Risk prediction method, device and system for soft soil foundation scene

Abstract

This application relates to the field of port foundation deformation monitoring technology, and discloses a risk prediction method for soft soil foundation scenarios. The method includes: acquiring deformation data and usage data at several sampling times within several grid areas; calculating the data difference between adjacent deformation data in each grid area; combining the data difference from a specific historical time to the current time and the grid position coordinates within each grid area into a deformation vector; clustering multiple deformation vectors to obtain one or more clusters; obtaining the correlation between historical overall usage data and historical overall deformation data; obtaining future overall usage data; determining the future overall deformation data corresponding to the future overall usage data based on the correlation; and determining the risk level corresponding to the future overall deformation data. This method can be used to predict the deformation risk of soft soil foundations in use. This application also discloses a risk prediction device and system for soft soil foundation scenarios.

Description

Technical Field

[0001] This application relates to the field of port foundation deformation monitoring technology, such as a risk prediction method, device and system for soft soil foundation scenarios. Background Technology

[0002] The Tianjin Port Nanjiang Yuanhang Yard, located in the Nanjiang Port Area of ​​Tianjin Port, was formed by land reclamation, resulting in a soft soil foundation. Natural soil is a three-phase system composed of a mineral particle skeleton and pores filled with pore water and gas. Soil particles and water have very low compressibility and are generally considered incompressible. In contrast, soft soil is characterized by its looseness, high porosity, high natural water content, high compressibility, low strength, low permeability, and sensitive structure, making it prone to deformation during initial use. Furthermore, as a specialized dry bulk cargo yard, the Yuanhang Yard experiences heavy loads and long periods of full-load operation, leading to various anomalies in facilities on the soft soil foundation. These include significant foundation settlement, deformation and displacement of stacker-reclaimer tracks, and tilting and deformation of high-mast lights. These phenomena severely impact the normal operation of the port; therefore, monitoring the deformation of the port's soft soil foundation is crucial.

[0003] Existing technologies can invert surface deformation through synthetic aperture radar interferometry (InSAR) to monitor soft soil foundation deformation. Furthermore, time-series deformation data can be processed based on long short-term memory (LSTM) networks to predict soft soil foundation deformation.

[0004] However, the deformation of soft soil foundations is determined by both the properties of the soft soil and the condition of the foundation in use. This technology only studies the apparent deformation data of soft soil foundations. It is suitable for idealized soft soil foundation models, but in actual use, the load on the soft soil foundation changes, and the internal consolidation state, void state, and moisture state of the soft soil foundation will also change. This makes it difficult for existing technologies to accurately predict the deformation results and deformation risks of soft soil foundations in use. Summary of the Invention

[0005] To provide a basic understanding of some aspects of the disclosed embodiments, a brief summary is given below. This summary is not intended as a general commentary, nor is it intended to identify key / important components or describe the scope of protection of these embodiments, but rather as a prelude to the detailed description that follows.

[0006] This application provides a risk prediction method, apparatus, and system for soft soil foundation scenarios, enabling the prediction of deformation risks of soft soil foundations in use.

[0007] In some embodiments, the risk prediction method for soft soil foundation scenarios includes:

[0008] The system acquires deformation and usage data at several sampling times within several grid areas. The grid areas are pre-divided, and the grid areas cover part or all of the port infrastructure. The infrastructure covered by the grids includes one or more of the following: yard area, yard edge, stacker-reclaimer track, and high mast lights, but not only high mast lights. The deformation data is the deformation data corresponding to the infrastructure covered in each grid area. The usage data is load-related data.

[0009] For the deformation data of each grid region, the data difference between every two adjacent deformation data points is calculated according to the order of sampling time;

[0010] The data difference from a specific historical moment to the current moment and the grid position coordinates within each grid region are combined into a deformation vector;

[0011] Clustering multiple deformed vectors yields one or more clusters; the clustering effect at a specific historical moment is better than the clustering effect at other historical moments;

[0012] The usage data of the grid region corresponding to all deformation vectors at the same time is taken as the historical overall usage data, and the deformation data of the grid region corresponding to all deformation vectors at the same time is taken as the historical overall deformation data. The correlation between the historical overall usage data and the historical overall deformation data is obtained.

[0013] Determine future overall usage data based on the changing trends of historical overall usage data;

[0014] Determine the future overall deformation data corresponding to the future overall usage data based on the correlation relationships;

[0015] The risk level corresponding to the overall future deformation data is determined based on the correspondence between deformation data and risk levels.

[0016] In some embodiments, the risk prediction device for soft soil foundation scenarios includes a first acquisition module, a difference calculation module, a combination module, a clustering module, a second acquisition module, a first determination module, a second determination module, and a first risk module.

[0017] The first acquisition module is used to acquire deformation data and usage data at several sampling times within several grid areas. The grid areas are pre-divided, and the several grid areas cover part or all of the port infrastructure. The infrastructure covered by the several grids includes one or more of the following: yard area, yard edge, stacker-reclaimer track, and high mast lights, but not only high mast lights. The deformation data is the deformation data corresponding to the infrastructure covered in each grid area. The usage data is load-related data.

[0018] The difference calculation module is used to calculate the data difference between every two adjacent deformation data points based on the order of sampling time for the deformation data of each grid region.

[0019] The combination module is used to combine the data difference from a specific historical moment to the current moment and the grid position coordinates within each grid region into a deformation vector.

[0020] The clustering module is used to cluster multiple deformed vectors to obtain one or more clusters; the clustering effect at a specific historical moment is better than the clustering effect at other historical moments.

[0021] The second acquisition module is used to take the usage data of the grid area corresponding to all deformation vectors at the same time as the historical overall usage data, and the deformation data of the grid area corresponding to all deformation vectors at the same time as the historical overall deformation data, and to obtain the correlation between the historical overall usage data and the historical overall deformation data.

[0022] The first determination module is used to determine future overall usage data based on the changing trends of historical overall usage data.

[0023] The second determining module is used to determine the future overall deformation data corresponding to the future overall usage data based on the correlation relationship.

[0024] The first risk module is used to determine the risk level corresponding to the overall future deformation data based on the correspondence between deformation data and risk levels.

[0025] In some embodiments, the risk prediction apparatus for soft soil foundation scenarios includes a processor and a memory storing program instructions. The processor is configured to execute the risk prediction method for soft soil foundation scenarios provided in the foregoing embodiments when executing the program instructions.

[0026] In some embodiments, the risk prediction system for soft soil foundation scenarios includes the risk prediction device for soft soil foundation scenarios provided in the foregoing embodiments.

[0027] The risk prediction method, apparatus, and system for soft soil foundation scenarios provided in this application can achieve the following technical effects:

[0028] Several grid areas are pre-divided to cover part or all of the port infrastructure. When predicting the deformation risk of part of the soft soil foundation of the port, several grid areas need to cover part of the port infrastructure. When predicting the deformation risk of the entire soft soil foundation of the port, several grid areas need to cover the entire port infrastructure.

[0029] Different areas on a soft soil foundation have different infrastructures, and these facilities all have corresponding deformation data. This deformation data can reflect the deformation state of the soft soil foundation in each grid area.

[0030] For the deformation data of each grid region, the data difference between every two adjacent deformation data points is calculated according to the sampling time sequence. These data differences reflect the deformation gradient of each grid region. After performing cluster analysis on the deformation vectors containing the deformation gradient, it is equivalent to "circling" grid regions with similar deformation gradients.

[0031] Under ideal conditions of land reclamation according to standards, port infrastructure set up according to standards, and then ideal use, the deformation of the soft soil foundation corresponding to each infrastructure should not be coupled. That is, the deformation gradient of different grid areas should present an idealized random distribution state, which is fundamentally difficult to cluster.

[0032] However, the complex conditions of soft soil foundations and the complex operating conditions of ports inevitably lead to coupling of foundation deformations corresponding to different infrastructures. Coupling factors include, but are not limited to: stockpile yards, stacker-reclaimer tracks, and high-mast lights being located in the same or adjacent backfill layers; high loads occurring at a stockpile location near the edge of the stockpile, causing severe lateral deformation at the edge of the stockpile, which in turn affects one side of the stacker-reclaimer track; and high loads at stockpile locations causing changes in drainage structures, which in turn lead to abnormalities in the internal gaps, moisture content, and consolidation rate of the soft soil foundation.

[0033] The specific historical moment in this application's technical solution reflects the moment when abnormal coupling occurs in the deformation of soft soil foundations corresponding to different infrastructures: In a normal, uncoupled state, the deformation of soft soil foundations corresponding to different infrastructures exhibits a random distribution, resulting in poor clustering of deformation gradients; if coupling occurs due to anomalies, the deformation gradients of adjacent grid regions within the coupling area will show varying degrees of similarity. Combining the characteristics of port soft soil foundations and the basic characteristics of clustering analysis algorithms, the "specific historical moment" in "the clustering effect at a specific historical moment is better than the clustering effect at other historical moments" can reflect the moment when abnormal coupling occurs in the deformation of soft soil foundations corresponding to different infrastructures.

[0034] Meanwhile, cluster analysis is performed on deformation vectors containing deformation gradients to "circle" grid regions with similar deformation gradients, which are then identified as grid regions with deformation coupling. For these "special" grid regions, the usage data of all special grid regions is used as a historical overall usage data set, and the deformation data of all special grid regions is used as a historical overall deformation data set. By treating the coupled terrain as a whole rather than decoupling it, the correlation between the overall usage and overall deformation of these special grid regions is extracted. Then, future overall usage data is used to predict future overall deformation data and determine the corresponding risks. By fully considering the complex changing state of soft soil foundations and the complex operating conditions of ports, the predicted deformation data and deformation risks are more consistent with the actual changing state of soft soil foundations and the actual operating conditions of ports, resulting in more accurate prediction results.

[0035] The above general description and the description below are exemplary and illustrative only and are not intended to limit this application. Attached Figure Description

[0036] One or more embodiments are illustrated by way of example with reference to the accompanying drawings. These illustrative descriptions and drawings do not constitute a limitation on the embodiments. Elements having the same reference numerals in the drawings are considered similar elements, and wherein:

[0037] Figure 1 This is a flowchart illustrating a risk prediction method for soft soil foundation scenarios provided in an embodiment of this application;

[0038] Figure 2 This is a flowchart illustrating another risk prediction method for soft soil foundation scenarios provided in an embodiment of this application;

[0039] Figure 3 This is a schematic diagram illustrating the principle of lateral deformation monitoring of soil at the edge of a stockpile, provided in an embodiment of this application.

[0040] Figure 4 This is a schematic diagram illustrating the principle of monitoring track deformation of a stacker-reclaimer provided in an embodiment of this application;

[0041] Figure 5 This is a schematic diagram of a risk prediction device for soft soil foundation scenarios provided in an embodiment of this application;

[0042] Figure 6 This is a schematic diagram of another risk prediction device for soft soil foundation scenarios provided in the embodiments of this application. Detailed Implementation

[0043] To provide a more detailed understanding of the features and technical content of the embodiments of this application, the implementation of the embodiments of this application will be described in detail below with reference to the accompanying drawings. The accompanying drawings are for illustrative purposes only and are not intended to limit the embodiments of this application. In the following technical description, for ease of explanation, several details are used to provide a full understanding of the disclosed embodiments. However, one or more embodiments may still be implemented without these details. In other cases, well-known structures and devices may be simplified in their depiction to simplify the drawings.

[0044] The terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of this application described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion.

[0045] Unless otherwise stated, the term "multiple" means two or more.

[0046] In this embodiment, the character " / " indicates that the objects before and after it are in an "or" relationship. For example, A / B means: A or B.

[0047] The term "and / or" describes an association between objects, indicating that three relationships can exist. For example, A and / or B means: A or B, or A and B.

[0048] The term "storeyard" refers to the area in a port where goods are stored.

[0049] The term "yard edge" refers to the area within a predetermined distance from the storage area, or the area where shear deformation occurs within the soil after goods are stored in the yard.

[0050] The term "stack reclaimer track" refers to the track on which mechanical equipment for stacking and reclaiming goods runs. This equipment includes, but is not limited to, overhead bridge belt stackers, rake stackers, overhead bridge belt stackers, and rotary cantilever belt stackers.

[0051] The technical solutions in this application are based on the application scenario of Tianjin Port Nanjiang Shipping Yard, but are not limited to the application scenario of Tianjin Port Nanjiang Shipping Yard. They can also be applied to other ports with similar infrastructure and soft soil foundations.

[0052] Figure 1 This is a flowchart illustrating a risk prediction method for soft soil foundation scenarios provided in an embodiment of this application. This risk prediction method can be executed via a local computer or a cloud server.

[0053] Combination Figure 1 As shown, risk prediction methods for soft soil foundation scenarios include:

[0054] S101. Obtain deformation data and usage data at several sampling times within several grid areas.

[0055] The grid areas are pre-divided, and some grid areas cover part or all of the port infrastructure. The infrastructure covered by some grids includes one or more of the following: yard area, yard edge, stacker-reclaimer track and high mast lights, but not only high mast lights.

[0056] When dividing the grid regions, each grid region can have the same area; alternatively, the more frequent the load changes in a region, the smaller the corresponding grid region area should be, and the less frequent the load changes, the larger the corresponding grid region area should be. In this way, without increasing the computational load, the calculated and predicted foundation deformation and its risks will be more accurate.

[0057] The aforementioned deformation data covers the deformation data corresponding to the infrastructure within each grid area.

[0058] Specifically, when the infrastructure is a stockpile area, the deformation data is the first uneven settlement data of the stockpile; when the infrastructure is the edge of the stockpile, the deformation data is the lateral deformation data of the edge of the stockpile; when the infrastructure is a stacker-reclaimer track, the deformation data is the second uneven settlement data of the cross-section of the stacker-reclaimer track; and when the infrastructure is a high-mast light, the deformation data is the tilt data of the high-mast light.

[0059] Optionally, when several grid areas cover two or more types of infrastructure, the deformation data corresponding to different types of infrastructure are normalized before calculating the data difference between every two adjacent deformation data.

[0060] In principle, one deformation data corresponds to one grid region. If a region corresponds to two or more deformation data, the grid region can be reduced so that the two or more deformation data are used as the deformation data corresponding to two or more grid regions.

[0061] The aforementioned data is load-related data.

[0062] Specifically, load-related data includes one or more of the following: load amount, load dwell time, and load change rate in this grid area or adjacent grid areas.

[0063] Based on this, the more frequent the load changes in the above-mentioned areas, including but not limited to: frequent changes in load magnitude, frequent changes in load dwell time, and frequent changes in load change rate that are sometimes fast and sometimes slow; the less frequent the load changes in the above-mentioned areas, including but not limited to: stable load magnitude with little change; stable load dwell time with little change; and stable load change rate with little change that is sometimes fast and sometimes slow.

[0064] Optionally, when the load data types are different in several grid regions, the load data of different types are normalized before using the usage data of the grid regions corresponding to all deformation vectors at the same time as the overall historical usage data.

[0065] In addition, the usage data corresponding to a grid area can contain one or more types of usage data.

[0066] S102. For the deformation data of each grid region, calculate the data difference between every two adjacent deformation data according to the order of sampling time.

[0067] S103. Combine the data difference from a specific historical time to the current time and the grid position coordinates within each grid region into a deformation vector.

[0068] The grid position coordinates can be represented by the absolute coordinates of the grid region center point in the port, or by the relative coordinates of the grid region center point in the overall region formed by several grid regions; they can also be represented by the absolute coordinate sequence of the grid region edge in the port, or by the relative coordinate sequence of the grid region edge in the overall region formed by several grid regions.

[0069] S104. Cluster multiple deformed vectors to obtain one or more clusters.

[0070] The clustering method here refers to the algorithm that can group similar objects together. This clustering method includes, but is not limited to, the K-Nearest Neighbor (KNN) algorithm and Fuzzy Clustering Mean (FCM).

[0071] The clustering results at a specific historical moment in this step are better than those at other historical moments.

[0072] Optionally, the process of determining a specific historical moment includes: acquiring several historical moments; for each historical moment, combining the data difference from the historical moment to the current moment and the grid position coordinates within each grid region into a deformation vector, performing clustering processing on multiple deformation vectors, evaluating the clustering effect, and selecting the historical moment with the best clustering effect as the specific historical moment.

[0073] Alternatively, the process of determining a specific historical moment includes: acquiring a first historical moment; combining the data difference from the first historical moment to the current moment and the grid position coordinates within each grid region into a deformed vector; performing clustering processing on multiple deformed vectors and evaluating the clustering effect; acquiring a second historical moment before the first historical moment and a third historical moment after the first historical moment, and sequentially acquiring the clustering effect corresponding to the second historical moment and the third historical moment; if the clustering effect of the second historical moment is better than the clustering effect corresponding to the first historical moment and the third historical moment, then continuing to acquire other historical moments before the second historical moment and acquiring the clustering effect corresponding to other historical moments, and so on, until the clustering effect corresponding to a newly acquired historical moment no longer becomes better, and the historical moment corresponding to the best clustering effect is taken as the specific historical moment; if the clustering effect of the third historical moment is better than the clustering effect corresponding to the first historical moment and the second historical moment, then continuing to acquire other historical moments after the third historical moment and acquiring the clustering effect corresponding to other historical moments, and so on, until the clustering effect corresponding to a newly acquired historical moment no longer becomes better, and the historical moment corresponding to the best clustering effect is taken as the specific historical moment.

[0074] The first historical moment can be randomly obtained, or it can be the moment when a special event occurs. Special events include, but are not limited to, the moment when a stockpile is above its historical maximum load, the moment when heavy rainfall occurs, the moment when the stacker-reclaimer malfunctions or is under maintenance, and the moment when a high-mast light is accidentally struck.

[0075] Metrics for evaluating clustering performance include, but are not limited to: Silhouette Coefficient, Calinski Harabasz Index, and Davies-Bouldin Index (DBI).

[0076] The evaluation of clustering effect in this embodiment is used to determine a specific historical moment. It only requires obtaining the evaluation result of the clustering effect, and does not limit the evaluation indicators and their corresponding evaluation methods. Those skilled in the art can select the evaluation indicators and their corresponding evaluation methods based on experience, which will not be elaborated here.

[0077] S105. Obtain the correlation between historical overall usage data and historical overall deformation data.

[0078] Historical overall usage data and historical overall deformation data are determined as follows: the usage data of the grid area corresponding to all deformation vectors at the same moment is taken as historical overall usage data, and the deformation data of the grid area corresponding to all deformation vectors at the same moment is taken as historical overall deformation data.

[0079] The methods for obtaining correlations mentioned above include, but are not limited to: chart analysis, correlation coefficient analysis, regression analysis, and neural networks.

[0080] The main purpose of this application is to obtain the correlation between historical overall usage data and historical overall deformation data. No specific scheme for obtaining the correlation is limited. Those skilled in the art can adopt specific correlation analysis schemes based on experience.

[0081] S106. Determine future overall usage data based on the changing trends of historical overall usage data.

[0082] The trend of historical overall usage data is the trend of port operations. If the port loading and unloading plan remains unchanged, the future overall usage data can be determined based on the trend of historical overall usage data.

[0083] Furthermore, when port loading and unloading plans change, future overall usage data can be determined based on historical overall usage data and the port loading and unloading plan itself. The port loading and unloading plan is the core data source. Using historical overall usage data as a reference, the future usage data for each infrastructure (grid area) corresponding to the changed port loading and unloading plan is determined. The usage data at the same moment for all grid areas corresponding to the deformation vectors is then extracted and used as the future overall usage data. Alternatively, using the port loading and unloading plan as the core and historical overall usage data as a reference, the usage data at the same moment for all grid areas corresponding to the deformation vectors can be directly determined and used as the future overall usage data.

[0084] S107. Determine the future overall deformation data corresponding to the future overall usage data based on the correlation relationship.

[0085] S108. Determine the risk level corresponding to the overall future deformation data based on the correspondence between deformation data and risk level.

[0086] In the risk prediction method for soft soil foundation scenarios provided in this application embodiment, several grid areas are pre-divided to cover part or all of the port infrastructure. When predicting the deformation risk of part of the soft soil foundation of the port, several grid areas need to cover part of the port infrastructure. When predicting the deformation risk of all soft soil foundation of the port, several grid areas need to cover all of the port infrastructure.

[0087] Different areas on a soft soil foundation have different infrastructures, and these facilities all have corresponding deformation data. This deformation data can reflect the deformation state of the soft soil foundation in each grid area.

[0088] For the deformation data of each grid region, the data difference between every two adjacent deformation data points is calculated according to the sampling time sequence. These data differences reflect the deformation gradient of each grid region. After performing cluster analysis on the deformation vectors containing the deformation gradient, it is equivalent to "circling" grid regions with similar deformation gradients.

[0089] Under ideal conditions of land reclamation according to standards, port infrastructure set up according to standards, and then ideal use, the deformation of the soft soil foundation corresponding to each infrastructure should not be coupled. That is, the deformation gradient of different grid areas should present an idealized random distribution state, which is fundamentally difficult to cluster.

[0090] However, the complex conditions of soft soil foundations and the complex operating conditions of ports inevitably lead to coupling of foundation deformations corresponding to different infrastructures. Coupling factors include, but are not limited to: stockpile yards, stacker-reclaimer tracks, and high-mast lights being located in the same or adjacent backfill layers; high loads occurring at a stockpile location near the edge of the stockpile, causing severe lateral deformation at the edge of the stockpile, which in turn affects one side of the stacker-reclaimer track; and high loads at stockpile locations causing changes in drainage structures, which in turn lead to abnormalities in the internal gaps, moisture content, and consolidation rate of the soft soil foundation.

[0091] The specific historical moment in this application's technical solution reflects the moment when abnormal coupling occurs in the deformation of soft soil foundations corresponding to different infrastructures: In a normal, uncoupled state, the deformation of soft soil foundations corresponding to different infrastructures exhibits a random distribution, resulting in poor clustering of deformation gradients; if coupling occurs due to anomalies, the deformation gradients of adjacent grid regions within the coupling area will show varying degrees of similarity. Combining the characteristics of port soft soil foundations and the basic characteristics of clustering analysis algorithms, the "specific historical moment" in "the clustering effect at a specific historical moment is better than the clustering effect at other historical moments" can reflect the moment when abnormal coupling occurs in the deformation of soft soil foundations corresponding to different infrastructures.

[0092] Meanwhile, cluster analysis is performed on deformation vectors containing deformation gradients to "circle" grid regions with similar deformation gradients, which are then identified as grid regions with deformation coupling. For these "special" grid regions, the usage data of all special grid regions is used as a historical overall usage data set, and the deformation data of all special grid regions is used as a historical overall deformation data set. By treating the coupled terrain as a whole rather than decoupling it, the correlation between the overall usage and overall deformation of these special grid regions is extracted. Then, future overall usage data is used to predict future overall deformation data and determine the corresponding risks. By fully considering the complex changing state of soft soil foundations and the complex operating conditions of ports, the predicted deformation data and deformation risks are more consistent with the actual changing state of soft soil foundations and the actual operating conditions of ports, resulting in more accurate prediction results.

[0093] In addition, this application provides two schemes for obtaining the correlation between historical overall usage data and historical overall deformation data. The first scheme uses a fitting formula, i.e., regression method, to determine the correlation between the two, and the second scheme uses a graph network to determine the correlation between the two. The first scheme is introduced first below.

[0094] Optionally, the usage data of the grid regions corresponding to all deformation vectors at the same time are taken as historical overall usage data, and the deformation data of the grid regions corresponding to all deformation vectors at the same time are taken as historical overall deformation data. The correlation between historical overall usage data and historical overall deformation data is obtained, including: combining the usage data of the grid regions corresponding to all deformation vectors at the same time into a first usage data matrix, and using the first usage data matrix as historical overall usage data; combining the deformation data of the grid regions corresponding to all deformation vectors at the same time into a first deformation data matrix, and using the first deformation data matrix as historical overall usage data; fitting multiple historical overall usage data and historical overall deformation data according to a first preset fitting formula; the fitted formula is the correlation.

[0095] Optionally, the risk level corresponding to the future overall deformation data is determined based on the correspondence between deformation data and risk level, including: splitting the future deformation data of each grid region from the future overall deformation data, and determining the risk level corresponding to the future deformation data of each grid region based on the correspondence between single deformation data and single risk level.

[0096] Optionally, when the first usage data matrix is ​​used as the historical overall usage data and the first deformation data matrix is ​​used as the historical overall deformation data, the correspondence between a single deformation data and a single risk level includes: each risk level corresponds to a range of deformation data.

[0097] Optionally, based on the correspondence between single deformation data and single risk level, the risk level corresponding to the future deformation data of each grid area is determined, including: matching the current deformation data range corresponding to the future deformation data of the grid area from multiple deformation data ranges; and using the risk level corresponding to the current deformation data range as the risk level of the grid area.

[0098] Optionally, based on the correspondence between single deformation data and single risk level, the risk level corresponding to the future deformation data of each grid area is determined, including: processing the future deformation data of the grid area according to a preset method, matching the current deformation data range corresponding to the processed future deformation data from multiple deformation data ranges; and using the risk level corresponding to the current deformation data range as the risk level of the grid area.

[0099] The algorithms corresponding to the above fitting formulas include, but are not limited to, one or a combination of the following: Fourier base-based fitting algorithms, sine function-based fitting methods, high-order polynomial-based fitting algorithms, Smooth fitting algorithms, least squares-based fitting algorithms, and nonlinear functions such as exponential and logarithmic fitting algorithms.

[0100] For example, risk levels can be classified as low risk, medium risk, high risk, and critical risk.

[0101] When the deformation data represents the first uneven settlement data of the stockpile, the numerical ranges corresponding to different risks are shown below:

[0102] Low risk corresponds to longitudinal uneven settlement difference of less than 12mm for a horizontal span of 1 to 10m;

[0103] Medium risk corresponds to a longitudinal uneven settlement difference of 12-25mm over a horizontal span of 1-10m;

[0104] High risk corresponds to a longitudinal uneven settlement difference of 25-50mm over a horizontal span of 1-10m;

[0105] Critical risks correspond to longitudinal uneven settlement differences of more than 50mm over a horizontal span of 1 to 10m.

[0106] The boundaries of the above-mentioned numerical ranges have overlapping numerical points. The technical solution of this application does not specifically limit which region the overlapping numerical points belong to. Those skilled in the art can place the overlapping numerical points in the numerical range corresponding to lower risk or in the numerical range corresponding to higher risk.

[0107] When the deformation data is lateral deformation data at the edge of the stockpile, the numerical ranges corresponding to different risks are as follows:

[0108] Low risk corresponds to a lateral deformation distance of less than 15mm for a longitudinal span of 2-5m;

[0109] Medium risk corresponds to a lateral deformation distance of 15-40mm over a longitudinal span of 2-5m;

[0110] High risk corresponds to a lateral deformation distance of 40-100mm with a longitudinal span of 2-5m;

[0111] Critical risks correspond to lateral deformation distances of over 100mm across a longitudinal span of 2–5m.

[0112] The boundaries of the above-mentioned numerical ranges have overlapping numerical points. The technical solution of this application does not specifically limit which region the overlapping numerical points belong to. Those skilled in the art can place the overlapping numerical points in the numerical range corresponding to lower risk or in the numerical range corresponding to higher risk.

[0113] When the deformation data is the second uneven settlement data of the stacker-reclaimer track cross-section, the numerical ranges corresponding to different risks are as follows:

[0114] Low risk corresponds to an uneven settlement difference of less than 6mm between the two rails;

[0115] Medium risk corresponds to an uneven settlement difference of 6-15mm between the two tracks;

[0116] High risk corresponds to an uneven settlement difference of 15-25mm between the two tracks;

[0117] A critical risk corresponds to an uneven settlement difference of more than 25mm between the two tracks.

[0118] The boundaries of the above-mentioned numerical ranges have overlapping numerical points. The technical solution of this application does not specifically limit which region the overlapping numerical points belong to. Those skilled in the art can place the overlapping numerical points in the numerical range corresponding to lower risk or in the numerical range corresponding to higher risk.

[0119] When the deformation data is the tilt data of high-mast lights, the numerical ranges corresponding to different risks are as follows:

[0120] Low risk corresponds to a tilt range of less than 0.2°;

[0121] Medium risk corresponds to a tilt range of 0.2° to 0.5°.

[0122] High risk corresponds to a tilt range of 0.5° to 1.0° or less;

[0123] Critical risk corresponds to a tilt range of 1.0° or higher.

[0124] The boundaries of the above-mentioned numerical ranges have overlapping numerical points. The technical solution of this application does not specifically limit which region the overlapping numerical points belong to. Those skilled in the art can place the overlapping numerical points in the numerical range corresponding to lower risk or in the numerical range corresponding to higher risk.

[0125] In cases of critical risk, the device should be immediately shut down and rectified; in cases of high risk, operation should be restricted and maintenance should be carried out; in cases of medium risk, calibration or leveling should be performed; and in cases of low risk, normal operation may be permitted.

[0126] The second approach is to use graph networks to determine the relationship between historical overall usage data and historical overall deformation data.

[0127] Optionally, the usage data of the grid regions corresponding to all deformation vectors at the same time are taken as historical overall usage data, and the deformation data of the grid regions corresponding to all deformation vectors at the same time are taken as historical overall deformation data. The correlation between historical overall usage data and historical overall deformation data is obtained, including:

[0128] The usage data of the grid region corresponding to each deformation vector at the same time is taken as a node. If two grid regions are adjacent in actual space, the nodes corresponding to the two grid regions are neighbor nodes. The graph network formed by the nodes and their adjacency relationships is updated once or multiple times to obtain the updated usage data of each node. The updated usage data of all nodes are combined into a second usage data matrix, and the second usage data matrix is ​​used as the historical overall usage data.

[0129] The deformation data of the grid region corresponding to each deformation vector at the same time is taken as a node. If two grid regions are adjacent in actual space, the nodes corresponding to the two grid regions are neighbor nodes. The graph network formed by the nodes and their adjacency relationships is updated once or multiple times to obtain the updated deformation data of each node. The updated deformation data of all nodes are combined into a second deformation data matrix, and the second deformation data matrix is ​​used as the historical overall deformation data.

[0130] The second preset fitting formula is used to fit multiple historical overall usage data and historical overall deformation data, and the fitted formula is the correlation relationship; or, the time series neural network is fine-tuned using multiple historical overall usage data and historical overall deformation data, and the fine-tuned time series neural network is the correlation relationship.

[0131] Optionally, the risk level corresponding to the future overall deformation data is determined based on the correspondence between deformation data and risk level, including: determining the overall risk level corresponding to the future overall deformation data based on the correspondence between overall deformation data and overall risk level; the overall risk level corresponding to the overall deformation data is used to represent the overall risk of the grid area corresponding to all deformation vectors within each cluster.

[0132] Optionally, when the second usage data matrix is ​​used as the historical overall usage data and the second deformation data matrix is ​​used as the historical overall deformation data, the correspondence between the overall deformation data and the overall risk level includes: each overall risk level corresponds to a range of mean deformation data.

[0133] Optionally, based on the correspondence between overall deformation data and overall risk level, the overall risk level corresponding to future overall deformation data is determined, including: obtaining the mean of future overall deformation data; matching the current mean data range corresponding to the mean of future overall deformation data within multiple mean deformation data ranges; and determining the overall risk level of multiple grid regions corresponding to clusters based on the current mean data range.

[0134] The graph network mentioned above refers to Graph Neural Networks (GNNs), which have the ability to extract features from a holistic perspective. The "one or more updates" mentioned refers to updating node data in a GNN, where the update method is to update the current node using the data of its neighbors based on the similarity between the current node and its neighbors. The number of updates is generally limited to one, two, or three.

[0135] In this embodiment, the usage data of the grid regions corresponding to all deformation vectors within a cluster at the same time are used as the historical overall usage data, and the deformation data of the grid regions corresponding to all deformation vectors within a cluster at the same time are used as the historical overall deformation data. The historical overall deformation data is extracted using a graph network, which can reflect the coupling relationship of the grid regions corresponding to all deformation vectors within a cluster. In addition, when there is coupling between the grid regions corresponding to all deformation vectors within a cluster, the usage data in a single grid region not only affects the deformation of that grid region, but also affects the deformation of other grid regions. Therefore, by using a graph network to extract the historical overall deformation data, the usage data is coupled before it is applied to the coupled deformation data, which can avoid the complex multi-angle analysis of the usage data of multiple single grid regions being applied to the coupled grid regions.

[0136] In addition, the deformation after coupling of multiple adjacent grid regions is usually caused by a large load and long load residence time in a single grid region. The deformation trend shows the phenomenon of radiation influence from a single grid region to the surrounding adjacent grid regions. The update process of the graph network is the process of updating the data of the current node using the data of the adjacent nodes. This is similar to the deformation trend after coupling of multiple adjacent grid regions. Therefore, compared with other schemes, the historical overall deformation data extracted by the graph network can more accurately reflect the coupling characteristics of multiple adjacent grid regions.

[0137] The aforementioned temporal neural network refers to LSTM and its deformation network. During the use of soft soil foundations in ports, the external deformation corresponds to changes in the internal structure of the soil. That is, the input and output of the previous moment have an impact on the input and output of the next moment. Therefore, temporal neural networks can be used to represent the correlation between historical overall usage data and historical overall deformation data.

[0138] The overall risk level mentioned above corresponds to a mean deformation data range, which can be obtained in the following way: for each type of risk, the numerical ranges corresponding to the first uneven settlement data of the stockpile, the lateral deformation data of the edge of the stockpile, and the second uneven settlement data of the cross section of the stockpile-reclaimer track in the aforementioned embodiment are converted into a gradient, that is, the deformation amount per unit horizontal span or per unit longitudinal span.

[0139] Next, the tilt data of the three gradients and high-mast lights were normalized, and weights were assigned to the normalized three gradients and tilt data according to the degree of influence of data type on usage risk. The weighted sum of the four was used as the mean deformation value range of different risks.

[0140] Figure 2 This is a flowchart illustrating a risk prediction method for soft soil foundation scenarios provided in an embodiment of this application. This risk prediction method can be executed via a local computer or a cloud server.

[0141] Combination Figure 2 As shown, risk prediction methods for soft soil foundation scenarios include:

[0142] S201. Obtain deformation data and usage data at several sampling times within several grid areas.

[0143] The grid areas are pre-divided, and some grid areas cover part or all of the port infrastructure. The infrastructure covered by some grids includes one or more of the following: yard area, yard edge, stacker-reclaimer track and high mast lights, but not only high mast lights.

[0144] S202. For the deformation data of each grid region, calculate the data difference between every two adjacent deformation data according to the order of sampling time.

[0145] S203. Combine the data difference from a specific historical moment to the current moment and the grid position coordinates within each grid region into a deformation vector.

[0146] S204. Cluster multiple deformed vectors to obtain one or more clusters.

[0147] S205. The grid regions corresponding to one or more deformation vectors that are closest to the cluster center in a cluster are designated as high-risk grid regions.

[0148] S206. Obtain the correlation between historical overall usage data and historical overall deformation data.

[0149] According to the aforementioned embodiments, this step has two schemes: one is to extract the correlation between historical overall usage data and historical overall deformation data based on the fitting formula, and the other is to extract the correlation between historical overall usage data and historical overall deformation data based on the graph network.

[0150] S207. Determine future overall usage data based on the changing trends of historical overall usage data.

[0151] S208. Determine the future overall deformation data corresponding to the future overall usage data based on the correlation relationship.

[0152] S209. Determine the risk level corresponding to the overall future deformation data based on the correspondence between deformation data and risk level.

[0153] The future overall deformation data in this step is the same as the future overall deformation data determined based on the graph network mentioned above.

[0154] That is, when confirming the risk level in this step, a scheme based on graph networks is adopted to extract the correlation between historical overall usage data and historical overall deformation data.

[0155] S210. Based on the correspondence between deformation data and risk levels, determine the risk level corresponding to high-risk areas.

[0156] The process for determining this step corresponds to the aforementioned process of determining the correlation between historical overall usage data and historical overall deformation data based on the fitting formula. In other words, when confirming the risk level in this step, a scheme based on the fitting formula to extract the correlation between historical overall usage data and historical overall deformation data is adopted.

[0157] For example, after obtaining the overall future deformation data, the future deformation data of high-risk grid areas is extracted from the overall future deformation data. Based on the correspondence between individual deformation data and individual risk levels, the risk level corresponding to the future deformation data of high-risk grid areas is determined.

[0158] S211. Determine the final risk level based on the risk level corresponding to the overall future deformation data and the risk level corresponding to the high-risk area.

[0159] For example, the final risk level can be an array composed of the risk level corresponding to the overall future deformation data and the risk level corresponding to the high-risk area. The risk level corresponding to the overall future deformation data represents the overall risk level, while the risk level corresponding to the high-risk area represents the maximum risk level.

[0160] In another feasible approach, a high-risk weight is set based on the degree of risk preference, and the overall risk weight is determined based on the high-risk weight, with the sum of the high-risk weight and the overall risk weight being 1; or, the overall risk weight is set based on the degree of risk preference, and the high-risk weight is determined based on the high-risk weight, with the sum of the overall risk weight and the high-risk weight being 1.

[0161] Among them, the more users value high risk, the greater the weight of high risk; the more users value overall risk, the higher the weight of overall risk.

[0162] Alternatively, the weights of the overall risk level corresponding to future deformation data and the risk level corresponding to high-risk areas can be determined separately. The higher the risk level corresponding to the overall future deformation data, the greater the overall risk weight; the higher the risk level corresponding to the high-risk area, the greater the high-risk weight. The weighted sum of the risk levels corresponding to the overall future deformation data and the high-risk areas is then used to obtain the final risk level. Monitoring and maintaining the port based on this final risk level is more conducive to the safe operation of the port.

[0163] The sampling process for each deformation data will be illustrated below.

[0164] The principle of monitoring uneven settlement in a stockpile is as follows: The change in the sine or tangent of the angle within the conduit, measured by an angle sensor, can represent the change in vertical quantity within a finite length, accurately measuring the angle of deflection of an object. When the sensor is in a horizontal position, it measures the corresponding capacitance between the electrodes. When the sensor tilts, the elastic electrode changes its relative position to the fixed electrode, and the capacitance measured by the sensor unit between the two electrodes also changes accordingly. This change in capacitance is converted into a corresponding tilt value.

[0165] Figure 3 This is a schematic diagram illustrating the principle of lateral deformation monitoring of soil at the edge of a stockpile, provided in an embodiment of this application.

[0166] refer to Figure 3 The principle of monitoring lateral deformation of the soil at the edge of the stockpile is as follows: a clinometer tube, divided into several sections, is pre-embedded in the soil and connected to each other with joints, allowing the entire clinometer tube to deform with the foundation. After deformation, each section of the clinometer tube has a tilt angle α. i The change in inclination produces a corresponding horizontal displacement difference δ at both ends of the inclinometer tube. i ,have:

[0167] ;

[0168] In the formula: Li represents the length of the i-th section of the inclinometer tube, taken as the center-to-center distance between two adjacent ends, and Li is usually taken as an integer. The difference in horizontal displacement of the soil at both ends of the entire inclinometer tube can be expressed as:

[0169] ;

[0170] The measurement results are compiled into a horizontal displacement change curve to reflect the horizontal displacement of each soil layer.

[0171] High mast light tilt monitoring principle: A bidirectional inclinometer is used to record the current state of the pole in real time through automatic acquisition and wireless transmission. When the pole tilts and deforms, the tilt change is transmitted to the inclinometer's built-in sensor. The tilt angle corresponds to the output power. The tilt angle of the measured structure is calculated by converting the power change. At the same time, the tilt angle and the positive and negative directions of the change are displayed with the zero point as the reference value.

[0172] Figure 4 This is a schematic diagram illustrating the principle of monitoring track deformation of a stacker-reclaimer provided in an embodiment of this application.

[0173] refer to Figure 4 The principle of monitoring the deformation of the stacker-reclaimer track is as follows: tilt sensors, displacement sensors and connecting pipe shafts are installed at the same cross-section of the track.

[0174] Working principle of uneven settlement between two tracks: The distance between two adjacent tilt sensors is... When track sections settle, the connecting tube shafts deflect. By observing the changes in the inclination angles of each section, the track settlement can be calculated. After deformation, each tube shaft section has an inclination angle θ. The change in inclination angle produces a corresponding vertical displacement difference Δ at both ends of the tube shaft. i ,have

[0175] ;

[0176] in: - indicates the distance between the two tilt sensors in the i-th section.

[0177] The difference in axial displacement between the two rails on the same cross section can be expressed as:

[0178] ;

[0179] Taking one of the tracks as a reference, δ is the uneven settlement value between the two tracks on the same cross section.

[0180] Figure 5 This is a schematic diagram of a risk prediction device for soft soil foundation scenarios provided in an embodiment of this application. The risk prediction device can be implemented through software, hardware, or a combination of both.

[0181] Combination Figure 5As shown, the risk prediction device for soft soil foundation scenarios includes a first acquisition module 51, a difference calculation module 52, a combination module 53, a clustering module 54, a second acquisition module 55, a first determination module 56, a second determination module 57, and a first risk module 58.

[0182] The first acquisition module 51 is used to acquire deformation data and usage data at several sampling times within several grid areas; wherein, the grid areas are pre-divided, and the several grid areas cover part or all of the port infrastructure. The infrastructure covered by the several grids includes one or more of the following: yard area, yard edge, stacker-reclaimer track, and high mast lights, but will not include only high mast lights; the deformation data is the deformation data corresponding to the infrastructure covered in each grid area; the usage data is load-related data.

[0183] The difference calculation module 52 is used to calculate the data difference between every two adjacent deformation data points based on the order of sampling time for the deformation data of each grid region.

[0184] The combination module 53 is used to combine the data difference from a specific historical time to the current time and the grid position coordinates within each grid region into a deformation vector;

[0185] Clustering module 54 is used to cluster multiple deformed vectors to obtain one or more clusters; the clustering effect at a specific historical moment is better than the clustering effect at other historical moments;

[0186] The second acquisition module 55 is used to take the usage data of the grid area corresponding to all deformation vectors at the same time as the historical overall usage data, and take the deformation data of the grid area corresponding to all deformation vectors at the same time as the historical overall deformation data, and acquire the correlation between the historical overall usage data and the historical overall deformation data.

[0187] The first determining module 56 is used to determine future overall usage data based on the changing trends of historical overall usage data;

[0188] The second determining module 57 is used to determine the future overall deformation data corresponding to the future overall usage data based on the correlation relationship;

[0189] The first risk module 58 is used to determine the risk level corresponding to the overall future deformation data based on the correspondence between deformation data and risk level.

[0190] Optionally, the second acquisition module 55 is specifically used for: combining the usage data of the grid regions corresponding to all deformation vectors at the same time into a first usage data matrix, and using the first usage data matrix as the historical overall usage data; combining the deformation data of the grid regions corresponding to all deformation vectors at the same time into a first deformation data matrix, and using the first deformation data matrix as the historical overall usage data; fitting multiple historical overall usage data and historical overall deformation data according to a first preset fitting formula; the fitted formula is the correlation relationship.

[0191] Optionally, the first risk module 58 is specifically used to: extract the future deformation data of each grid region from the overall future deformation data, and determine the risk level corresponding to the future deformation data of each grid region based on the correspondence between single deformation data and single risk level.

[0192] Optionally, the second acquisition module 55 is specifically used to: take the usage data of the grid region corresponding to each deformation vector at the same time as a node; if two grid regions are adjacent in actual space, then the nodes corresponding to the two grid regions are neighbor nodes; update the graph network formed by the nodes and their adjacency relationships once or multiple times to obtain the updated usage data of each node; combine the updated usage data of all nodes into a second usage data matrix; and use the second usage data matrix as the historical overall usage data.

[0193] The deformation data of the grid region corresponding to each deformation vector at the same time is taken as a node. If two grid regions are adjacent in actual space, the nodes corresponding to the two grid regions are neighbor nodes. The graph network formed by the nodes and their adjacency relationships is updated once or multiple times to obtain the updated deformation data of each node. The updated deformation data of all nodes are combined into a second deformation data matrix, and the second deformation data matrix is ​​used as the historical overall deformation data.

[0194] The second preset fitting formula is used to fit multiple historical overall usage data and historical overall deformation data, and the fitted formula is the correlation relationship; or, the time series neural network is fine-tuned using multiple historical overall usage data and historical overall deformation data, and the fine-tuned time series neural network is the correlation relationship.

[0195] Optionally, the first risk module 58 is specifically used to: determine the overall risk level corresponding to the future overall deformation data based on the correspondence between the overall deformation data and the overall risk level; the overall risk level corresponding to the overall deformation data is used to represent the overall risk of the grid area corresponding to all deformation vectors within each cluster.

[0196] Optionally, when the first usage data matrix is ​​used as the historical overall usage data and the first deformation data matrix is ​​used as the historical overall deformation data, the correspondence between a single deformation data and a single risk level includes: each risk level corresponds to a range of deformation data.

[0197] Optionally, based on the correspondence between single deformation data and single risk level, the risk level corresponding to the future deformation data of each grid area is determined, including: matching the current deformation data range corresponding to the future deformation data of the grid area from multiple deformation data ranges, or processing the future deformation data of the grid area according to a preset method, and matching the current deformation data range corresponding to the processed future deformation data from multiple deformation data ranges; using the risk level corresponding to the current deformation data range as the risk level of the grid area.

[0198] Optionally, when the second usage data matrix is ​​used as the historical overall usage data and the second deformation data matrix is ​​used as the historical overall deformation data, the correspondence between the overall deformation data and the overall risk level includes: each overall risk level corresponds to a range of mean deformation data.

[0199] Optionally, based on the correspondence between overall deformation data and overall risk level, the overall risk level corresponding to future overall deformation data is determined, including: obtaining the mean of future overall deformation data; matching the current mean data range corresponding to the mean of future overall deformation data within multiple mean deformation data ranges; and determining the overall risk level of multiple grid regions corresponding to clusters based on the current mean data range.

[0200] Optionally, the risk prediction device for soft soil foundation scenarios also includes an extraction module. The extraction module, after clustering multiple deformation vectors to obtain one or more clusters, identifies the grid regions corresponding to the one or more deformation vectors closest to the cluster center within each cluster as high-risk grid regions.

[0201] Optionally, the risk prediction device for soft soil foundation scenarios further includes a second risk module and a third risk module. The second risk module is used to determine the risk level corresponding to the overall future deformation data based on the correspondence between deformation data and risk levels, and then, based on the same correspondence, to determine the risk level corresponding to the high-risk area. The third risk module is used to determine the final risk level based on the risk level corresponding to the overall future deformation data and the risk level corresponding to the high-risk area.

[0202] Optionally, the process of determining a specific historical moment includes: acquiring several historical moments; for each historical moment, combining the data difference from the historical moment to the current moment and the grid position coordinates within each grid region into a deformation vector, performing clustering processing on multiple deformation vectors, evaluating the clustering effect, and selecting the historical moment with the best clustering effect as the specific historical moment.

[0203] Alternatively, the process of determining a specific historical moment includes: acquiring a first historical moment; combining the data difference from the first historical moment to the current moment and the grid position coordinates within each grid region into a deformed vector; performing clustering processing on multiple deformed vectors and evaluating the clustering effect; acquiring a second historical moment before the first historical moment and a third historical moment after the first historical moment, and sequentially acquiring the clustering effect corresponding to the second historical moment and the third historical moment; if the clustering effect of the second historical moment is better than the clustering effect corresponding to the first historical moment and the third historical moment, then continuing to acquire other historical moments before the second historical moment and acquiring the clustering effect corresponding to other historical moments, and so on, until the clustering effect corresponding to a newly acquired historical moment no longer becomes better, and the historical moment corresponding to the best clustering effect is taken as the specific historical moment; if the clustering effect of the third historical moment is better than the clustering effect corresponding to the first historical moment and the second historical moment, then continuing to acquire other historical moments after the third historical moment and acquiring the clustering effect corresponding to other historical moments, and so on, until the clustering effect corresponding to a newly acquired historical moment no longer becomes better, and the historical moment corresponding to the best clustering effect is taken as the specific historical moment.

[0204] Optionally, when the infrastructure is a stockpile area, the deformation data is the first uneven settlement data of the stockpile; when the infrastructure is the edge of the stockpile, the deformation data is the lateral deformation data of the edge of the stockpile; when the infrastructure is a stacker-reclaimer track, the deformation data is the second uneven settlement data of the cross section of the stacker-reclaimer track; and when the infrastructure is a high-mast light, the deformation data is the tilt data of the high-mast light.

[0205] Optionally, the load data may include one or more of the following: load quantity, load dwell time, and load change rate in the current or adjacent grid area.

[0206] Optionally, the risk prediction device for soft soil foundation scenarios also includes a normalization module. The normalization module is used to normalize the deformation data corresponding to different types of infrastructure before calculating the data difference between any two adjacent deformation data when several grid areas cover two or more types of infrastructure.

[0207] In some embodiments, the risk prediction apparatus for soft soil foundation scenarios includes a processor and a memory storing program instructions. The processor is configured to execute the risk prediction method for soft soil foundation scenarios provided in the foregoing embodiments when executing the program instructions.

[0208] In some embodiments, the risk prediction system for soft soil foundation scenarios includes the risk prediction device for soft soil foundation scenarios provided in the foregoing embodiments.

[0209] Figure 6 This is a schematic diagram of a risk prediction device for soft soil foundation scenarios provided in an embodiment of this application.

[0210] Combination Figure 6 As shown, the risk prediction device for soft soil foundation scenarios includes:

[0211] The processor 61 and memory 62 may also include a communication interface 63 and a bus 64. The processor 61, communication interface 63, and memory 62 can communicate with each other via the bus 64. The communication interface 63 can be used for information transmission. The processor 61 can call logical instructions in the memory 62 to execute the risk prediction method for soft soil foundation scenarios provided in the foregoing embodiments.

[0212] Furthermore, the logical instructions in the aforementioned memory 62 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium.

[0213] The memory 62, as a computer-readable storage medium, can be used to store software programs and computer-executable programs, such as program instructions / modules corresponding to the methods in the embodiments of this application. The processor 61 executes functional applications and data processing by running the software programs, instructions, and modules stored in the memory 62, thereby implementing the methods in the above-described method embodiments.

[0214] The memory 62 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the terminal device. Furthermore, the memory 62 may include high-speed random access memory and may also include non-volatile memory.

[0215] This application provides a computer-readable storage medium storing computer-executable instructions configured to execute the risk prediction method for soft soil foundation scenarios provided in the foregoing embodiments.

[0216] This application provides a computer program product, which includes a computer program stored on a computer-readable storage medium. The computer program includes program instructions, which, when executed by a computer, cause the computer to perform the risk prediction method for soft soil foundation scenarios provided in the foregoing embodiments.

[0217] The aforementioned computer-readable storage medium may be a transient computer-readable storage medium or a non-transitory computer-readable storage medium.

[0218] The technical solutions of this application embodiment can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes one or more instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods in this application embodiment. The aforementioned storage medium can be a non-transitory storage medium, including: USB flash drive, portable hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk, and other media capable of storing program code; it can also be a transient storage medium.

[0219] The foregoing description and accompanying drawings fully illustrate embodiments of this application to enable those skilled in the art to practice them. Other embodiments may include structural, logical, electrical, procedural, and other changes. The embodiments represent only possible variations. Individual components and functions are optional unless explicitly required, and the order of operations may vary. Parts and features of some embodiments may be included in or replace parts and features of other embodiments. Moreover, the terminology used in this application is for describing embodiments only and is not intended to limit the claims. As used in the description of embodiments and claims, the singular forms “a,” “an,” and “the” are intended to equally include the plural forms unless the context clearly indicates otherwise. Additionally, when used in this application, the terms “comprise” and its variations “comprises” and / or “comprising” refer to the presence of stated features, integrals, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or groups thereof. Unless otherwise specified, an element defined by the phrase "comprising a..." does not exclude the presence of other identical elements in the process, method, or apparatus that includes that element. In this document, each embodiment may focus on describing the differences from other embodiments, and similar or identical parts between embodiments can be referred to mutually. For methods, products, etc., disclosed in the embodiments, if they correspond to the method section disclosed in the embodiments, then the relevant parts can be referred to the description of the method section.

[0220] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the embodiments of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0221] The methods and products (including but not limited to devices and equipment) disclosed in the embodiments herein can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of units may be merely a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the shown or discussed units may be through some interfaces, and the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms. Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, i.e., they may be located in one place or distributed across multiple network units. Some or all of the units may be selected to implement this embodiment according to actual needs. Furthermore, the functional units in the embodiments of this application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0222] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. Each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

Claims

1. A risk prediction method for soft soil foundation scenarios, characterized in that, include: Deformation data and usage data at several sampling times within several grid areas are acquired; the grid areas are pre-divided, and the several grid areas cover part or all of the port infrastructure. The infrastructure covered by the several grids includes one or more of the following: yard area, yard edge, stacker-reclaimer track, and high mast lights, but not only high mast lights; the deformation data is the deformation data corresponding to the infrastructure covered in each grid area; the usage data is load-related data. For the deformation data of each grid region, the data difference between every two adjacent deformation data points is calculated according to the order of sampling time; The data difference from a specific historical moment to the current moment and the grid position coordinates within each grid region are combined into a deformation vector; Multiple deformed vectors are clustered to obtain one or more clusters; the clustering effect at a specific historical moment is better than the clustering effect at other historical moments; The usage data of the grid region corresponding to all deformation vectors within the cluster at the same time is taken as the historical overall usage data, and the deformation data of the grid region corresponding to all deformation vectors within the cluster at the same time is taken as the historical overall deformation data. The correlation between the historical overall usage data and the historical overall deformation data is obtained. Based on the changing trends of the historical overall usage data, determine the future overall usage data; Based on the aforementioned correlation, determine the future overall deformation data corresponding to the future overall usage data; The risk level corresponding to the future overall deformation data is determined based on the correspondence between deformation data and risk level.

2. The risk prediction method for soft soil foundation scenarios according to claim 1, characterized in that, The process involves taking the usage data of the grid regions corresponding to all deformation vectors within a cluster at the same time as the historical overall usage data, and taking the deformation data of the grid regions corresponding to all deformation vectors within a cluster at the same time as the historical overall deformation data. The process then obtains the correlation between the historical overall usage data and the historical overall deformation data, including: combining the usage data of the grid regions corresponding to all deformation vectors within a cluster at the same time into a first usage data matrix, and using the first usage data matrix as the historical overall usage data; combining the deformation data of the grid regions corresponding to all deformation vectors within a cluster at the same time into a first deformation data matrix, and using the first deformation data matrix as the historical overall usage data; fitting multiple sets of historical overall usage data and historical overall deformation data according to a first preset fitting formula; and the fitted formula is the correlation. Determining the risk level corresponding to the future overall deformation data based on the correspondence between deformation data and risk level includes: splitting the future deformation data of each grid region from the future overall deformation data, and determining the risk level corresponding to the future deformation data of each grid region based on the correspondence between single deformation data and single risk level; or, Using the usage data of the grid regions corresponding to all deformation vectors within a cluster at the same time as the historical overall usage data, and using the deformation data of the grid regions corresponding to all deformation vectors within a cluster at the same time as the historical overall deformation data, the correlation between the historical overall usage data and the historical overall deformation data is obtained, including: The usage data of the grid region corresponding to each deformation vector at the same time is taken as a node. If two grid regions are adjacent in actual space, the nodes corresponding to the two grid regions are neighbor nodes. The graph network formed by the nodes and their adjacency relationships is updated once or multiple times to obtain the updated usage data of each node. The updated usage data of all nodes are combined into a second usage data matrix, and the second usage data matrix is ​​used as the historical overall usage data. The deformation data of the grid region corresponding to each deformation vector at the same time is taken as a node. If two grid regions are adjacent in actual space, the nodes corresponding to the two grid regions are neighbor nodes. The graph network formed by the nodes and their adjacency relationships is updated once or multiple times to obtain the updated deformation data of each node. The updated deformation data of all nodes are combined into a second deformation data matrix. The second deformation data matrix is ​​used as the historical overall deformation data. The correlation relationship is obtained by fitting multiple historical overall usage data and historical overall deformation data according to the second preset fitting formula; or, the correlation relationship is obtained by fine-tuning the temporal neural network using multiple historical overall usage data and historical overall deformation data. Determining the risk level corresponding to the future overall deformation data based on the correspondence between deformation data and risk level includes: determining the overall risk level corresponding to the future overall deformation data based on the correspondence between overall deformation data and overall risk level; the overall risk level corresponding to the overall deformation data is used to represent the overall risk of the grid area corresponding to all deformation vectors within each cluster.

3. The risk prediction method for soft soil foundation scenarios according to claim 2, characterized in that, When the first usage data matrix is ​​used as the overall historical usage data and the first deformation data matrix is ​​used as the overall historical deformation data, the correspondence between single deformation data and single risk level includes: each risk level corresponds to a range of deformation data. Based on the correspondence between single deformation data and single risk level, the risk level corresponding to the future deformation data of each grid area is determined, including: matching the current deformation data range corresponding to the future deformation data of the grid area from multiple deformation data ranges, or processing the future deformation data of the grid area according to a preset method, and matching the current deformation data range corresponding to the processed future deformation data from multiple deformation data ranges; the risk level corresponding to the current deformation data range is used as the risk level of the grid area. or, When the second usage data matrix is ​​used as the historical overall usage data and the second deformation data matrix is ​​used as the historical overall deformation data, the correspondence between the overall deformation data and the overall risk level includes: each overall risk level corresponds to a range of mean deformation data. Based on the correspondence between overall deformation data and overall risk level, the overall risk level corresponding to the future overall deformation data is determined, including: obtaining the mean of the future overall deformation data; matching the current mean data range corresponding to the mean of the future overall deformation data within multiple mean deformation data ranges; and determining the overall risk level of multiple grid regions corresponding to the cluster based on the current mean data range.

4. The risk prediction method for soft soil foundation scenarios according to any one of claims 1 to 3, characterized in that, After clustering multiple deformation vectors to obtain one or more clusters, the risk prediction method further includes: taking the grid region corresponding to one or more deformation vectors closest to the cluster center in the cluster as the high-risk grid region; After determining the risk level corresponding to the future overall deformation data based on the correspondence between deformation data and risk levels, the risk prediction method further includes: determining the risk level corresponding to the high-risk grid area based on the correspondence between deformation data and risk levels; and determining the final risk level based on the risk level corresponding to the future overall deformation data and the risk level corresponding to the high-risk grid area.

5. The risk prediction method for soft soil foundation scenarios according to claim 1, characterized in that, The process of determining the specific historical moment includes: Acquire several historical moments; for each historical moment, combine the data difference from the historical moment to the current moment and the grid position coordinates within each grid region into a deformation vector, perform clustering processing on multiple deformation vectors, evaluate the clustering effect, and select the historical moment with the best clustering effect as the specific historical moment; or, A first historical moment is obtained. The data difference from the first historical moment to the current moment within each grid region, along with the grid position coordinates, is combined into a deformation vector. Multiple deformation vectors are clustered, and the clustering effect is evaluated. A second historical moment before the first historical moment and a third historical moment after the first historical moment are obtained, and the clustering effect corresponding to the second and third historical moments is obtained sequentially. If the clustering effect of the second historical moment is better than the clustering effects corresponding to the first and third historical moments, other historical moments before the second historical moment are obtained, and the clustering effects corresponding to these other historical moments are obtained. This process is repeated until the clustering effect corresponding to a newly obtained historical moment no longer improves. The historical moment corresponding to the optimal clustering effect is then taken as the specific historical moment. If the clustering effect of the third historical moment is better than the clustering effects corresponding to the first and second historical moments, other historical moments after the third historical moment are obtained, and the clustering effects corresponding to these other historical moments are obtained. This process is repeated until the clustering effect corresponding to a newly obtained historical moment no longer improves. The historical moment corresponding to the optimal clustering effect is then taken as the specific historical moment.

6. The risk prediction method for soft soil foundation scenarios according to claim 1, characterized in that, When the infrastructure is a stockpile area, the deformation data is the first uneven settlement data of the stockpile; when the infrastructure is the edge of the stockpile, the deformation data is the lateral deformation data of the edge of the stockpile; when the infrastructure is a stacker-reclaimer track, the deformation data is the second uneven settlement data of the cross section of the stacker-reclaimer track; when the infrastructure is a high-mast light, the deformation data is the tilt data of the high-mast light. And / or, Load data includes one or more of the following: load quantity, load dwell time, and load change rate in the current or adjacent grid area.

7. The risk prediction method for soft soil foundation scenarios according to claim 1 or 6, characterized in that, When the grid regions cover two or more types of infrastructure, the deformation data corresponding to different types of infrastructure are normalized before calculating the data difference between every two adjacent deformation data.

8. A risk prediction device for soft soil foundation scenarios, characterized in that, include: The first acquisition module is used to acquire deformation data and usage data at several sampling times within several grid areas; the grid areas are pre-divided, and the several grid areas cover part or all of the port infrastructure. The infrastructure covered by the several grids includes one or more of the following: yard area, yard edge, stacker-reclaimer track, and high mast lights, but not only high mast lights; the deformation data is the deformation data corresponding to the infrastructure covered in each grid area; the usage data is load-related data. The difference calculation module is used to calculate the data difference between every two adjacent deformation data points based on the order of sampling time for the deformation data of each grid region. The combination module is used to combine the data difference from a specific historical time to the current time and the grid position coordinates within each grid region into a deformation vector; The clustering module is used to cluster multiple deformed vectors to obtain one or more clusters; the clustering effect at a specific historical moment is better than the clustering effect at other historical moments; The second acquisition module is used to take the usage data of the grid area corresponding to all deformation vectors in the cluster at the same time as the historical overall usage data, and take the deformation data of the grid area corresponding to all deformation vectors in the cluster at the same time as the historical overall deformation data, and acquire the correlation between the historical overall usage data and the historical overall deformation data. The first determining module is used to determine future overall usage data based on the changing trends of the historical overall usage data; The second determining module is used to determine the future overall deformation data corresponding to the future overall usage data based on the association relationship; The first risk module is used to determine the risk level corresponding to the future overall deformation data based on the correspondence between deformation data and risk level.

9. A risk prediction device for soft soil foundation scenarios, comprising a processor and a memory storing program instructions, characterized in that, The processor is configured to execute the risk prediction method for soft soil foundation scenarios as described in any one of claims 1 to 7 when executing the program instructions.

10. A risk prediction system for soft soil foundation scenarios, characterized in that, Includes the risk prediction device for soft soil foundation scenarios as described in claim 8 or 9.

Images