIM-based foundation pit monitoring system, comprising:

[0008] Data preprocessing module, which cleans, verifies and calibrates the collected data;

[0009] Real-time data stream processing module, which continuously processes and analyzes real-time data;

[0010] Data transmission and communication module, responsible for transmitting data from sensors to edge computing devices or central servers;

[0011] Real-time data storage module, storing real-time processed data;

[0012] Real-time analysis and prediction module, which uses machine learning and prediction models to conduct real-time analysis of foundation pit status and predict future trends;

[0013] Early warning and alarm module, triggering early warning and alarm mechanisms based on analysis results;

[0014] BIM model integration and visualization module integrates monitoring data with BIM models to achieve three-dimensional visualization;

[0015] Decision support and optimization module, providing suggestions for construction plan optimization and risk control measures;

[0016] System management and monitoring module, responsible for the overall monitoring, management and maintenance of the system;

[0017] The data acquisition module includes displacement sensors, stress sensors, meteorological sensors and geological sensors, and the number and location of the sensors are determined according to the geometric shape and geological conditions of the foundation pit;

[0018] The multimodal data fusion module standardizes different types of data to make them have the same scale, unifies temperature, humidity, displacement and other data into the range of [0,1], integrates the standardized environmental, geological and structural data together, establishes a data association model, and mines the association relationship between different types of data;

[0019] The edge computing module performs preliminary data processing and analysis, such as data filtering, aggregation, and simple prediction, close to the data source, reducing the amount of data transmission, reducing dependence on network bandwidth, increasing data processing speed, and reducing latency.

[0020] Preferably, the detailed steps of the data acquisition module to collect multiple data are as follows:

[0021] S1. Determine the monitoring objectives and clearly define the foundation pit parameters that need to be monitored, including displacement, stress, groundwater level, temperature, humidity, wind speed, wind direction, rainfall, and geological conditions;

[0022] S2. Select sensor type: Select the appropriate sensor type based on the monitoring objective. Displacement sensors are used to monitor the horizontal and vertical displacement of the foundation pit wall. Stress sensors are used to monitor stress changes in the support structure. Water level sensors are used to monitor the groundwater level. Meteorological sensors are used to monitor temperature, humidity, wind speed, wind direction, and rainfall. Geological sensors are used to monitor soil moisture and stratum displacement.

[0023] S3. Determine the number and location of sensors. Based on the geometry, size and geological conditions of the foundation pit, determine the number of sensors and the optimal installation location to ensure full coverage of the monitoring area.

[0024] Preferably, the multimodal data fusion module integrates and analyzes multi-source data from different sensors and external data sources to provide more comprehensive and accurate monitoring results. The detailed steps are as follows:

[0025] A1. Data feature extraction: Based on the monitoring objectives, the features that have a greater impact on the foundation pit status are selected, such as displacement change rate, stress peak value, and temperature change trend;

[0026] A2. Feature Engineering: Time series feature extraction calculates the statistical characteristics of time series data, including mean, variance, trend, and period. Frequency domain feature extraction performs a Fourier transform on the signal to extract frequency domain features, including frequency components and power spectral density. Spatial feature extraction extracts spatial features from spatially distributed data, such as displacement differences at different locations in the foundation pit.

[0027] A3. Determine the fusion level. Early fusion is performed at the data level, directly combining different sensor data into a single feature vector. Mid-term fusion is performed at the feature level, combining environmental and geological features. Late fusion is performed at the model output level, taking a weighted average of the prediction results from different models.

[0028] A4. Select a fusion algorithm. Use the weighted average method to average data from different data sources, assigning weights based on the reliability or importance of the data source.

[0029] A5. Verify the data fusion results. Verify the fused data to ensure its rationality and consistency. Use cross-validation or holdout methods to verify the accuracy of the fusion results.

[0030] A6. Data smoothing and filtering: Smoothing the fused data to remove noise and unnecessary fluctuations, using filtering algorithms to smooth the data.

[0031] A7. Data storage and transmission: The fused data is stored in a database for subsequent analysis and decision-making, and the data is transmitted to the real-time data stream processing module and the early warning and alarm module.

[0032] A8. Visualize the results: Integrate the fused data with the BIM model to achieve 3D visualization. Key monitoring points are marked in the BIM model to display fused real-time data and historical trends.

[0033] A9. Decision support: Based on the integrated data and analysis results, it provides suggestions for construction plan optimization and risk control measures, provides early warning and alarm information, and guides on-site personnel to take corresponding measures.

[0034] Preferably, the edge computing module performs data processing and analysis on edge devices close to the data source, which can significantly reduce data transmission delays, reduce bandwidth consumption, and improve the real-time response capability of the system. The detailed steps are as follows:

[0035] B1. Edge device deployment and configuration: industrial-grade edge gateways, embedded systems, or sensor nodes with computing capabilities. Consider the device's processing power, storage capacity, power consumption, reliability, and environmental adaptability. The device must be waterproof, dustproof, and shockproof.

[0036] B2. Data reception: receiving raw data from sensors and data acquisition devices to ensure efficient data transmission;

[0037] B3. Data verification: Perform real-time verification of received data to check data range and integrity, and identify and mark abnormal data;

[0038] B4. Data cleaning: remove noise and outliers using filtering algorithms. Adjust the data to eliminate systematic errors based on the sensor's calibration curve.

[0039] B5. Data normalization: normalize different types of data to the same scale. For example, use minimum-to-maximum normalization to map data to the range [0, 1].

[0040] B6. Real-time data processing and analysis: Aggregate data to calculate the average, maximum, minimum, and variance over a period of time, extract features useful for pit condition monitoring, such as displacement change rate, stress peak, and temperature trend. Analyze real-time data, including pit displacement, stress, and water level parameters, and use machine learning models for real-time prediction.

[0041] B7. Edge decision-making and control: Based on real-time analysis results, local decisions are made at the edge, triggering local alarms and controlling field equipment. Decision results are sent to field equipment in the form of control instructions. Processed data and analysis results are stored locally on the edge device for subsequent query and analysis. Data is compressed to reduce data transmission volume, and key data and analysis results are transmitted to the central control system for further analysis and decision-making.

[0042] B8. The edge collaborates with the central system, synchronizes data with the central system to ensure data consistency between the edge and central systems, receives model update instructions from the central system, updates the edge's machine learning model, and receives remote monitoring and management instructions from the central system.

[0043] (3) Beneficial effects

[0044] The present invention provides a foundation pit monitoring system based on BIM. It has the following beneficial effects:

[0045] 1. In the present invention, the edge computing module performs preliminary data processing and analysis on the edge device close to the data source, and only transmits the processed key data and analysis results to the central control system, reducing the amount of data transmission and reducing the dependence on network bandwidth. Through local processing, the data transmission delay is reduced, the data processing speed is improved, and real-time response is achieved.

[0046] 2. In the present invention, the multimodal data fusion module solves the problems of data silos and intelligence, standardizes different types of data to make them have the same scale, integrates the standardized environmental, geological and structural data together, establishes a data association model, and explores the association relationship between different types of data. By comprehensively analyzing multiple data sources, the accuracy of early warning and prediction is improved.

[0047] 3. In the present invention, the real-time data stream processing module improves real-time performance and processing efficiency, realizes real-time monitoring and analysis of foundation pit status, provides timely warning and decision support, can process high-frequency data streams, ensures the efficiency and timeliness of data processing, and provides a fault-tolerant mechanism to ensure that data processing tasks can be restored when a node fails or the network is interrupted.

[0048] 4. In the present invention, the BIM model integration and visualization module improves the visualization effect and decision-making support capabilities, realizes the seamless integration of monitoring data and three-dimensional models through the BIM model, provides intuitive visualization, facilitates communication and collaboration between different departments, improves work efficiency, provides more accurate decision-making support, and optimizes construction plans and risk control measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 Schematic diagram of the system of the present invention. DETAILED DESCRIPTION

[0050] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0051] Example 1:

[0052] like Figure 1As shown, an embodiment of the present invention provides a BIM-based foundation pit monitoring system, including:

[0053] Data preprocessing module, which cleans, verifies and calibrates the collected data;

[0054] Real-time data stream processing module, which continuously processes and analyzes real-time data;

[0055] Data transmission and communication module, responsible for transmitting data from sensors to edge computing devices or central servers;

[0056] Real-time data storage module, storing real-time processed data;

[0057] Real-time analysis and prediction module, which uses machine learning and prediction models to conduct real-time analysis of foundation pit status and predict future trends;

[0058] Early warning and alarm module, triggering early warning and alarm mechanisms based on analysis results;

[0059] BIM model integration and visualization module integrates monitoring data with BIM models to achieve three-dimensional visualization;

[0060] Decision support and optimization module, providing suggestions for construction plan optimization and risk control measures;

[0061] System management and monitoring module, responsible for the overall monitoring, management and maintenance of the system;

[0062] The data acquisition module includes displacement sensors, stress sensors, meteorological sensors and geological sensors. The number and location of sensors are determined according to the geometric shape and geological conditions of the foundation pit.

[0063] The multimodal data fusion module standardizes different types of data to make them have the same scale, unifying data such as temperature, humidity, and displacement into the range of [0,1]. It integrates the standardized environmental, geological, and structural data together, establishes a data association model, and explores the association relationships between different types of data.

[0064] The edge computing module performs preliminary data processing and analysis, such as data filtering, aggregation, and simple prediction, close to the data source, reducing the amount of data transmission, reducing dependence on network bandwidth, improving data processing speed, and reducing delays. The edge computing module performs preliminary data processing and analysis on edge devices close to the data source, and only transmits processed key data and analysis results to the central control system, reducing the amount of data transmission, reducing dependence on network bandwidth, and reducing data transmission delays through local processing, improving data processing speed, and achieving real-time response.

[0065] The detailed steps for the data acquisition module to collect various data are as follows:

[0066] S1. Determine the monitoring objectives and clearly define the foundation pit parameters that need to be monitored, including displacement, stress, groundwater level, temperature, humidity, wind speed, wind direction, rainfall, and geological conditions;

[0067] S2. Select sensor type: Select the appropriate sensor type based on the monitoring objective. Displacement sensors are used to monitor the horizontal and vertical displacement of the foundation pit wall. Stress sensors are used to monitor stress changes in the support structure. Water level sensors are used to monitor the groundwater level. Meteorological sensors are used to monitor temperature, humidity, wind speed, wind direction, and rainfall. Geological sensors are used to monitor soil moisture and stratum displacement.

[0068] S3. Determine the number and location of sensors. Based on the geometry, size, and geological conditions of the foundation pit, determine the number of sensors and the optimal installation location to ensure full coverage of the monitoring area. The multimodal data fusion module solves the problems of data silos and intelligence, standardizes different types of data to make them have the same scale, integrates the standardized environmental, geological, and structural data together, establishes a data association model, and explores the correlation between different types of data. By comprehensively analyzing multiple data sources, the accuracy of early warning and prediction is improved.

[0069] The multimodal data fusion module integrates and analyzes multi-source data from different sensors and external data sources to provide more comprehensive and accurate monitoring results. The detailed steps are as follows:

[0070] A1. Data feature extraction: Based on the monitoring objectives, the features that have a greater impact on the foundation pit status are selected, such as displacement change rate, stress peak value, and temperature change trend;

[0071] A2. Feature Engineering: Time series feature extraction calculates the statistical characteristics of time series data, including mean, variance, trend, and period. Frequency domain feature extraction performs a Fourier transform on the signal to extract frequency domain features, including frequency components and power spectral density. Spatial feature extraction extracts spatial features from spatially distributed data, such as displacement differences at different locations in the foundation pit.

[0072] A3. Determine the fusion level. Early fusion is performed at the data level, directly combining different sensor data into a single feature vector. Mid-term fusion is performed at the feature level, combining environmental and geological features. Late fusion is performed at the model output level, taking a weighted average of the prediction results from different models.

[0073] A4. Select a fusion algorithm. Use the weighted average method to average data from different data sources, assigning weights based on the reliability or importance of the data source.

[0074] A5. Verify the data fusion results. Verify the fused data to ensure its rationality and consistency. Use cross-validation or holdout methods to verify the accuracy of the fusion results.

[0075] A6. Data smoothing and filtering: Smoothing the fused data to remove noise and unnecessary fluctuations, using filtering algorithms to smooth the data.

[0076] A7. Data storage and transmission: The fused data is stored in a database for subsequent analysis and decision-making, and the data is transmitted to the real-time data stream processing module and the early warning and alarm module.

[0077] A8. Visualize the results: Integrate the fused data with the BIM model to achieve 3D visualization. Key monitoring points are marked in the BIM model to display fused real-time data and historical trends.

[0078] A9. Decision support: Based on the fused data and analysis results, it provides suggestions for optimizing construction plans and risk control measures, provides early warning and alarm information, and guides on-site personnel to take corresponding measures. The multimodal data fusion module solves the problems of data silos and intelligence, standardizes different types of data to make them have the same scale, integrates standardized environmental, geological, and structural data, establishes a data association model, and explores the associations between different types of data. By comprehensively analyzing multiple data sources, it improves the accuracy of early warnings and predictions.

[0079] The edge computing module processes and analyzes data on edge devices close to the data source, which can significantly reduce data transmission delays, lower bandwidth consumption, and improve the system's real-time responsiveness. The detailed steps are as follows:

[0080] B1. Edge device deployment and configuration: industrial-grade edge gateways, embedded systems, or sensor nodes with computing capabilities. Consider the device's processing power, storage capacity, power consumption, reliability, and environmental adaptability. The device must be waterproof, dustproof, and shockproof.

[0081] B2. Data reception: receiving raw data from sensors and data acquisition devices to ensure efficient data transmission;

[0082] B3. Data verification: Perform real-time verification of received data to check data range and integrity, and identify and mark abnormal data;

[0083] B4. Data cleaning: remove noise and outliers using filtering algorithms. Adjust the data to eliminate systematic errors based on the sensor's calibration curve.

[0084] B5. Data normalization: normalize different types of data to the same scale. For example, use minimum-to-maximum normalization to map data to the range [0, 1].

[0085] B6. Real-time data processing and analysis: Aggregate data to calculate the average, maximum, minimum, and variance over a period of time, extract features useful for pit condition monitoring, such as displacement change rate, stress peak, and temperature trend. Analyze real-time data, including pit displacement, stress, and water level parameters, and use machine learning models for real-time prediction.

[0086] B7. Edge decision-making and control: Based on real-time analysis results, local decisions are made at the edge, triggering local alarms and controlling field equipment. Decision results are sent to field equipment in the form of control instructions. Processed data and analysis results are stored locally on the edge device for subsequent query and analysis. Data is compressed to reduce data transmission volume, and key data and analysis results are transmitted to the central control system for further analysis and decision-making.

[0087] B8. The edge collaborates with the central system, synchronizes data with the central system to ensure data consistency between the edge and central systems, receives model update instructions from the central system, updates the edge's machine learning model, and receives remote monitoring and management instructions from the central system.

[0088] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A BIM-based foundation pit monitoring system, characterized by: include: Data acquisition module, data preprocessing module, multimodal data fusion module, edge computing module, real-time data stream processing module, data storage and communication module, real-time data storage module, real-time analysis and prediction module, early warning and alarm module, BIM model integration and visualization module, decision support and optimization module, and system management and monitoring module; The data acquisition module includes displacement sensors, stress sensors, meteorological sensors and geological sensors, and the number and location of the sensors are determined according to the geometric shape and geological conditions of the foundation pit; The multimodal data fusion module standardizes different types of data to make them have the same scale, unifies temperature, humidity, displacement and other data into the range of [0,1], integrates the standardized environmental, geological and structural data together, establishes a data association model, and mines the association relationship between different types of data; The edge computing module performs preliminary data processing and analysis, such as data filtering, aggregation, and simple prediction, close to the data source, reducing the amount of data transmission, reducing dependence on network bandwidth, increasing data processing speed, and reducing latency.

2. The BIM-based foundation pit monitoring system according to claim 1, characterized in that: The detailed steps of the data acquisition module to collect various data are as follows: S1. Determine the monitoring objectives and clearly define the foundation pit parameters that need to be monitored, including displacement, stress, groundwater level, temperature, humidity, wind speed, wind direction, rainfall, and geological conditions; S2. Select sensor type: Select the appropriate sensor type based on the monitoring objective. Displacement sensors are used to monitor the horizontal and vertical displacement of the foundation pit wall. Stress sensors are used to monitor stress changes in the support structure. Water level sensors are used to monitor the groundwater level. Meteorological sensors are used to monitor temperature, humidity, wind speed, wind direction, and rainfall. Geological sensors are used to monitor soil moisture and stratum displacement. S3. Determine the number and location of sensors. Based on the geometry, size and geological conditions of the foundation pit, determine the number of sensors and the optimal installation location to ensure full coverage of the monitoring area.

3. The BIM-based foundation pit monitoring system according to claim 1, characterized in that: The multimodal data fusion module integrates and analyzes multi-source data from different sensors and external data sources to provide more comprehensive and accurate monitoring results. The detailed steps are as follows: A1. Data feature extraction: Based on the monitoring objectives, the features that have a greater impact on the foundation pit status are selected, such as displacement change rate, stress peak value, and temperature change trend; A2. Feature Engineering: Time series feature extraction calculates the statistical characteristics of time series data, including mean, variance, trend, and period. Frequency domain feature extraction performs a Fourier transform on the signal to extract frequency domain features, including frequency components and power spectral density. Spatial feature extraction extracts spatial features from spatially distributed data, such as displacement differences at different locations in the foundation pit. A3. Determine the fusion level. Early fusion is performed at the data level, directly combining different sensor data into a single feature vector. Mid-term fusion is performed at the feature level, combining environmental and geological features. Late fusion is performed at the model output level, taking a weighted average of the prediction results from different models. A4. Select a fusion algorithm. Use the weighted average method to average data from different data sources, assigning weights based on the reliability or importance of the data source. A5. Verify the data fusion results. Verify the fused data to ensure its rationality and consistency. Use cross-validation or holdout methods to verify the accuracy of the fusion results. A6. Data smoothing and filtering: Smoothing the fused data to remove noise and unnecessary fluctuations, using filtering algorithms to smooth the data. A7. Data storage and transmission: The fused data is stored in a database for subsequent analysis and decision-making, and the data is transmitted to the real-time data stream processing module and the early warning and alarm module. A8. Visualize the results: Integrate the fused data with the BIM model to achieve 3D visualization. Key monitoring points are marked in the BIM model to display fused real-time data and historical trends. A9. Decision support: Based on the integrated data and analysis results, it provides suggestions for construction plan optimization and risk control measures, provides early warning and alarm information, and guides on-site personnel to take corresponding measures.

4. The BIM-based foundation pit monitoring system according to claim 1, characterized in that: The edge computing module processes and analyzes data on edge devices close to the data source, which can significantly reduce data transmission delays, reduce bandwidth consumption, and improve the system's real-time responsiveness. The detailed steps are as follows: B1. Edge device deployment and configuration: industrial-grade edge gateways, embedded systems, or sensor nodes with computing capabilities. Consider the device's processing power, storage capacity, power consumption, reliability, and environmental adaptability. The device must be waterproof, dustproof, and shockproof. B2. Data reception: receiving raw data from sensors and data acquisition devices to ensure efficient data transmission; B3. Data verification: Perform real-time verification of received data to check data range and integrity, and identify and mark abnormal data; B4. Data cleaning: remove noise and outliers using filtering algorithms. Adjust the data to eliminate systematic errors based on the sensor's calibration curve. B5. Data normalization: normalize different types of data to the same scale. For example, use minimum-to-maximum normalization to map data to the range [0, 1]. B6. Real-time data processing and analysis: Aggregate data to calculate the average, maximum, minimum, and variance over a period of time, extract features useful for pit condition monitoring, such as displacement change rate, stress peak, and temperature trend. Analyze real-time data, including pit displacement, stress, and water level parameters, and use machine learning models for real-time prediction. B7. Edge decision-making and control: Based on real-time analysis results, local decisions are made at the edge, triggering local alarms and controlling field equipment. Decision results are sent to field equipment in the form of control instructions. Processed data and analysis results are stored locally on the edge device for subsequent query and analysis. Data is compressed to reduce data transmission volume, and key data and analysis results are transmitted to the central control system for further analysis and decision-making. B8. The edge collaborates with the central system, synchronizes data with the central system to ensure data consistency between the edge and central systems, receives model update instructions from the central system, updates the edge's machine learning model, and receives remote monitoring and management instructions from the central system.

Images