Foundation Collapsibility Monitoring Data Analysis Platform

Abstract

The present invention discloses a foundation collapsibility monitoring data analysis platform, which relates to the technical field of foundation monitoring and data analysis. The platform includes: a data acquisition module, a data fusion and analysis module, a risk assessment module, and a result display and update module. The present invention organically combines geological survey data, environmental monitoring data, and foundation physical parameter data through the innovative design of multi-source data fusion, realizes a comprehensive and accurate assessment of foundation collapsibility, constructs a multi-dimensional risk assessment index system with the help of advanced algorithms and dynamically updates it, timely reflects the changes in foundation collapsibility risks, and facilitates engineering personnel to intuitively understand the foundation conditions and obtain targeted risk management and control suggestions through visual display and interactive functions, effectively improves the scientificity and rationality of engineering decision-making, reduces the incidence of safety accidents caused by foundation collapsibility problems, reduces engineering construction and maintenance costs, and ensures long-term stable operation of the project.

Description

Technical Field

[0001] The present invention relates to the technical field of foundation monitoring and data analysis, and in particular to a foundation collapsible monitoring data analysis platform. Background Art

[0002] Foundation collapsibility refers to the property that when the foundation soil is soaked by water, the structure is rapidly destroyed and significant additional subsidence occurs. In many areas, especially in areas where loess is widely distributed, foundation collapsibility is a factor that must be considered in civil engineering construction, and it is crucial to monitor foundation collapsibility. In the process of engineering construction, accurate understanding of foundation collapsibility can provide a key basis for the formulation of foundation design and construction plans, and ensure the stability and safety of buildings during use. For example, in the construction of large buildings, bridges, water conservancy facilities, etc., if the foundation collapsibility is not effectively monitored, it may lead to uneven foundation settlement, cracks, tilts, or even collapse of buildings, and other serious problems, endangering people's lives and causing huge economic losses.

[0003] The traditional foundation collapsibility monitoring data platform has certain defects when used. When only simple measurement and analysis of foundation physical parameters are performed, there is a lack of consideration of geological foundations and environmental factors, and it is impossible to fully understand the causes and trends of foundation collapsibility. Empirical judgments are made based only on geological survey data, which is difficult to reflect the dynamic changes in actual projects. Only the unilateral influence of environmental factors is considered, which ignores the importance of the foundation's own characteristics. This single data source analysis and evaluation method has significant limitations and cannot accurately present the true situation and potential risks of foundation collapsibility. It is easy to cause engineering decision-making errors, which brings great safety hazards and economic losses to engineering construction and maintenance. Therefore, it is necessary to propose a foundation collapsibility monitoring data analysis platform to solve the problems in the existing technology. Summary of the invention

[0004] The purpose of the present invention is to make up for the shortcomings of the existing technology and provide a foundation collapsible monitoring data analysis platform, which can effectively solve the shortcomings of the traditional single data source analysis and evaluation method through multi-source data fusion algorithm and advanced risk assessment system, and realize comprehensive and accurate assessment and dynamic monitoring of foundation collapsible risk.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a foundation collapsible monitoring data analysis platform, which includes: a data acquisition module, a data fusion and analysis module, a risk assessment module, and a result display and update module;

[0006] The data acquisition module is used to collect geological survey data, environmental monitoring data and foundation physical parameter data, wherein the geological survey data includes stratum structure data and soil property data, the environmental monitoring data includes rainfall data and groundwater level change data, and the foundation physical parameter data includes foundation humidity, pressure, displacement and pore water pressure data;

[0007] The data fusion and analysis module includes a data integration unit, a correlation analysis unit and a feature extraction unit;

[0008] The data integration unit is used to convert various types of data collected by the data collection module into a unified format and store them;

[0009] The correlation analysis unit uses a multivariate dynamic Bayesian network algorithm to perform correlation analysis on the integrated data. The algorithm constructs a dynamic Bayesian network model including geological survey data nodes, environmental monitoring data nodes, and foundation physical parameter data nodes, and calculates the conditional probability distribution between each node at different time steps based on the Bayesian theorem and time series information to determine the internal connection and mutual influence law between different data types;

[0010] The feature extraction unit extracts feature data of great significance to the assessment of foundation collapsibility based on the correlation analysis results, and constructs a multi-dimensional feature vector; wherein the feature extraction calculation method is as follows: define For the The first data points, For the The number of data points for each class of data is calculated first. , then calculate the covariance matrix , by performing eigendecomposition on the covariance matrix, selecting the original data corresponding to the eigenvector with larger eigenvalue as the feature data, and constructing the eigenvector ,in is the number of feature data selected;

[0011] The risk assessment module includes an indicator system establishment unit and a fuzzy comprehensive evaluation unit;

[0012] The indicator system establishment unit establishes a multi-dimensional risk assessment indicator system based on the multi-dimensional feature vector, and the indicator system covers geological factors, environmental factors and foundation physical state factors;

[0013] The fuzzy comprehensive evaluation unit adopts an improved fuzzy comprehensive evaluation algorithm, which first determines the membership function of each risk assessment indicator for different risk levels and defines For the risk assessment indicators, For indicators For The membership degree of each risk level is calculated as follows: ,in and The parameters are determined based on experimental data and engineering experience, and then the weights of each indicator are determined by combining the entropy weight method and the hierarchical analysis method. For indicators The entropy value is calculated by the entropy weight method. ,in is the number of risk levels, and then the coefficient of difference is obtained , then the entropy weight , the weight determined by the hierarchical analysis method is , the final weight ,in The adjustment coefficient is determined based on expert experience. Finally, the comprehensive evaluation result of foundation collapse risk is obtained through fuzzy synthesis operation, so as to determine the collapse risk level of the foundation.

[0014] The result display and update module includes a visualization display unit and a dynamic update unit. The visualization display unit displays the risk assessment results in an intuitive visualization manner. The dynamic update unit dynamically updates the risk assessment results as new data is continuously collected and the data fusion and analysis module and the risk assessment module continue to run.

[0015] Furthermore, when collecting geological survey data, the data acquisition module adopts a combination of high-density electrical exploration technology and traditional drilling technology to obtain stratum structure data. The high-density electrical exploration technology arranges multiple electrodes on the surface, applies currents of different frequencies and intensities, measures the potential difference between different electrodes, and uses an inversion algorithm to construct a resistivity profile image of the underground stratum to infer the stratum structure. The traditional drilling technology performs drilling and sampling at key locations to obtain actual physical samples of the soil layer and measure soil property data. The soil property data includes the particle size distribution, porosity, compression modulus and natural water content of the soil.

[0016] Furthermore, in the environmental monitoring data collection process, for the collection of rainfall data, a distributed network consisting of multiple high-precision rainfall sensors is used. The rainfall sensors are distributed in different terrains and locations within the monitoring area. After the sensors collect the rainfall data, preliminary data processing is performed locally to remove abnormal data points caused by wind and vibration factors, and the data is sent to the data collection module through the wireless transmission module. For the collection of groundwater level change data, a high-precision pressure water level sensor is used and installed in the groundwater level monitoring well to monitor the slight changes in the water level in real time, and the water level depth data is converted into an electrical signal, which is amplified and filtered by the signal conditioning circuit and then transmitted to the data collection module.

[0017] Furthermore, the data integration unit of the data fusion and analysis module adopts a distributed database storage architecture when storing data, and stores different types of data in different database nodes respectively. The nodes are connected through a high-speed network. When constructing a multivariate dynamic Bayesian network model, the correlation analysis unit subdivides the geological survey data nodes according to the different levels and attributes of the stratigraphic structure data and soil property data, and subdivides different soil layer thickness nodes and soil layer type nodes as stratigraphic structure data nodes, and soil porosity nodes and compression modulus nodes as soil property data nodes.

[0018] Furthermore, the correlation analysis unit uses a multivariate dynamic Bayesian network algorithm to perform correlation analysis on the integrated data. The specific calculation process is as follows: represents a set of geological survey data nodes, represents the set of environmental monitoring data nodes, Represents the set of nodes for the data of the foundation physical parameters, at the time step , for the node , its conditional probability distribution Calculated as: ,in For Node The parent node set of are the relevant parameters, is the number of parent nodes, For Node The number of possible values ​​can be calculated in the same way as the conditional probability distribution of environmental monitoring data nodes and ground-based physical parameter data nodes.

[0019] Furthermore, when constructing the feature vector, the feature extraction unit, in addition to selecting feature data according to the covariance matrix feature decomposition, introduces the time series characteristics of the data as an auxiliary screening condition, and calculates the trend slope and periodic characteristic parameters of the data with obvious time trend, and defines For the In the data of The time series eigenvalues ​​of the data points are integrated into the eigenvector by weighted summation ,in is the weight coefficient determined according to the importance of the data, is the number of data categories considering time series characteristics.

[0020] Furthermore, when determining the membership function of each risk assessment indicator for different risk levels, the fuzzy comprehensive evaluation unit adopts a method combining expert scoring and fuzzy clustering to determine the membership for indicators that are difficult to quantify directly. The scoring data is used as the input of the fuzzy clustering algorithm. The geological structure is divided into three categories: simple, medium, and complex through the clustering algorithm. The simple structure corresponds to a low risk level, with a membership range of 0.8-1.0, the medium structure corresponds to a medium risk level, with a membership range of 0.5-0.8, and the complex structure corresponds to a high risk level, with a membership range of 0-0.5. When displaying the risk assessment results, the visualization display unit of the result display and update module constructs a three-dimensional foundation model and displays the collapsible risk levels of different locations on the model with different colors and marks, providing an interactive risk detail viewing function.

[0021] Furthermore, the dynamic update unit adopts an adaptive time interval adjustment algorithm when dynamically updating the risk assessment results. The algorithm adjusts the update period according to the data change rate and the stability of the risk level. The stability index of risk level is obtained by calculating the slope of data within a certain period of time. The update cycle is obtained by calculating the variance of several consecutive risk assessment results. ,in and It is a coefficient determined according to engineering requirements and data characteristics.

[0022] Compared with the existing technology, the foundation collapsibility monitoring data analysis platform has the following beneficial effects:

[0023] The present invention organically combines geological survey data, environmental monitoring data and foundation physical parameter data through the innovative design of multi-source data fusion, so as to realize a comprehensive and accurate assessment of foundation collapsibility. With the help of advanced algorithms, a multi-dimensional risk assessment index system is constructed and dynamically updated to timely reflect changes in foundation collapsibility risks. Through visual display and interactive functions, it is convenient for engineering personnel to intuitively understand the foundation conditions and obtain targeted risk management and control suggestions, effectively improving the scientificity and rationality of engineering decision-making, reducing the incidence of safety accidents caused by foundation collapsibility problems, reducing engineering construction and maintenance costs, and ensuring the long-term stable operation of the project.

[0024] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0026] Figure 1 This is a schematic diagram of the structure of the foundation collapsible monitoring data analysis platform. DETAILED DESCRIPTION

[0027] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] Embodiment 1

[0029] In a large commercial center construction project in a certain city, the foundation collapsibility monitoring data analysis platform plays a key role. This commercial center covers a large area and the geological conditions in the area are relatively complex. The groundwater level fluctuates significantly due to seasonal influences and the surrounding environmental factors are changeable. Therefore, it is extremely important to accurately monitor and evaluate the foundation collapsibility.

[0030] First, the data acquisition module starts working, acquiring geological survey data by combining high-density electrical exploration technology with traditional drilling technology. High-density electrical exploration arranges electrodes at a certain interval within the commercial center, applies currents with different frequencies of 10-100Hz and 1-10A intensities, measures the potential difference between the electrodes, and constructs a stratum resistivity profile image through an inversion algorithm, clearly presenting the distribution of each soil layer, such as the thickness range of different soil layers such as clay and sand. Traditional drilling technology conducts drilling and sampling at key supporting parts of buildings and areas with suspected complex geological structures, and accurately determines soil property data, including soil particle size distribution between 0.002-2mm, porosity of 0.5-1.2, compression modulus of 5-20MPa, natural water content of 10%-30%, etc.

[0031] At the same time, environmental monitoring data collection is also carried out synchronously. A distributed network consisting of 10 high-precision rainfall sensors evenly distributed in the commercial center and surrounding areas collects rainfall data in real time. The sensor has an automatic calibration function, which can control the data deviation caused by wind, vibration and other factors within a very small range. The collected data is preliminarily processed locally, and after removing abnormal data points, the data is sent to the data acquisition module through the wireless transmission module. In addition, the high-precision pressure water level sensor installed in the groundwater level monitoring well monitors the groundwater level changes in real time. It can accurately monitor the water level depth change to a small change of 0.01 meters, convert the water level depth data into an electrical signal, and transmit it to the data acquisition module after amplification and filtering by the signal conditioning circuit. The foundation physical parameter data is collected by the sensor array installed at different depths of the foundation (such as setting a layer of sensors every 2-5 meters) and key locations (such as building corners, column foundations, etc.), including foundation humidity changes between 15%-40%, pressure range of 0-500kPa, displacement data with a displacement accuracy of 0.1mm, and pore water pressure data.

[0032] After data collection is completed, it enters the data fusion and analysis module. The data integration unit converts various types of data into a unified format and stores them in a distributed database storage architecture. The geological survey data, environmental monitoring data, and foundation physical parameter data are stored in different database nodes. The nodes are connected by a high-speed network. Based on the relevance and frequency of use of the data, the intelligent caching mechanism caches the recently frequently used data into the memory. At the same time, the distributed ledger based on blockchain technology records the data update history and operation records to ensure the integrity and security of the data.

[0033] The correlation analysis unit constructs a multivariate dynamic Bayesian network model. The geological survey data nodes are subdivided into different soil layer thickness nodes (such as clay thickness nodes, sand thickness nodes, etc.), soil layer type nodes, etc. The soil property data nodes are divided into soil porosity nodes, compression modulus nodes, etc. The environmental monitoring data nodes include rainfall data nodes, groundwater level change data nodes, rainfall intensity change rate nodes (obtained by calculating the ratio of the rainfall difference in adjacent time periods to the time interval), groundwater level fluctuation frequency nodes (counting the number of water level fluctuations per unit time), etc. The foundation physical parameter data nodes are hierarchically set up with humidity nodes, pressure nodes, displacement nodes, and pore water pressure nodes according to the foundation geographical location and depth information. Then, based on the Bayesian theorem and time series information, the conditional probability distribution between each node at different time steps is calculated. For example, for a certain soil layer thickness node , its conditional probability distribution According to the formula ,in For Node The parent node set of are the relevant parameters, is the number of parent nodes, For Node The number of possible values ​​is calculated, and the conditional probability distribution of other nodes is calculated similarly, so as to determine the internal connection and mutual influence between different data types.

[0034] The feature extraction unit extracts feature data that are of great significance to the assessment of foundation collapsibility based on the correlation analysis results. For the The first data points, For the The number of data points in the data, first calculate the mean of each type of data , then calculate the covariance matrix By performing eigendecomposition on the covariance matrix, the original data corresponding to the eigenvector with larger eigenvalue is selected as the feature data to construct the eigenvector ,in is the number of characteristic data selected. At the same time, for data with obvious time trends, such as the curve of foundation humidity changing over time, its trend slope, periodic characteristics and other parameters are calculated. , and integrate it into the feature vector by weighted summation ,in is the weight coefficient determined according to the importance of the data, is the number of data categories considering time series characteristics.

[0035] Next, the risk assessment module works. The indicator system establishment unit establishes a multi-dimensional risk assessment indicator system based on the multi-dimensional characteristic vector. In addition to the conventional stratum stability indicators (such as calculating the safety factor based on the shear strength of the soil layer) and soil sensitivity indicators (such as determining the soil collapsibility coefficient), the geological structure complexity indicators are added according to the geological structure characteristics of the region. The environmental factor indicators include extreme rainfall indicators (such as the single maximum rainfall of 200mm in the past month, the cumulative rainfall of 300mm for three consecutive days, etc.) and abnormal groundwater level fluctuation indicators (such as the number of times the water level has risen and fallen by more than 2 meters three times in the past week).

[0036] The indicators of foundation physical state factors include the degree of excess of foundation humidity (such as the current foundation humidity exceeds the safety humidity threshold of 2596 by 2096, and the duration is 5 days), the unevenness of pressure distribution (such as the ratio of the pressure difference in different areas of the foundation to the average value is 3096), and the abnormal change of displacement (such as the number of times the displacement mutation amplitude exceeds the warning value of 5mm is 2 times, and the duration of continuous increase in displacement exceeding the warning value of 10mm is 3 days).

[0037] The fuzzy comprehensive evaluation unit adopts an improved fuzzy comprehensive evaluation algorithm. First, the membership function of each risk assessment index for different risk levels is determined. For the risk assessment indicators, For indicators For The membership degree of each risk level is calculated as follows: ,in and The parameters are determined based on experimental data and engineering experience, and then the weights of each index are determined by combining the entropy weight method and the hierarchical analysis method. For indicators The entropy value is calculated by the entropy weight method. ,in is the number of risk levels, and then the coefficient of difference is obtained , then the entropy weight , the weight determined by the hierarchical analysis method is , the final weight ,in The adjustment coefficient is determined based on expert experience. Finally, the comprehensive evaluation result of the foundation collapse risk is obtained through fuzzy synthesis operation, and the collapse risk level of the foundation is determined to be medium risk. Finally, the result display and update module comes into play.

[0038] The visualization display unit constructs a three-dimensional foundation model, displays medium-risk areas with yellow marks on the model, and provides an interactive risk detail viewing function. When the user clicks on a specific area, a detailed risk information window pops up, showing the values ​​of various risk assessment indicators in the area, the historical curve of risk level changes, related data charts, and corresponding risk management suggestions. For example, it is recommended to increase foundation reinforcement measures in medium-risk areas, such as using deep mixing piles to reinforce the foundation, and at the same time strengthen the construction of drainage facilities to prevent the groundwater level from further rising and causing adverse effects on the foundation. The dynamic update unit adjusts the algorithm to run continuously according to the adaptive time interval, and the data change rate The stability index of risk level is obtained by calculating the slope of data within a certain period of time. The update cycle is obtained by calculating the variance of several consecutive risk assessment results. ,in The coefficient is determined according to engineering requirements and data characteristics. When the data changes drastically, such as a sudden increase in rainfall or a rapid rise in groundwater level, the update cycle is shortened to 1 hour to timely capture changes in the risk of foundation subsidence. When the data changes slowly and the risk level is stable, the update cycle is extended to 6 hours to reduce unnecessary consumption of computing resources.

[0039] Effects of this embodiment: This embodiment fully integrates multi-source data through the foundation collapsibility monitoring data analysis platform, overcomes the limitations of traditional single data source monitoring and analysis, and uses advanced algorithms to accurately calculate the correlation and risk assessment indicators between various data, making the foundation collapsibility risk assessment results more accurate and reliable. Visual display and interactive functions provide engineering personnel with intuitive and detailed risk information and management suggestions, which helps to take effective response measures in a timely manner, ensure the stability and safety of the foundation of large-scale commercial center construction projects, reduce the risks of project delays and structural damage that may be caused by foundation collapsibility problems, reduce project construction and maintenance costs, and improve the quality and benefits of the entire project.

[0040] Embodiment 2

[0041] In a railway bridge construction project, since railway bridges have extremely high requirements for foundation stability and the geological conditions along the line are complex and changeable, they may pass through a variety of different stratigraphic structure areas. At the same time, they face large differences in rainfall in different seasons and factors such as surrounding water conservancy facilities that affect groundwater level changes. Therefore, the local foundation collapsibility monitoring data analysis platform plays an indispensable role.

[0042] After the data acquisition module is started, high-density electrical exploration electrodes are arranged along the bridge at a density of one exploration point every 50 meters based on the geological survey data, and a 5-80Hz frequency and 0.5-8A current strength are applied to obtain the formation resistivity profile to determine the formation structure, such as the presence of a sandy soil layer with a thickness of 3-8 meters and a clay layer of 10-15 meters. Drilling and sampling are carried out at key pier locations to determine soil property data, with particle size distribution between 0.005-3mm, porosity ratio of 0.4-1.0, compression modulus of 8-25MPa, and natural water content of 8%-28%.

[0043] In terms of environmental monitoring data, a high-precision rainfall sensor station is set up every 2 kilometers along the railway bridge line to form a distributed network to collect rainfall data. These sensors have powerful data error correction capabilities and can effectively remove abnormal data caused by environmental interference. After collection, the data is transmitted wirelessly to the data acquisition module. The groundwater level change data is obtained by installing high-precision pressure water level sensors in groundwater observation wells near the bridge. It can accurately sense tiny changes in water levels of 0.005 meters, convert them into electrical signals and transmit them. The foundation physical parameter data is collected by sensor arrays installed at different depths (3-6 meters apart) on the pier foundation and surrounding foundations, including foundation humidity fluctuations of 12%-35%, pressure range of 0-600kPa, displacement data with a displacement accuracy of 0.05mm, and pore water pressure data.

[0044] In the data fusion and analysis module, the data integration unit organizes the data into a unified format and stores it in a distributed database. Various types of data are stored in different nodes and connected through a high-speed network. An intelligent caching mechanism is used to improve data access efficiency, while blockchain technology is used to ensure data consistency and security.

[0045] The correlation analysis unit constructs a multivariate dynamic Bayesian network model, subdividing geological survey data nodes into nodes of different soil layer thicknesses (such as sandy soil layer thickness nodes, clay layer thickness nodes), soil layer types, etc. The soil property data nodes are porosity nodes, compression modulus nodes, etc. The environmental monitoring data nodes include rainfall data nodes, groundwater level change data nodes, rainfall intensity change rate nodes (ratio of rainfall difference in adjacent time periods to time intervals), groundwater level fluctuation frequency nodes (number of water level fluctuations per unit time), and foundation physical parameter data nodes are set according to location and depth. Humidity nodes, pressure nodes, displacement nodes, and pore water pressure nodes, etc., calculate the conditional probability distribution of each node according to the Bayesian theorem and time series information. For example, for a clay layer thickness node , at time step , its conditional probability distribution According to the formula Calculate and determine the internal connections and influence patterns between different data.

[0046] The feature extraction unit extracts feature data based on the correlation results. For the The first data points, For the The number of data points for each class of data is calculated first. , then calculate the covariance matrix , after eigendecomposition, the original data corresponding to the eigenvector with larger eigenvalue is selected as the feature data to construct the feature vector For data with time trends, such as the ground humidity change curve, calculate its trend slope, periodic characteristics and other parameters , weighted into the feature vector , The weight coefficient is determined according to the importance of the data. is the number of data categories considering time series characteristics.

[0047] The indicator system establishment unit of the risk assessment module builds an indicator system based on the characteristic vector. In addition to the stratum stability index (calculated based on the shear strength of the soil layer) and the soil sensitivity index (determining the soil collapsibility coefficient), for the geological structure, if there is a fault structure, according to the number of faults , fault drop In 1-3 meters, fault length 20-50 meters and the thickness of the strata in the area In the range of 8-20 meters, according to the formula The geological structure complexity index is calculated, and the environmental factor indicators include the extreme rainfall index (the maximum single rainfall in the past half month is 180mm, and the cumulative rainfall for three consecutive days is 250mm), the abnormal groundwater level fluctuation index (the water level has risen and fallen by more than 1.5 meters 4 times in the past ten days), and the foundation physical state factor indicators include the foundation humidity exceedance degree index (the current humidity exceeds the safety threshold of 22% by 15%, and the duration is 3 days), the pressure distribution unevenness index (the ratio of the pressure difference in different areas to the average value is 2596), and the displacement abnormal change index (the displacement suddenly exceeded the warning value by 4mm 3 times, and the displacement continued to increase and exceeded the warning value by 8mm for 2 days).

[0048] The fuzzy comprehensive evaluation unit uses an improved algorithm to determine the membership function of each indicator. , and Determined based on experience, the weights are determined by combining the entropy weight method with the hierarchical analysis method, and the entropy value is calculated by the entropy weight method ,in is the number of risk levels, and then the coefficient of difference is obtained , then the entropy weight , the weight determined by the hierarchical analysis method is , the final weight , , the fuzzy synthesis operation determines that the foundation collapse risk level is medium risk. In the result display and update module, the visualization display unit constructs a three-dimensional bridge foundation model, marks the medium risk area in yellow and provides interactive functions. Click the area to view the risk details, including indicator values, change curves, data charts and management suggestions, such as using grouting to reinforce the foundation around the medium risk pier foundation, adding drainage pipes to reduce the impact of groundwater levels, etc. The dynamic update unit adjusts the algorithm operation according to the adaptive time interval, and the data change rate Stability index with risk level According to the formula, update cycle , , When data changes drastically, the update cycle is shortened to 40 minutes, and the stable delay is up to 4 hours.

[0049] Effects brought about by this embodiment: This embodiment uses the foundation collapsibility monitoring data analysis platform in the railway bridge construction scenario, fully integrates multi-source data, accurately analyzes the relationship between data, and provides a scientific basis for accurate risk assessment to ensure the stability of the railway bridge foundation, effectively avoiding safety hazards such as deformation and settlement of the bridge structure caused by foundation collapsibility. The visualization and dynamic update functions help engineering personnel to grasp the foundation status in a timely manner and take targeted measures, improve the efficiency and quality of project management, reduce the subsequent maintenance costs and potential risks, ensure the safe operation of railway bridges under complex geological and environmental conditions, and ensure the efficiency and safety of railway transportation.

[0050] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the same elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. The foundation collapsibility monitoring data analysis platform is characterized by: The platform includes: data acquisition module, data fusion and analysis module, risk assessment module and result display and update module; The data acquisition module is used to collect geological survey data, environmental monitoring data and foundation physical parameter data, wherein the geological survey data includes stratum structure data and soil property data, the environmental monitoring data includes rainfall data and groundwater level change data, and the foundation physical parameter data includes foundation humidity, pressure, displacement and pore water pressure data; The data fusion and analysis module includes a data integration unit, a correlation analysis unit and a feature extraction unit; The data integration unit is used to convert various types of data collected by the data collection module into a unified format and store them; The correlation analysis unit uses a multivariate dynamic Bayesian network algorithm to perform correlation analysis on the integrated data. The algorithm constructs a dynamic Bayesian network model including geological survey data nodes, environmental monitoring data nodes, and foundation physical parameter data nodes, and calculates the conditional probability distribution between each node at different time steps based on the Bayesian theorem and time series information to determine the internal connection and mutual influence law between different data types; The feature extraction unit extracts feature data of great significance to the assessment of foundation collapsibility based on the correlation analysis results, and constructs a multi-dimensional feature vector; wherein the feature extraction calculation method is as follows: define For the The first data points, For the The number of data points for each class of data is calculated first. , then calculate the covariance matrix , by performing eigendecomposition on the covariance matrix, selecting the original data corresponding to the eigenvector with larger eigenvalue as the feature data, and constructing the eigenvector ,in is the number of feature data selected; The risk assessment module includes an indicator system establishment unit and a fuzzy comprehensive evaluation unit; The indicator system establishment unit establishes a multi-dimensional risk assessment indicator system based on the multi-dimensional feature vector, and the indicator system covers geological factors, environmental factors and foundation physical state factors; The fuzzy comprehensive evaluation unit adopts an improved fuzzy comprehensive evaluation algorithm, which first determines the membership function of each risk assessment indicator for different risk levels and defines For the risk assessment indicators, For indicators For The membership degree of each risk level is calculated as follows: ,in and The parameters are determined based on experimental data and engineering experience, and then the weights of each indicator are determined by combining the entropy weight method and the hierarchical analysis method. For indicators The entropy value is calculated by entropy weight method ,in is the number of risk levels, and then the coefficient of difference is obtained , then the entropy weight , the weight determined by the hierarchical analysis method is , the final weight ,in The adjustment coefficient is determined based on expert experience. Finally, the comprehensive evaluation result of foundation collapse risk is obtained through fuzzy synthesis operation, so as to determine the collapse risk level of the foundation. The result display and update module includes a visualization display unit and a dynamic update unit. The visualization display unit displays the risk assessment results in an intuitive visualization manner. The dynamic update unit dynamically updates the risk assessment results as new data is continuously collected and the data fusion and analysis module and the risk assessment module continue to run.

2. The foundation collapsibility monitoring data analysis platform according to claim 1, characterized in that: When collecting geological survey data, the data acquisition module adopts a combination of high-density electrical exploration technology and traditional drilling technology to obtain stratum structure data. The high-density electrical exploration technology arranges multiple electrodes on the surface, applies currents of different frequencies and intensities, measures the potential difference between different electrodes, and uses an inversion algorithm to construct a resistivity profile image of the underground stratum to infer the stratum structure. The traditional drilling technology performs drilling and sampling at key locations to obtain actual physical samples of the soil layer and measure soil property data. The soil property data includes soil particle size distribution, porosity, compression modulus and natural water content.

3. The foundation collapsibility monitoring data analysis platform according to claim 1, characterized in that: In the environmental monitoring data collection process, for the collection of rainfall data, a distributed network composed of multiple high-precision rainfall sensors is used. The rainfall sensors are distributed in different terrains and locations within the monitoring area. After the sensors collect the rainfall data, preliminary data processing is performed locally to remove abnormal data points caused by wind and vibration factors, and the data is sent to the data collection module through the wireless transmission module. For the collection of groundwater level change data, a high-precision pressure water level sensor is used and installed in the groundwater level monitoring well to monitor the slight changes in the water level in real time, and the water level depth data is converted into an electrical signal, which is amplified and filtered by the signal conditioning circuit and then transmitted to the data collection module.

4. The foundation collapsibility monitoring data analysis platform according to claim 1, characterized in that: The data integration unit of the data fusion and analysis module adopts a distributed database storage architecture when storing data, and stores different types of data in different database nodes respectively. The nodes are connected through a high-speed network. When constructing a multivariate dynamic Bayesian network model, the correlation analysis unit subdivides the geological survey data nodes according to the different levels and attributes of the stratum structure data and the soil property data, and subdivides different soil layer thickness nodes and soil layer type nodes as stratum structure data nodes, and soil porosity nodes and compression modulus nodes as soil property data nodes.

5. The foundation collapsibility monitoring data analysis platform according to claim 1, characterized in that: The specific calculation process of the correlation analysis unit for performing correlation analysis on the integrated data using a multivariate dynamic Bayesian network algorithm is as follows: represents a set of geological survey data nodes, represents the set of environmental monitoring data nodes, Represents the set of nodes for the data of the foundation physical parameters, at the time step , for the node , its conditional probability distribution Calculated as: ,in For Node The parent node set of are the relevant parameters, is the number of parent nodes, For Node The number of possible values ​​can be calculated in the same way as the conditional probability distribution of environmental monitoring data nodes and ground-based physical parameter data nodes.

6. The foundation collapsibility monitoring data analysis platform according to claim 1, characterized in that: When constructing the feature vector, the feature extraction unit, in addition to selecting feature data based on the covariance matrix feature decomposition, introduces the time series characteristics of the data as an auxiliary screening condition. For data with obvious time trends, the trend slope and periodic feature parameters are calculated to define For the In the data of The time series eigenvalues ​​of the data points are integrated into the eigenvector by weighted summation ,in is the weight coefficient determined according to the importance of the data, is the number of data categories considering time series characteristics.

7. The foundation collapsibility monitoring data analysis platform according to claim 1, characterized in that: When determining the membership function of each risk assessment indicator for different risk levels, the fuzzy comprehensive evaluation unit adopts a method combining expert scoring and fuzzy clustering to determine the membership for indicators that are difficult to quantify directly, and uses the scoring data as the input of the fuzzy clustering algorithm. The geological structure is divided into three categories: simple, medium, and complex through the clustering algorithm. The simple structure corresponds to a low risk level, with a membership range of 0.8-1.0, the medium structure corresponds to a medium risk level, with a membership range of 0.5-0.8, and the complex structure corresponds to a high risk level, with a membership range of 0-0.

5. When displaying the risk assessment results, the visualization display unit of the result display and update module constructs a three-dimensional foundation model and displays the collapsible risk levels of different locations on the model with different colors and marks, providing an interactive risk detail viewing function.

8. The foundation collapsibility monitoring data analysis platform according to claim 1, characterized in that: The dynamic update unit adopts an adaptive time interval adjustment algorithm when dynamically updating the risk assessment results. The algorithm adjusts the update period according to the data change rate and the stability of the risk level. The stability index of risk level is obtained by calculating the slope of data within a certain period of time. The update cycle is obtained by calculating the variance of several consecutive risk assessment results. ,in and It is a coefficient determined according to engineering requirements and data characteristics.

Images