A geotechnical material quality evaluation method and system based on big data

Abstract

The application relates to the technical field of quality evaluation, and particularly discloses a geotechnical material quality evaluation method and system based on big data, which comprises the following steps: in a proposed application engineering environment, applying an initial mechanical index of geotechnical material to abnormal characteristic parameters generated by environmental influence factors, and instantiating the abnormal characteristic parameters as abnormal event nodes of the geotechnical material; instantiating the environmental influence factors as environmental factor nodes, and establishing a dynamic abnormal correlation network of time evolution of the geotechnical material according to spatial adjacency relationships and time sequences among the abnormal event nodes; and deducing continuous quality risks of the geotechnical material according to the dynamic abnormal correlation network, and generating a comprehensive quality evaluation result of the geotechnical material. The application can improve the continuity of geotechnical material quality evaluation and the accuracy of risk deduction.

Description

Technical Field

[0001] This invention relates to the field of quality assessment technology, and in particular to a method and system for assessing the quality of geosynthetic materials based on big data. Background Technology

[0002] The quality assessment of geotechnical materials often relies on static and discrete testing methods, which make it difficult to continuously characterize the dynamic evolution of material properties under the complex spatiotemporal conditions of the intended engineering environment. Traditional methods often analyze mechanical properties in isolation at a single moment or in a local area, ignoring the continuous interaction between environmental influencing factors and the initial mechanical properties of the material. This results in lagging and one-sided assessment results, failing to capture subtle anomalies in material properties and their development trends in a timely manner.

[0003] Furthermore, existing technologies lack effective modeling methods for the spatiotemporal correlations between abnormal events when processing large-scale monitoring data, making it difficult to quantify the spatial propagation patterns and temporal cumulative effects of abnormal characteristics. This fragmented analysis approach keeps quality risk assessment at a qualitative or semi-quantitative level, failing to accurately predict continuous quality risk trajectories. This results in insufficient accuracy in the comprehensive quality assessment of geosynthetic materials throughout their entire lifecycle, making it difficult to meet the refined management requirements of major projects for the safety and durability of materials during service. Therefore, how to overcome the limitations of traditional static and discrete assessment models, achieve continuous and accurate characterization of the dynamic evolution of geosynthetic material performance under the complex spatiotemporal conditions of the intended engineering environment, and how to effectively model the spatiotemporal correlations between abnormal events to predict continuous quality risk trajectories have become urgent technical challenges to be solved in this field. Summary of the Invention

[0004] In view of the above problems, the purpose of this invention is to provide a method and system for geosynthetic material quality assessment based on big data, so as to solve the problems of static and fragmented assessment models, as well as the lagging, one-sided and inaccurate quality risk assessment in existing geosynthetic material quality assessment technologies.

[0005] This invention provides a method for assessing the quality of geosynthetic materials based on big data, comprising:

[0006] Step 1: In the proposed engineering environment, the initial mechanical properties of the geotextile material are applied to the abnormal characteristic parameters generated by the environmental impact factors, and these parameters are instantiated as abnormal event nodes of the geotextile material.

[0007] Step 2: Instantiate the environmental impact factors as environmental factor nodes, and establish a dynamic anomaly association network for the evolution of the geotechnical material over time based on the spatial adjacency relationship and temporal sequence between each abnormal event node.

[0008] Step 3: Based on the dynamic anomaly correlation network, deduce the continuous quality risk of the geosynthetic material and generate the comprehensive quality assessment result of the geosynthetic material.

[0009] Preferably, in the proposed application engineering environment, the abnormal characteristic parameters generated by applying the initial mechanical properties of the geosynthetic material to environmental impact factors are instantiated as abnormal event nodes of the geosynthetic material, and the process is as follows:

[0010] In the proposed engineering environment, a dynamic benchmark deviation field for the geosynthetic material is constructed based on the correlation logic between the initial mechanical properties of the geosynthetic material and environmental impact factors.

[0011] Based on the dynamic benchmark deviation field, the degree of dispersion of the geomaterial between the actual performance degradation and the theoretical performance degradation curve is quantified, and the abnormal characteristic parameters of the geomaterial are generated.

[0012] Based on the time node and spatial coordinates of the aforementioned abnormal feature parameters, the spatial coordinates under each time node are instantiated into an abnormal event node of the geotechnical material.

[0013] Preferably, the process of constructing the dynamic benchmark deviation field of the geosynthetic material based on the correlation logic of the initial mechanical properties of the geosynthetic material acting on environmental impact factors is as follows:

[0014] The initial mechanical indices, aligned according to time sequence, are combined with the environmental impact factors to form the environmental-mechanical data pair for the geotechnical material.

[0015] The environmental-mechanical data pairs are projected onto a coordinate system with environmental impact factors on the horizontal axis and initial mechanical indices on the vertical axis to form a data scatter point, which is then fitted into a smooth curve. The change relationship represented by the smooth curve is then determined as the baseline attenuation mapping relationship.

[0016] Substituting the spatiotemporal distribution coordinates of the environmental impact factors into the benchmark attenuation mapping relationship, the dynamic benchmark deviation field of the geomaterial under no-abnormal conditions is extrapolated.

[0017] Preferably, the process of instantiating the environmental impact factors into environmental factor nodes and establishing a dynamic anomaly association network of the geotechnical material's evolution over time based on the spatial adjacency and temporal sequence between each anomalous event node is as follows:

[0018] Based on the spatial coordinates of the abnormal event nodes, pairs of abnormal event nodes whose spatial distance between each other falls within a preset adjacency interval are identified as spatial adjacency pairs.

[0019] Draw the first type of directed edges from the first node to the second node in chronological order;

[0020] Each of the aforementioned abnormal event nodes is connected to spatiotemporally matched environmental factor nodes through a second type of undirected edge, and a time label is attached to each edge to obtain the dynamic abnormal association network of the geotechnical material.

[0021] Preferably, the formula for calculating the spatial distance between any two of the abnormal event nodes is as follows:

[0022] ;

[0023] In the formula, Indicates the spatial distance. This represents the spatial coordinates of one of the abnormal event nodes in the abnormal event node pair. This represents the spatial coordinates of another abnormal event node in the abnormal event node pair. This represents the preset abnormal difference adjustment coefficient. Representation of spatial coordinates The abnormal characteristic parameter values ​​of the corresponding abnormal event nodes, Representation of spatial coordinates The abnormal characteristic parameter values ​​of the corresponding abnormal event node.

[0024] Preferably, the process of connecting each of the anomalous event nodes with spatiotemporally matched environmental factor nodes through a second type of undirected edge, and attaching a time label to each edge to obtain the dynamic anomaly association network of the geotechnical material, is as follows:

[0025] Based on the timestamp and spatial coordinates of the abnormal event node, the environmental factor nodes carrying the same timestamp and spatial coordinates are identified as the spatiotemporal matching nodes of the abnormal event node.

[0026] Between each pair of abnormal event nodes and their corresponding spatiotemporal matching nodes, draw a second-type undirected edge and assign a time label to each second-type undirected edge;

[0027] Arrange the first type of directed edges and the second type of undirected edges according to the order of the time tags to form the dynamic anomaly association network of the geotechnical material.

[0028] Preferably, the process of drawing a second-type undirected edge between each pair of abnormal event nodes and their corresponding spatiotemporal matching nodes, and assigning a time label to each second-type undirected edge, is as follows:

[0029] Using the earlier time among the abnormal event node and the environmental factor node as the start time and the later time as the end time, the start time and the end time are jointly encapsulated into a time label for the second type of undirected edge;

[0030] Create a time tag entry in the attribute field of the second type of undirected edge, and write the time tag into the time tag entry.

[0031] Preferably, the process of inferring the continuous quality risk of the geosynthetic material based on the dynamic anomaly correlation network and generating a comprehensive quality assessment result for the geosynthetic material is as follows:

[0032] Along the propagation direction indicated by the first type of directed edge in the dynamic anomaly association network, the change in the abnormal feature parameter value between the start and end points of each first type of directed edge is paired with the time label of the corresponding first type of directed edge to obtain the abnormal change amount-time data pair of the geotechnical material.

[0033] Arrange the abnormal change-time data pairs in chronological order to form the dynamic risk curve of the geosynthetic material;

[0034] In the dynamic risk curve, the cumulative value of abnormal changes within different time windows is extracted, and the matching result between the cumulative value of abnormal changes and the preset quality risk level interval is integrated into the comprehensive quality assessment result of the geotechnical material.

[0035] Preferably, the process of integrating the cumulative value of the abnormal changes with the preset quality risk level range into the comprehensive quality assessment result of the geosynthetic material is as follows:

[0036] The cumulative value of abnormal changes in each time window is sequentially mapped to a preset quality risk level range for comparison;

[0037] If the cumulative value of the abnormal change falls into the low-risk range, the time window is marked as a low-risk window, and so on, to obtain the low-risk window, medium-risk window and high-risk window of the time window.

[0038] The time windows carrying risk markers are arranged in chronological order to generate a time series consisting of low-risk, medium-risk, and high-risk periods distributed sequentially or alternately, and the time series is used as the comprehensive quality assessment result of the geotechnical material.

[0039] This invention also provides a geosynthetic material quality assessment system based on big data, the system comprising:

[0040] An anomaly node generation module is used to instantiate the abnormal characteristic parameters generated by the initial mechanical properties of the geomaterial acting on environmental impact factors in the proposed application engineering environment into anomaly event nodes of the geomaterial.

[0041] The dynamic network modeling module is used to instantiate the environmental impact factors into environmental factor nodes, and establish a dynamic anomaly association network of the geotechnical material over time based on the spatial adjacency relationship and temporal sequence between each abnormal event node.

[0042] The quality risk assessment module is used to deduce the continuous quality risk of the geomaterial based on the dynamic anomaly correlation network and generate a comprehensive quality assessment result of the geomaterial.

[0043] As can be seen from the above technical solution, the present invention provides a method and system for geotechnical material quality assessment based on big data. By constructing a dynamic anomaly correlation network, the initial mechanical indicators of geotechnical materials are deeply integrated with environmental impact factors, realizing the transformation from traditional static and discrete detection to continuous and dynamic assessment. This method can quantify the spatiotemporal adjacency relationship and propagation direction between abnormal event nodes, accurately predict the continuous quality risk evolution trajectory of geotechnical materials over time, thereby significantly improving the continuity of geotechnical material quality assessment and the accuracy of risk prediction, and providing a more reliable and refined basis for the service safety of materials in engineering environments. Attached Figure Description

[0044] Other objects and results of the invention will become more apparent and readily understood by referring to the following description taken in conjunction with the accompanying drawings, and with a more complete understanding of the invention. In the drawings:

[0045] Figure 1 This is a flowchart illustrating a method for assessing the quality of geosynthetic materials based on big data, according to an embodiment of the present invention.

[0046] Figure 2 This is a functional module diagram of a geosynthetic material quality assessment system based on big data, according to an embodiment of the present invention. Detailed Implementation

[0047] Existing geosynthetic material quality assessment technologies suffer from static and fragmented assessment models, as well as lagging, one-sided, and inaccurate quality risk assessments.

[0048] To address the aforementioned problems, this invention provides a method and system for assessing the quality of geotechnical materials based on big data. The specific embodiments of this invention will be described in detail below with reference to the accompanying drawings.

[0049] To illustrate the big data-based method and system for assessing the quality of geosynthetic materials provided by this invention, Figure 1 An exemplary illustration is provided for a geosynthetic material quality assessment method based on big data according to an embodiment of the present invention; Figure 2 An exemplary illustration is provided for a geotechnical material quality assessment system based on big data, according to an embodiment of the present invention.

[0050] The following description of exemplary embodiments is merely illustrative and is in no way intended to limit the invention or its application or use. Techniques and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques and equipment should be considered part of the specification.

[0051] Reference Figure 1 The diagram shown is a flowchart illustrating a geosynthetic material quality assessment method based on big data, according to an embodiment of the present invention. In this embodiment, the geosynthetic material quality assessment method based on big data includes:

[0052] Step 1: In the proposed engineering environment, the initial mechanical properties of the geotextile material are applied to the abnormal characteristic parameters generated by the environmental impact factors, and these parameters are instantiated as abnormal event nodes of the geotextile material.

[0053] In this embodiment of the invention, the process of instantiating the abnormal characteristic parameters generated by the application of the initial mechanical properties of the geotextile to environmental impact factors in the proposed engineering environment into abnormal event nodes of the geotextile is as follows:

[0054] In the proposed engineering environment, a dynamic benchmark deviation field for the geosynthetic material is constructed based on the correlation logic between the initial mechanical properties of the geosynthetic material and environmental impact factors.

[0055] Based on the dynamic benchmark deviation field, the degree of dispersion of the geomaterial between the actual performance degradation and the theoretical performance degradation curve is quantified, and the abnormal characteristic parameters of the geomaterial are generated.

[0056] Based on the time node and spatial coordinates of the aforementioned abnormal feature parameters, the spatial coordinates under each time node are instantiated into an abnormal event node of the geotechnical material.

[0057] The dynamic benchmark deviation field of the geosynthetic material is constructed based on the correlation logic between the initial mechanical properties of the geosynthetic material and environmental impact factors. The process is as follows:

[0058] The initial mechanical indices, aligned according to time sequence, are combined with the environmental impact factors to form the environmental-mechanical data pair for the geotechnical material.

[0059] The environmental-mechanical data pairs are projected onto a coordinate system with environmental impact factors on the horizontal axis and initial mechanical indices on the vertical axis to form a data scatter point, which is then fitted into a smooth curve. The change relationship represented by the smooth curve is then determined as the baseline attenuation mapping relationship.

[0060] Substituting the spatiotemporal distribution coordinates of the environmental impact factors into the benchmark attenuation mapping relationship, the dynamic benchmark deviation field of the geomaterial under no-abnormal conditions is extrapolated.

[0061] Various sensors and data acquisition terminals deployed at the engineering site systematically acquire the initial mechanical property values ​​of geomaterials at various monitoring points. These properties include, but are not limited to, key physical parameters such as tensile strength, elongation, and tear resistance. Simultaneously, environmental impact factor data for the corresponding locations are collected. These environmental impact factors encompass external conditions such as temperature gradient, humidity content, pH value, and changes in internal soil stress. The system then performs time-series alignment processing on all acquired data streams, binding the initial mechanical property values ​​and environmental impact factor values ​​within the same timestamp and spatial grid into an indivisible data unit. These paired units are defined as environmental-mechanical data pairs. This step ensures the accuracy of causal relationships between variables in subsequent analyses and avoids logical fallacies caused by time misalignment.

[0062] All generated environmental-mechanical data pairs are projected into a two-dimensional Cartesian coordinate system. The horizontal axis of this system is set as the quantified values ​​of environmental impact factors, and the vertical axis is set as the quantified values ​​of initial mechanical indicators. Each data pair is represented as a discrete data point in the coordinate system. To extract universal patterns from these discrete observations, the system uses a combination of moving average and cubic spline interpolation to perform smooth curve fitting on these data points. By setting the sliding window size, noise points caused by instantaneous disturbances are eliminated, and interpolation algorithms are used to fill in the gaps in sparse regions, ultimately forming a continuous and smooth curve. The relationship represented by this curve is established as the baseline attenuation mapping relationship. Its physical meaning lies in describing the natural attenuation or evolution trend of the mechanical properties of geotechnical materials under ideal conditions without any abnormal disturbances, as external environmental factors change.

[0063] The spatiotemporal distribution coordinates of all environmental impact factors monitored in the actual engineering environment, including specific longitude, latitude, elevation, and timestamps, are successively substituted into the established benchmark attenuation mapping relationship. For points that can be precisely matched in the coordinate system, the corresponding ordinate value is directly read as the theoretical mechanical index. For floating-point coordinates that cannot be directly matched, the system uses a bilinear interpolation algorithm to perform weighted calculations using the values ​​of its four nearest neighbors, thereby extrapolating the theoretical mechanical index values ​​that the geomaterial should exhibit under that specific spatiotemporal coordinate. Arranging all these calculated theoretical values ​​according to their original spatial location and time order, a theoretical performance field covering the entire engineering area and dynamically updated over time is constructed. This is the dynamic benchmark deviation field of the geomaterial under anomaly-free conditions, which provides an absolute reference benchmark for subsequent anomaly determination.

[0064] The system tracks the mechanical properties of geosynthetic materials under actual working conditions in real time, plots the actual performance degradation curve, and compares this curve point-by-point with the aforementioned dynamic benchmark deviation field under the same spatiotemporal coordinates. The system calculates the numerical deviation between the two, i.e., the difference between the actual measured value and the theoretical predicted value, and performs statistical analysis on these differences. By setting a dynamic threshold based on the principle of three standard deviations, statistically significant deviation values ​​are selected. These selected significant deviation values, along with their specific time nodes and spatial coordinate information, are quantified into anomaly characteristic parameters of the geosynthetic materials. These parameters not only reflect the severity of the anomaly but also record the spatiotemporal imprint of the anomaly.

[0065] The specific occurrence time and geographical coordinates corresponding to each anomalous feature parameter are extracted. This time and coordinates are treated as independent entity objects, and corresponding record entries are created in a graph database or relational database. This instantiates the abstract, numerical anomalous feature parameters into geotechnical material anomaly event nodes with clear physical meaning. Each node stores its unique attribute information, including node number, occurrence time, spatial coordinates, and anomalous feature value. These nodes form the foundation for subsequent network analysis, completing the crucial transformation from data to information.

[0066] The original environmental impact factor data stream is also instantiated as environmental factor nodes, serving as another type of basic entity in the network to facilitate the establishment of a direct link between environmental disturbances and material responses. The system then uses the spatial coordinates of all anomalous event nodes as the calculation benchmark, calculating the Euclidean distance between each pair, and determining their spatial proximity based on a preset adjacency interval threshold. Node pairs whose distance falls within this interval are identified as spatially adjacent node pairs. Following chronological order, the system draws a first-type directed edge from the earlier-occurring node to the later-occurring node between every two adjacent nodes with causal probability, thus representing the propagation direction of the anomaly along the time axis. Furthermore, the system establishes a second-type undirected edge between each pair of anomalous event nodes and their spatiotemporally matched environmental factor nodes, and attaches precise time labels to all edges, thereby successfully establishing a dynamic anomaly association network for geotechnical materials evolving over time. This network fully preserves the spatiotemporal topology and evolutionary dynamics of the anomalous events.

[0067] Following the propagation direction indicated by the first type of directed edges in the dynamic anomaly association network, the system extracts the changes in anomaly characteristic parameter values ​​between the start and end points of each edge's anomaly event nodes. These changes are then paired with the time labels of the corresponding edges, resulting in a series of anomaly change-time data pairs. The system arranges these data pairs chronologically to plot a continuous dynamic risk curve, visually demonstrating the fluctuations in quality risk. The system further extracts time windows of different lengths, calculates the cumulative value of anomaly changes within each window, and compares these cumulative values ​​with preset quality risk level intervals. If the cumulative value falls within a low-risk interval, it is marked as a low-risk window, and so on. Finally, the system arranges all time windows carrying risk labels chronologically, generating a time series consisting of low-risk, medium-risk, and high-risk periods, either sequentially or alternately. This complete series is output as the comprehensive quality assessment result of geotechnical materials, providing a comprehensive quantitative basis for engineering decisions, from microscopic anomalies to macroscopic risks.

[0068] Step 2: Instantiate the environmental impact factors as environmental factor nodes, and establish a dynamic anomaly association network for the evolution of the geotechnical material over time based on the spatial adjacency relationship and temporal sequence between each abnormal event node.

[0069] In this embodiment of the invention, the process of instantiating the environmental impact factors as environmental factor nodes and establishing a dynamic anomaly association network of the geotechnical material's evolution over time based on the spatial adjacency and temporal sequence between each anomalous event node is as follows:

[0070] Based on the spatial coordinates of the abnormal event nodes, pairs of abnormal event nodes whose spatial distance between each other falls within a preset adjacency interval are identified as spatial adjacency pairs.

[0071] Draw the first type of directed edges from the first node to the second node in chronological order;

[0072] Each of the aforementioned abnormal event nodes is connected to spatiotemporally matched environmental factor nodes through a second type of undirected edge, and a time label is attached to each edge to obtain the dynamic abnormal association network of the geotechnical material.

[0073] The formula for calculating the spatial distance between any two abnormal event nodes is as follows:

[0074] ;

[0075] In the formula, Indicates the spatial distance. This represents the spatial coordinates of one of the abnormal event nodes in the abnormal event node pair. This represents the spatial coordinates of another abnormal event node in the abnormal event node pair. This represents the preset abnormal difference adjustment coefficient. Representation of spatial coordinates The abnormal characteristic parameter values ​​of the corresponding abnormal event nodes, Representation of spatial coordinates The abnormal characteristic parameter values ​​of the corresponding abnormal event node.

[0076] The process of connecting each of the anomalous event nodes with spatiotemporally matched environmental factor nodes through a second type of undirected edge, and attaching a time label to each edge, to obtain the dynamic anomaly association network of the geotechnical material is as follows:

[0077] Based on the timestamp and spatial coordinates of the abnormal event node, the environmental factor nodes carrying the same timestamp and spatial coordinates are identified as the spatiotemporal matching nodes of the abnormal event node.

[0078] Between each pair of abnormal event nodes and their corresponding spatiotemporal matching nodes, draw a second-type undirected edge and assign a time label to each second-type undirected edge;

[0079] Arrange the first type of directed edges and the second type of undirected edges according to the order of the time tags to form the dynamic anomaly association network of the geotechnical material.

[0080] The process of drawing a second-type undirected edge between each pair of abnormal event nodes and their corresponding spatiotemporal matching nodes, and assigning a time label to each second-type undirected edge, is as follows:

[0081] Using the earlier time among the abnormal event node and the environmental factor node as the start time and the later time as the end time, the start time and the end time are jointly encapsulated into a time label for the second type of undirected edge;

[0082] Create a time tag entry in the attribute field of the second type of undirected edge, and write the time tag into the time tag entry.

[0083] The initial mechanical properties of the geomaterials at each monitoring point are obtained. These properties include, but are not limited to, key physical parameters such as tensile strength, elongation, and tear resistance. At the same time, environmental impact factor data at the corresponding locations are collected synchronously. These environmental impact factors cover external conditions such as temperature gradient, humidity content, pH value, and changes in internal soil stress, ensuring the comprehensiveness and synchronicity of the data source.

[0084] All collected data streams are time-series aligned, binding the initial mechanical index values ​​and environmental impact factor values ​​within the same timestamp and spatial grid into an indivisible data unit. These paired units are defined as environmental-mechanical data pairs. This step eliminates time-series errors caused by clock drift or transmission delay of the acquisition equipment through data cleaning and matching mechanisms, ensuring the accuracy of causal relationships between variables in subsequent analysis.

[0085] All generated environmental-mechanical data pairs are projected into a two-dimensional Cartesian coordinate system. The horizontal axis of this coordinate system is set as the quantified value of the environmental impact factor, and the vertical axis is set as the quantified value of the initial mechanical index. Each data pair is represented as a discrete data point in the coordinate system. Then, a combination of moving average method and cubic spline interpolation method is used to perform smooth curve fitting on these data points. By setting the sliding window size, noise points caused by instantaneous interference are eliminated, and the interpolation algorithm is used to fill the gaps in sparse areas, finally forming a continuous and smooth curve. The change relationship represented by this curve is established as the benchmark attenuation mapping relationship. Its physical meaning is to describe the natural attenuation or evolution trend of the mechanical properties of geotechnical materials under ideal conditions without any abnormal interference, as external environmental factors change.

[0086] The spatiotemporal distribution coordinates of all environmental impact factors monitored in the actual engineering environment are successively substituted into the above-mentioned benchmark attenuation mapping relationship. For points that can be accurately matched in the coordinate system, the corresponding ordinate value is directly read as the theoretical mechanical index. For floating-point coordinates that cannot be directly matched, the system adopts a bilinear interpolation algorithm to perform weighted calculation using the values ​​of the four nearest neighbor points around it, thereby extrapolating the theoretical mechanical index value that the geomaterial should exhibit under the specific spatiotemporal coordinate. By arranging these calculated theoretical values ​​according to their original spatial location and time order, a theoretical performance field covering the entire engineering area and dynamically updated over time is constructed, which is the dynamic benchmark deviation field of geomaterials under no abnormal conditions.

[0087] The mechanical properties of geosynthetic materials under actual working conditions are tracked in real time, and the actual performance decay curve is plotted. This curve is then compared point by point with the aforementioned dynamic benchmark deviation field under the same spatiotemporal coordinates. The numerical deviation between the two is calculated, that is, the difference between the actual measured value and the theoretical predicted value. By setting a dynamic threshold based on the principle of three standard deviations, deviation values ​​with significant statistical significance are screened out. These screened significant deviation values, along with the specific time nodes and spatial coordinate information when they occur, are quantified into the abnormal characteristic parameters of the geosynthetic materials. These parameters not only reflect the severity of the anomaly, but also record the spatiotemporal imprint of the anomaly.

[0088] Extract the specific occurrence time and geographical coordinates corresponding to each abnormal feature parameter, treat the time and coordinates as an independent entity object, and create corresponding record entries in a graph database or relational database. This instantiates the abstract, numerical abnormal feature parameters into geotechnical material abnormal event nodes with clear physical meaning. Each node stores its unique attribute information, including node number, occurrence time, spatial coordinates, and abnormal feature value. These nodes form the cornerstone of subsequent network analysis, completing the key transformation from data to information.

[0089] Using the spatial coordinates of all anomalous event nodes as the calculation benchmark, the Euclidean distance between each pair is calculated, and their spatial proximity is judged according to the preset adjacency interval threshold. Node pairs whose distance falls within the interval are identified as spatially adjacent node pairs. Then, in chronological order, the first type of directed edges are drawn from the earlier occurrence node to the later occurrence node. At the same time, the second type of undirected edges are established between each pair of anomalous event nodes and their spatiotemporally matching environmental factor nodes, and precise time labels are attached to all edges. Thus, a dynamic anomaly correlation network of geotechnical materials evolving over time is successfully established, and the risk is deduced along the network to generate the final assessment result.

[0090] The calculation of the spatial distance depends on the spatial coordinates of the two abnormal event nodes and the values ​​of the abnormal characteristic parameters. These values ​​are collected by the system or preset, wherein the spatial coordinates... and These correspond to the position data of two nodes in three-dimensional space, and the values ​​of the abnormal feature parameters. and These are then bound to the feature identifiers of the two nodes mentioned above. These data are recorded during system operation and used for subsequent calculations.

[0091] This method quantifies the spatial correlation strength between two anomalous event nodes. It considers not only the geometric interval between the nodes in three-dimensional space but also introduces an adjustment mechanism based on the difference in anomalous feature parameters. This allows the spatial distance to reflect not only physical proximity but also the similarity of the degree of anomalousness. When two nodes are spatially close and have similar anomalous feature parameter values, the adjustment term approaches zero, and the final spatial distance approaches pure geometric distance, indicating that they are highly consistent in the spatiotemporal and anomalous dimensions. However, when the spatial locations are close but the anomalous parameters differ significantly, the adjustment term increases, and the final spatial distance is amplified, indicating that although they are spatially adjacent, their anomalous natures are different and they need to be distinguished in network modeling. This formula, by mapping the difference in anomalous feature parameters to a correction factor for spatial distance, achieves an accurate characterization of the true correlation strength between anomalous event nodes, providing a quantitative basis for determining spatial adjacency relationships in subsequent dynamic anomalous association networks.

[0092] As the spatial coordinate difference between two anomalous event nodes increases, the base distance monotonically increases, and the overall spatial distance also monotonically increases accordingly; simultaneously, as the anomalous characteristic parameter values ​​of the two nodes increase... and When the difference increases, the numerator in the adjustment term increases while the denominator remains unchanged, leading to an increase in the adjustment term and consequently a further increase in the final spatial distance; if and As the difference decreases, the adjustment term decreases, and the final spatial distance approaches the basic Euclidean distance; when and When they are completely equal, the adjustment term is zero, and the final spatial distance equals the basic Euclidean distance; when and When the difference is greatest, the adjustment term reaches its maximum value, and the final spatial distance is significantly amplified; this trend is reflected in the adjustment coefficient. For positive time to be established, if A negative value indicates a trend reversal, but this is subject to system settings. As a pre-set positive coefficient, the overall trend is as follows: spatial distance increases with the expansion of spatial interval, further increases with the expansion of differences in abnormal feature parameters, and decreases to the basic geometric distance as abnormal feature parameters converge. This trend ensures that in spatial adjacency determination, both physical location and abnormality properties are considered, avoiding misjudging nodes with different abnormal features but spatial proximity as related nodes, thereby improving the accuracy and physical interpretability of dynamic abnormal association networks.

[0093] Step 3: Based on the dynamic anomaly correlation network, deduce the continuous quality risk of the geosynthetic material and generate the comprehensive quality assessment result of the geosynthetic material.

[0094] In this embodiment of the invention, the process of inferring the continuous quality risk of the geosynthetic material based on the dynamic anomaly correlation network and generating a comprehensive quality assessment result for the geosynthetic material is as follows:

[0095] Along the propagation direction indicated by the first type of directed edge in the dynamic anomaly association network, the change in the abnormal feature parameter value between the start and end points of each first type of directed edge is paired with the time label of the corresponding first type of directed edge to obtain the abnormal change amount-time data pair of the geotechnical material.

[0096] Arrange the abnormal change-time data pairs in chronological order to form the dynamic risk curve of the geosynthetic material;

[0097] In the dynamic risk curve, the cumulative value of abnormal changes within different time windows is extracted, and the matching result between the cumulative value of abnormal changes and the preset quality risk level interval is integrated into the comprehensive quality assessment result of the geotechnical material.

[0098] The process of integrating the matching results between the cumulative value of the abnormal changes and the preset quality risk level range into the comprehensive quality assessment result of the geosynthetic material is as follows:

[0099] The cumulative value of abnormal changes in each time window is sequentially mapped to a preset quality risk level range for comparison;

[0100] If the cumulative value of the abnormal change falls into the low-risk range, the time window is marked as a low-risk window, and so on, to obtain the low-risk window, medium-risk window and high-risk window of the time window.

[0101] The time windows carrying risk markers are arranged in chronological order to generate a time series consisting of low-risk, medium-risk, and high-risk periods distributed sequentially or alternately, and the time series is used as the comprehensive quality assessment result of the geotechnical material.

[0102] Along the propagation direction indicated by the first type of directed edge, the abnormal feature parameter values ​​of the abnormal event nodes at the start and end points of each directed edge are extracted. The difference between the corresponding parameter values ​​of these two nodes is calculated as the change. This change is then paired with the time label carried by the corresponding first type of directed edge to form an abnormal change-time data pair. Subsequently, all the generated data pairs are arranged in chronological order to form a dynamic risk curve. Then, time windows of different lengths are extracted from the dynamic risk curve, and the cumulative sum of all abnormal changes within each time window is calculated. The cumulative value of each time window is then compared with a pre-set quality risk level range. Based on the range in which the cumulative value falls, each time window is labeled as low-risk, medium-risk, or high-risk. Finally, all time windows with risk labels are arranged in chronological order and integrated to generate a time series consisting of low-risk, medium-risk, and high-risk periods distributed sequentially or alternately. This serves as the comprehensive quality assessment result of the geotechnical material.

[0103] The system identifies all first-type directed edges from the dynamic anomaly correlation network. For each first-type directed edge, the system locates the anomaly event node at the starting and ending points of the edge. It reads the anomaly feature parameter values ​​recorded in the starting and ending nodes, subtracts the parameter values ​​from the starting node's value, and the difference is the anomaly change amount corresponding to this directed edge. Simultaneously, it reads the time label attached to the first-type directed edge itself and combines the calculated anomaly change amount with this time label to form an anomaly change amount-time data pair. The system performs the above extraction and pairing operations on each first-type directed edge in the dynamic anomaly correlation network until all directed edges generate corresponding data pairs. All generated anomaly change amount-time data pairs are collected, and the time label recorded in each data pair is read. All data pairs are arranged sequentially according to the time represented by the time label from earliest to latest. After arrangement, these data pairs form a continuous curve, reflecting the dynamic process of anomaly feature parameters changing over time, i.e., the dynamic risk curve. On the pre-arranged dynamic risk curve, several consecutive time windows are set. The length of the time window can be determined according to actual needs. For each time window, the system extracts the abnormal change values ​​in all abnormal change-time data pairs contained in the window, and sums these values ​​to obtain the cumulative value of abnormal change within the time window. The system performs the same cumulative calculation on all set time windows to obtain the cumulative value corresponding to each time window.

[0104] The system compares the cumulative value of abnormal changes calculated for each time window with a pre-defined quality risk level range. The pre-defined quality risk level range includes low-risk, medium-risk, and high-risk ranges, each with a clearly defined numerical range. The system determines which range the cumulative value of the current time window falls into. If the cumulative value falls into the low-risk range, the time window is marked as a low-risk window; if it falls into the medium-risk range, it is marked as a medium-risk window; and if it falls into the high-risk range, it is marked as a high-risk window. After marking all time windows, the system obtains low-risk, medium-risk, and high-risk windows with risk labels. Collect all time windows marked as low-risk, medium-risk, and high-risk windows, read the time range of each window, and arrange these risk-marked windows in order from early to late according to the time range. The arrangement forms a continuous time series, which contains low-risk, medium-risk, and high-risk periods that appear alternately or consecutively in chronological order. The system outputs this time series as the final comprehensive quality assessment result of geotechnical materials.

[0105] As can be seen from the above embodiments, the geotechnical material quality assessment method based on big data provided by the present invention accurately quantifies the abnormal characteristic parameters by constructing a dynamic benchmark deviation field and establishing a dynamic abnormal correlation network that integrates spatiotemporal topology and causal propagation characteristics. This enables quantitative tracking of the entire process from micro-anomaly points to macro-risk evolution, and the final output of a continuous quality risk assessment time series provides an intuitive, advanced and spatiotemporally resolved scientific basis for engineering decision-making.

[0106] like Figure 2 The diagram shown is a functional block diagram of a geotechnical material quality assessment system 100 based on big data provided in an embodiment of the present invention, including an abnormal node generation module 101, a dynamic network modeling module 102, and a quality risk assessment module 103.

[0107] In this embodiment, the functions of each module are as follows:

[0108] The abnormal node generation module 101 is used to instantiate the abnormal characteristic parameters generated by the initial mechanical properties of the geomaterial acting on the environmental impact factors in the proposed application engineering environment into abnormal event nodes of the geomaterial.

[0109] The dynamic network modeling module 102 is used to instantiate the environmental impact factors into environmental factor nodes, and establish a dynamic anomaly association network of the geotechnical material over time based on the spatial adjacency relationship and temporal sequence between each abnormal event node.

[0110] The quality risk assessment module 103 is used to deduce the continuous quality risk of the geomaterial based on the dynamic anomaly correlation network and generate a comprehensive quality assessment result of the geomaterial.

[0111] As can be seen from the above embodiments, the geosynthetic material quality assessment system provided by the present invention, through modular collaborative operation, realizes a closed-loop assessment of the entire chain from the quantification of micro-anomaly characteristics and the construction of spatiotemporal dynamic networks to the macro-continuous risk extrapolation, which significantly improves the accuracy of geosynthetic material quality monitoring, the timeliness of early warning, and the scientific nature of engineering decision-making.

[0112] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0113] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for quality assessment of geosynthetic materials based on big data, characterized in that, The method includes: Step 1: In the proposed engineering environment, the initial mechanical properties of the geotextile material are applied to the abnormal characteristic parameters generated by the environmental impact factors, and these parameters are instantiated as abnormal event nodes of the geotextile material. Step 2: Instantiate the environmental impact factors as environmental factor nodes, and establish a dynamic anomaly association network for the evolution of the geotechnical material over time based on the spatial adjacency relationship and temporal sequence between each abnormal event node. Step 3: Based on the dynamic anomaly correlation network, deduce the continuous quality risk of the geosynthetic material and generate the comprehensive quality assessment result of the geosynthetic material.

2. The method for assessing the quality of geosynthetic materials based on big data as described in claim 1, characterized in that, In the proposed engineering environment, the abnormal characteristic parameters generated by applying the initial mechanical properties of the geomaterial to environmental impact factors are instantiated as abnormal event nodes of the geomaterial. The process is as follows: In the proposed engineering environment, a dynamic benchmark deviation field for the geosynthetic material is constructed based on the correlation logic between the initial mechanical properties of the geosynthetic material and environmental impact factors. Based on the dynamic benchmark deviation field, the degree of dispersion of the geomaterial between the actual performance degradation and the theoretical performance degradation curve is quantified, and the abnormal characteristic parameters of the geomaterial are generated. Based on the time node and spatial coordinates of the aforementioned abnormal feature parameters, the spatial coordinates under each time node are instantiated into an abnormal event node of the geotechnical material.

3. The method for assessing the quality of geosynthetic materials based on big data as described in claim 2, characterized in that, The dynamic benchmark deviation field of the geosynthetic material is constructed based on the correlation logic between the initial mechanical properties of the geosynthetic material and environmental impact factors. The process is as follows: The initial mechanical indices, aligned according to time sequence, are combined with the environmental impact factors to form the environmental-mechanical data pair for the geotechnical material. The environmental-mechanical data pairs are projected onto a coordinate system with environmental impact factors on the horizontal axis and initial mechanical indices on the vertical axis to form a data scatter point, which is then fitted into a smooth curve. The change relationship represented by the smooth curve is then determined as the baseline attenuation mapping relationship. Substituting the spatiotemporal distribution coordinates of the environmental impact factors into the benchmark attenuation mapping relationship, the dynamic benchmark deviation field of the geomaterial under no-abnormal conditions is extrapolated.

4. The method for assessing the quality of geosynthetic materials based on big data as described in claim 1, characterized in that, The process of instantiating the environmental impact factors into environmental factor nodes and establishing a dynamic anomaly association network for the evolution of the geotechnical material over time based on the spatial adjacency and temporal sequence of each anomaly event node is as follows: Based on the spatial coordinates of the abnormal event nodes, pairs of abnormal event nodes whose spatial distance between each other falls within a preset adjacency interval are identified as spatial adjacency pairs. Draw the first type of directed edges from the first node to the second node in chronological order; Each of the aforementioned abnormal event nodes is connected to spatiotemporally matched environmental factor nodes through a second type of undirected edge, and a time label is attached to each edge to obtain the dynamic abnormal association network of the geotechnical material.

5. The method for assessing the quality of geosynthetic materials based on big data as described in claim 4, characterized in that, The formula for calculating the spatial distance between any two abnormal event nodes is as follows: ； In the formula, Indicates the spatial distance. This represents the spatial coordinates of one of the abnormal event nodes in the abnormal event node pair. This represents the spatial coordinates of another abnormal event node in the abnormal event node pair. This represents the preset abnormal difference adjustment coefficient. Representation of spatial coordinates The abnormal characteristic parameter values ​​of the corresponding abnormal event nodes, Representation of spatial coordinates The abnormal characteristic parameter values ​​of the corresponding abnormal event node.

6. The method for assessing the quality of geosynthetic materials based on big data as described in claim 4, characterized in that, The process of connecting each of the anomalous event nodes with spatiotemporally matched environmental factor nodes through a second type of undirected edge, and attaching a time label to each edge, to obtain the dynamic anomaly association network of the geotechnical material is as follows: Based on the timestamp and spatial coordinates of the abnormal event node, the environmental factor nodes carrying the same timestamp and spatial coordinates are identified as the spatiotemporal matching nodes of the abnormal event node. Between each pair of abnormal event nodes and their corresponding spatiotemporal matching nodes, draw a second-type undirected edge and assign a time label to each second-type undirected edge; Arrange the first type of directed edges and the second type of undirected edges according to the order of the time tags to form the dynamic anomaly association network of the geotechnical material.

7. The method for assessing the quality of geosynthetic materials based on big data as described in claim 6, characterized in that, The process of drawing a second-type undirected edge between each pair of abnormal event nodes and their corresponding spatiotemporal matching nodes, and assigning a time label to each second-type undirected edge, is as follows: Using the earlier time among the abnormal event node and the environmental factor node as the start time and the later time as the end time, the start time and the end time are jointly encapsulated into a time label for the second type of undirected edge; Create a time tag entry in the attribute field of the second type of undirected edge, and write the time tag into the time tag entry.

8. The method for assessing the quality of geosynthetic materials based on big data as described in claim 1, characterized in that, The process of inferring the continuous quality risk of the geosynthetic material based on the dynamic anomaly correlation network and generating a comprehensive quality assessment result for the geosynthetic material is as follows: Along the propagation direction indicated by the first type of directed edge in the dynamic anomaly association network, the change in the abnormal feature parameter value between the start and end points of each first type of directed edge is paired with the time label of the corresponding first type of directed edge to obtain the abnormal change amount-time data pair of the geotechnical material. Arrange the abnormal change-time data pairs in chronological order to form the dynamic risk curve of the geosynthetic material; In the dynamic risk curve, the cumulative value of abnormal changes within different time windows is extracted, and the matching result between the cumulative value of abnormal changes and the preset quality risk level interval is integrated into the comprehensive quality assessment result of the geotechnical material.

9. The method for assessing the quality of geosynthetic materials based on big data as described in claim 8, characterized in that, The process of integrating the matching results between the cumulative value of the abnormal changes and the preset quality risk level range into the comprehensive quality assessment result of the geosynthetic material is as follows: The cumulative value of abnormal changes in each time window is sequentially mapped to a preset quality risk level range for comparison; If the cumulative value of the abnormal change falls into the low-risk range, the time window is marked as a low-risk window, and so on, to obtain the low-risk window, medium-risk window and high-risk window of the time window. The time windows carrying risk markers are arranged in chronological order to generate a time series consisting of low-risk, medium-risk, and high-risk periods distributed sequentially or alternately, and the time series is used as the comprehensive quality assessment result of the geotechnical material.

10. A geosynthetic material quality assessment system based on big data, characterized in that, The system for implementing the big data-based geosynthetic material quality assessment method according to any one of claims 1-9 includes: An anomaly node generation module is used to instantiate the abnormal characteristic parameters generated by the initial mechanical properties of the geomaterial acting on environmental impact factors in the proposed application engineering environment into anomaly event nodes of the geomaterial. The dynamic network modeling module is used to instantiate the environmental impact factors into environmental factor nodes, and establish a dynamic anomaly association network of the geotechnical material over time based on the spatial adjacency relationship and temporal sequence between each abnormal event node. The quality risk assessment module is used to deduce the continuous quality risk of the geomaterial based on the dynamic anomaly correlation network and generate a comprehensive quality assessment result of the geomaterial.

Images