A method for predicting settlement of a multi-layered soft soil foundation

Abstract

The application relates to the technical field of foundation settlement prediction, in particular to a multilayer soft soil foundation settlement prediction method, which comprises the following steps: analyzing real-time fluctuation of a settlement amount to identify a settlement jump time; constructing time dispersion degree according to the confusion degree of time interval distribution between all adjacent settlement jump times; comparing the difference of the settlement amount of a current point and each adjacent point at any soil layer at the settlement jump time to construct a settlement difference degree; forming a sequence of the settlement difference degrees of the current point and each adjacent point at all depths, analyzing the difference to quantify the local difference of the soil layer settlement trend, combining the settlement difference degree and the time dispersion degree to generate a settlement state comprehensive evaluation value, and adopting a smoothing algorithm to predict the settlement of the multilayer soft soil foundation. The application solves the problem of prediction deviation caused by geological anomalies and improves the accuracy of foundation settlement prediction.

Description

Technical Field

[0001] This application relates to the field of foundation settlement prediction technology, specifically to a method for predicting settlement of multi-layer soft soil foundations. Background Technology

[0002] Currently, multi-layered soft soil foundations are widely distributed in coastal, riverside, and mountainous areas, where major projects such as high-speed railways, cross-sea bridges, and airport runways are often constructed. Due to the significant settlement problem of soft soil foundations, safety hazards such as building cracking, structural tilting, and bridge misalignment are easily triggered, significantly increasing maintenance costs. Therefore, settlement prediction for multi-layered soft soil foundations is crucial for ensuring project safety.

[0003] In the prediction of settlement in multi-layered soft soil foundations, uneven soil layer distribution and increased random loads caused by external disturbances (such as heavy vehicles and construction) can lead to geological anomalies and drastic fluctuations in local settlement rates. Furthermore, the settlement impact area can spread outwards from the load center. In this situation, the degree of disturbance and data reliability vary depending on the location of the monitoring point within the anomaly area. Directly using traditional smoothing algorithms for prediction can easily result in serious biases due to neglecting the spatial variations in data reliability. Summary of the Invention

[0004] To address the aforementioned technical problems, this application provides a method for predicting settlement of multi-layer soft soil foundations, thereby resolving existing issues.

[0005] The settlement prediction method for multi-layer soft soil foundations proposed in this application adopts the following technical solution:

[0006] One embodiment of this application provides a method for predicting settlement of multi-layer soft soil foundations, the method comprising the following steps:

[0007] Monitoring points were set up in the multi-layered soft soil foundation area, and the settlement of the current point at different soil depths over time was obtained using a layered settlement gauge.

[0008] Before the current moment, for any soil layer at the current location, analyze the real-time fluctuation of settlement to identify the settlement jump moment; based on the degree of disorder in the distribution of time intervals between all adjacent settlement jump moments, construct a time dispersion to characterize the settlement jump time distribution law of the soil layer itself.

[0009] At the moment of settlement jump, the difference in settlement between the current point and its neighboring points at any soil layer is compared to construct the settlement difference degree between the current point and its neighboring points at any soil layer, which is used to reflect whether the current point and its neighboring points have undergone a sudden change at the same time.

[0010] The settlement difference between the current point and any of its nearest neighbor points at all soil depths is used to form a settlement difference sequence. By analyzing the differences between the settlement difference sequences corresponding to different nearest neighbor points, the local differences in the settlement trend of the soil layer at the current point are quantified. Combined with the settlement difference and the time dispersion, a comprehensive evaluation value of the settlement state of any soil layer at the current point is generated.

[0011] Based on the comprehensive settlement status assessment value, a smoothing algorithm is used to predict the settlement of multi-layer soft soil foundations.

[0012] Preferably, the method for identifying the settlement jump moment includes:

[0013] Before the current moment, for any soil layer at the current point, calculate the backward difference value of settlement at all sampling moments within the preset time period, and use it as the input of the threshold segmentation algorithm to output the segmentation threshold. The sampling moment corresponding to the backward difference value of settlement that is greater than the segmentation threshold is taken as the settlement jump moment.

[0014] Preferably, the method for constructing the time discreteness is as follows:

[0015] Divide all time intervals between all adjacent settlement jump moments of any soil layer at the current location within a preset time period before the current moment into multiple equal-width intervals. Count the frequency of time intervals falling into each interval. Take the ratio of the frequency of each interval to the total number of all time intervals as the probability value of each interval. The probability values ​​of all intervals constitute a probability distribution sequence. Calculate the information entropy of the probability distribution sequence as the time dispersion of any soil layer at the current location at the current moment.

[0016] Preferably, the nearest neighbor points of each point are the first preset number of points in the ascending order of the distances from each point to all other points in the multi-layer soft soil foundation area.

[0017] Preferably, the process of constructing the settlement difference between the current point and its nearest neighbor points at any soil layer is as follows:

[0018] The settlement amounts at all settlement jump moments below any soil layer at each point are compiled into a settlement sequence; the difference between the settlement sequence of the current point and its nearest neighbor points below any soil layer is taken as the settlement difference degree between the current point and its nearest neighbor points at any soil layer.

[0019] Preferably, the quantification process of the local differences in the soil settlement trend at the current location is as follows:

[0020] For the current location, the mean of the differences between the settlement difference sequences corresponding to all its nearest neighbor locations is used as the quantitative result of the local difference in the settlement trend of the soil layer under the current location.

[0021] Preferably, the settlement difference is arranged in a settlement difference sequence according to the soil layers from top to bottom.

[0022] Preferably, the comprehensive evaluation value of the settlement state of any soil layer at the current location is positively correlated with the quantitative result of the local difference in the settlement trend of the soil layer at the current location and the degree of settlement difference, and negatively correlated with the time dispersion.

[0023] Preferably, the step of generating a comprehensive assessment value of the settlement state of any soil layer at the current location further includes:

[0024] Comprehensive assessment value of settlement state of the i-th soil layer at the current location The expression is: In the formula, The quantitative results represent the local differences in the soil settlement trend at the current location; This represents the average settlement difference between the current location and all its nearest neighbor locations at the i-th soil layer. This represents the time dispersion of the i-th soil layer at the current location; This is a preset constant greater than 0; This represents the normalization function.

[0025] Preferably, the prediction of settlement of multi-layered soft soil foundations includes:

[0026] The settlement of any soil layer at the current location within a preset time period prior to the current time is used as the input to the smoothing algorithm. The comprehensive evaluation value of the settlement state of any soil layer at the current location is used as the smoothing coefficient in the smoothing algorithm. The predicted settlement value of any soil layer at the current location is output. The predicted settlement value of each soil layer at the current location is obtained by traversing all soil layers at all locations in the multi-layer soft soil foundation area.

[0027] This application has at least the following beneficial effects:

[0028] This application analyzes the real-time fluctuations in soil settlement to identify settlement jump moments that reflect geological anomalies. Furthermore, it constructs time dispersion based on the information entropy of the time interval between adjacent jump moments, thereby effectively quantifying the distribution pattern and disorder of settlement jumps in the time dimension. This method can deeply explore the dynamic characteristics of soil under the current loading state, distinguish between random impacts and regular loads, and provide a key data foundation and feature support for subsequent accurate assessment of geological conditions and realization of adaptive settlement prediction.

[0029] Furthermore, this application constructs a settlement difference degree by comparing the settlement difference between the current location and its neighboring locations at the settlement jump moment, thereby effectively quantifying the spatial synchronicity and uniformity of geological settlement in a local area. This method can accurately identify the inconsistency of soil settlement behavior and accurately distinguish between geological anomaly centers and stable areas, providing key spatial dimension features to support subsequent comprehensive assessment of geological status.

[0030] Furthermore, this application constructs settlement difference sequences at different depths and analyzes their local differences. Combining single-layer settlement difference degree and time dispersion, it generates a comprehensive assessment value of settlement state, thereby comprehensively quantifying the abnormal state and load center characteristics of the soil layer in the spatiotemporal dimension. This method can accurately capture the essential difference between the core area and the edge area of ​​geological anomalies, providing an accurate and reliable comprehensive evaluation basis for subsequent adaptive adjustment prediction strategies.

[0031] Finally, this application uses the aforementioned comprehensive settlement state assessment value as a smoothing coefficient to adaptively adjust the exponential moving average algorithm, thereby achieving dynamic control of data response sensitivity under different geological conditions. That is, it can quickly respond to drastic settlement changes in the abnormal core area, effectively filter measurement noise in the stable area, and ultimately accurately predict the settlement of each soil layer, thus improving the accuracy of foundation settlement prediction. Attached Figure Description

[0032] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1 A flowchart illustrating the steps of a method for predicting settlement of multi-layer soft soil foundations, provided in one embodiment of this application;

[0034] Figure 2 This is a flowchart illustrating the process of screening settlement jump moments according to one embodiment of this application. Detailed Implementation

[0035] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a multi-layer soft soil foundation settlement prediction method proposed in this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0036] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0037] The following description, in conjunction with the accompanying drawings, details the specific scheme of the multi-layer soft soil foundation settlement prediction method provided in this application.

[0038] This application provides an embodiment of a method for predicting settlement of multi-layer soft soil foundations. Specifically, the method is described below. Please refer to [link to relevant documentation]. Figure 1 The method includes the following steps:

[0039] Step S1: Set up monitoring points in the multi-layered soft soil foundation area and use layered settlement gauges to obtain the settlement of the current point at different soil depths over time.

[0040] In the process of predicting settlement of multi-layer soft soil foundations, firstly, layered settlement gauges are deployed in the target soft soil foundation area to monitor the settlement at each point. The interval between the points is set to 1 meter. At each point, a layered settlement gauge is installed to measure the settlement of soil layers at different depths. Since the consolidation settlement of soft soil is a relatively slow process, this embodiment sets the settlement sampling frequency to 0.5 hours / time. In actual application, the implementer can also set the point layout interval and sampling frequency according to the specific situation. This embodiment does not impose any special restrictions.

[0041] Furthermore, the acquired data is filled with missing values. In this embodiment, linear interpolation is used to fill in the missing data. In practical applications, as other implementation methods, implementers may also use other filling methods such as median filling in combination with specific circumstances. This embodiment does not impose any special restrictions.

[0042] The process of filling in missing data using linear interpolation is a well-known technique and will not be described in detail here.

[0043] Step S2: Before the current moment, analyze the real-time fluctuation of settlement for any soil layer at the current point to identify the settlement jump moment; based on the degree of disorder in the distribution of time intervals between all adjacent settlement jump moments, construct a time dispersion to characterize the settlement jump time distribution law of the soil layer itself.

[0044] In multi-layered soft soil foundation applications, such as airport runways, bridge approach sections, building foundations, and municipal roads, external loads often change dynamically over time. For example, prolonged heavy vehicle traffic, construction work, high-fill embankments, and airport runway embankment construction can all lead to a continuous increase in load and trigger geological anomalies. Under the combined influence of increased load and geological anomalies, the soil settlement rate often exhibits a surge, causing jumps in the monitored settlement data, and the rate of change fluctuates unstablely with changes in load stress.

[0045] Therefore, based on the above analysis, this embodiment analyzes the real-time fluctuation of settlement for any soil layer at the current location to identify settlement jump moments; based on the degree of disorder in the distribution of time intervals between all adjacent settlement jump moments, a time dispersion measure is constructed to characterize the settlement jump time distribution law of the soil layer itself. The specific process is as follows:

[0046] First, prior to the current moment, analyze the real-time fluctuations in settlement for any soil layer at the current location to identify the moment of settlement jump. Specifically:

[0047] Before the current moment, for any soil layer at the current point, calculate the backward difference value of settlement at all sampling moments within the preset time period, and use it as the input of the threshold segmentation algorithm to output the segmentation threshold. The sampling moment corresponding to the backward difference value of settlement that is greater than the segmentation threshold is taken as the settlement jump moment.

[0048] It should be noted that the preset time period length is set manually. In this embodiment, the preset time period length is 24 hours. The purpose is to cover a complete daily cycle of load change. Considering that the settlement response of multi-layer soft soil foundations has lag, external disturbances such as morning and evening rush traffic, daytime construction and nighttime rest usually show a periodic pattern in 24-hour units. Selecting this time period can ensure that the settlement data includes both the settlement surge information during the high load active period and the baseline information during the low load stable period, thereby comprehensively reflecting the dynamic fluctuation characteristics of the soil layer under the current loading state. This avoids losing key fluctuation trends due to a time window that is too short, or introducing redundant data interference due to a time window that is too long. This ensures the accuracy and timeliness of subsequent analysis of settlement jump moments and time intervals. In practical applications, as other implementation methods, implementers can also set their own settings according to specific circumstances. This embodiment does not impose any special restrictions.

[0049] It should be noted that there are many commonly used threshold segmentation algorithms. This embodiment uses the Otsu threshold segmentation algorithm to divide the backward difference value of settlement. In practical applications, as other implementation methods, implementers may also choose other threshold segmentation algorithms according to specific circumstances. This embodiment does not impose any special restrictions on the selection of threshold segmentation algorithms.

[0050] Preferably, the flowchart of the settlement jump time screening process provided in this embodiment is as follows: Figure 2 As shown.

[0051] The calculation process of the backward difference value and the process of partitioning using the Otsu threshold segmentation algorithm are well-known techniques and will not be elaborated further.

[0052] Furthermore, this embodiment constructs a time discreteness to characterize the time distribution law of soil layer settlement jumps based on the degree of disorder in the distribution of time intervals between all adjacent settlement jump moments. Specifically:

[0053] In this embodiment, the total time intervals between all adjacent settlement jump moments of any soil layer at the current location within a preset time period before the current moment are divided into K equal-width intervals. The frequency of time intervals falling into each interval is counted, and the ratio of the frequency of each interval to the total number of all time intervals is used as the probability value of each interval. The probability values ​​of all intervals constitute a probability distribution sequence, and the information entropy of the probability distribution sequence is calculated as the time dispersion of any soil layer at the current location at the current moment, which is used to characterize the time distribution law of the soil layer's own settlement jump.

[0054] It should be noted that in this embodiment, the value of K is set to 10. In actual applications, as other implementation methods, implementers may also set it according to the square root rule based on specific circumstances. This embodiment does not impose any special restrictions.

[0055] It should be noted that there are many commonly used methods for calculating information entropy. In this embodiment, the Shannon entropy of the time interval between all adjacent settlement jump times of any soil layer under the current point within a preset time period before the current time is used as the information entropy of the time interval between all adjacent settlement jump times of any soil layer under the current point within a preset time period before the current time. In actual applications, as other implementation methods, implementers may also use other methods such as permutation entropy according to specific circumstances. This embodiment does not impose any special restrictions.

[0056] Based on the concept of time dispersion, it can be understood that time dispersion is used to characterize the regularity and disorder of the distribution of soil settlement jumps over time. Its calculation logic is based on the time intervals between all adjacent settlement jump moments within a preset period, calculating the information entropy of these time intervals. Time dispersion is affected by the uniformity of the numerical distribution of settlement jump time intervals: when the occurrence time of settlement jumps is random and irregular, the time intervals vary greatly and are disordered, resulting in a larger calculated information entropy. In this case, the time dispersion is high, reflecting that the soil layer is subjected to unpredictable and highly random load impacts, the geological state is extremely unstable, and the settlement fluctuations are disordered. Conversely, when the occurrence time of settlement jumps shows a relatively uniform periodic or clustered pattern, the time intervals are relatively consistent, resulting in a smaller calculated information entropy. In this case, the time dispersion is low, reflecting that the soil layer settlement is affected by a certain continuous or periodic stable load, and although geological anomalies exist, they have certain predictable patterns.

[0057] Thus, this embodiment identifies settlement jump moments reflecting geological anomalies by analyzing real-time fluctuations in soil settlement. Furthermore, it constructs time dispersion based on the information entropy of time intervals between adjacent jump moments, thereby effectively quantifying the distribution pattern and disorder of settlement jumps in the time dimension. This method can deeply explore the dynamic characteristics of soil layers under the current loading state, distinguish between random impacts and regular loads, and provide a key data foundation and feature support for subsequent accurate assessment of geological conditions and the realization of adaptive settlement prediction.

[0058] Step S3: At the moment of settlement jump, compare the difference in settlement between the current point and its neighboring points at any soil layer, and construct the settlement difference degree between the current point and its neighboring points at any soil layer to reflect whether the current point and its neighboring points have undergone a sudden change at the same time.

[0059] Under stable conditions without external abnormal loads, soil settlement in adjacent areas of multi-layered soft soil foundations typically exhibits significant spatial synchronicity, meaning that the settlement of soil layers at similar locations under the same geological conditions shows a highly consistent trend over time. However, when subjected to continuous increases in load, such as construction or heavy vehicle traffic, leading to geological anomalies, the load stress diffuses within the soil. Due to the non-homogeneity of the physical properties of the foundation soil (such as soil stiffness and permeability), the effective load and settlement response at different locations will show significant differences. Specifically, areas with higher stiffness exhibit stronger resistance to deformation, resulting in smaller and more gradual settlement; while weaker areas with lower stiffness are prone to compressive settlement, with drastic changes in settlement. Furthermore, this asynchrony in settlement characteristics has a spatial distance attenuation effect. That is, the farther away a point is from the center of load disturbance or abnormal area, the weaker the stress diffusion effect, and the greater the difference between its settlement pattern and that of the severely affected points. The correlation of characteristics decreases rapidly with increasing distance, resulting in complex local differences in the spatiotemporal distribution of settlement data among various points.

[0060] Therefore, based on the above analysis, this embodiment constructs a settlement difference degree between the current point and its neighboring points at any soil layer by comparing the differences in settlement at any soil layer. This degree is used to reflect whether a sudden change has occurred simultaneously at the current point and its neighboring points. Specifically:

[0061] In this embodiment, the first preset number of points in the ascending order of the distances from each point to all other points within the multi-layer soft soil foundation area are taken as the nearest neighbor points of each point.

[0062] It should be noted that in this embodiment, the preset number is 8. Selecting 8 nearest neighbor points can construct a relatively complete local spatial topology centered on the target point. This order of magnitude is sufficient to cover geological information in different directions and at different distances around the target point, effectively avoiding the problem of one-sided spatial feature sampling or excessive influence from individual noise points due to too few neighbors. At the same time, this number is not set too large to prevent the introduction of data from points that are too far away and have weak geological correlation with the target point. Thus, while ensuring the accuracy of local geological anomaly feature analysis, the computational complexity of the algorithm is effectively controlled, ensuring the real-time performance and effectiveness of settlement difference calculation. In practical applications, as other implementation methods, implementers can also set their own settings according to specific circumstances. This embodiment does not impose any special restrictions.

[0063] Furthermore, in this embodiment, the settlement amounts at all settlement jump moments of each point under any soil layer are formed into a settlement amount sequence; the difference between the settlement amount sequences of the current point and its nearest neighboring points under any soil layer is taken as the settlement difference degree between the current point and its nearest neighboring points under any soil layer.

[0064] It should be noted that there are many methods to measure the differences between sequences. In this embodiment, the Euclidean distance between the current point and its nearest neighbor points at any soil layer is used as the degree of settlement difference between the current point and its nearest neighbor points at any soil layer. In actual application, as another implementation method, the implementer can adopt other methods to measure the differences between sequences according to the specific situation. This embodiment does not impose any special restrictions.

[0065] The calculation process for Euclidean distance is a well-known technique and will not be elaborated further.

[0066] The settlement difference degree can be understood as a measure of the synchronicity between the current location and its neighboring locations in the same soil layer settlement trend. It reflects the uniformity of geological settlement in a local area. Its calculation logic is to measure the difference between the settlement sequence of the current location at the settlement jump moment and the settlement sequence of each neighboring location at the same moment. The settlement difference degree is affected by the difference in the settlement waveform morphology and phase shift between the two locations: when the settlement sequence morphology of the two locations is large and the settlement value difference at the corresponding moment is large, it indicates that the settlement difference degree is higher, reflecting that the settlement behavior of the current location and the neighboring locations is seriously out of sync, and it is highly likely to be located in the central area of ​​geological anomaly or non-uniform settlement. Conversely, when the settlement sequence morphology of the two locations is similar and the value changes synchronously, the smaller the calculated Euclidean distance value, the lower the settlement difference degree, reflecting that the settlement behavior of the current location and the neighboring locations is highly consistent, the geological load in the area is uniform, and no obvious local anomalies have appeared.

[0067] Thus, this embodiment constructs a settlement difference degree by comparing the settlement difference between the current point and its neighboring points at the settlement jump moment, thereby effectively quantifying the spatial synchronicity and uniformity of geological settlement in a local area. This method can accurately identify the inconsistency of soil settlement behavior and accurately distinguish between geological anomaly centers and stable areas, providing key spatial dimension features to support subsequent comprehensive assessment of geological conditions.

[0068] Step S4: Combine the settlement difference between the current point and any of its nearest neighbor points at all soil depths to form a settlement difference sequence. By analyzing the differences between the settlement difference sequences corresponding to different nearest neighbor points, the local differences in the settlement trend of the soil layer at the current point are quantified. Combine the settlement difference and the time dispersion to generate a comprehensive evaluation value of the settlement state of any soil layer at the current point.

[0069] When a surge in the genetic load of multi-layered soft soil induces abnormal settlement, the central area of ​​the load often experiences the fastest settlement rate and the largest settlement, which then diffuses outwards, creating tiny "settlement funnels" or concave surfaces on the surface. Under this non-uniform deformation mode, although the continuity of the soil and rock mass is disrupted, specific distribution characteristics still exist in the local space: on the one hand, for monitoring points close to the load center, the difference between their own settlement and that of the surrounding undisturbed areas (distant neighbors) is significant, indicating that this point is in a high-risk area of ​​geological anomaly; on the other hand, the mechanical transmission characteristics between the point at the anomaly center and its surrounding adjacent points exhibit a specific "central symmetry." This means that although the differences between the center point and each of its neighbors are large, the degree of difference between these neighbors relative to the center point—that is, the direction and magnitude of the settlement gradient—is geometrically relatively uniform and similar.

[0070] Therefore, based on the above analysis, this embodiment constructs a settlement difference sequence by combining the settlement difference between the current location and any of its nearest neighbor locations at all soil depths. By analyzing the differences between the settlement difference sequences corresponding to different nearest neighbor locations, the local variability of the soil settlement trend at the current location is quantified. Combined with the settlement difference and the time dispersion, a comprehensive evaluation value of the settlement state of any soil layer at the current location is generated. The specific process is as follows:

[0071] In this embodiment, firstly, the settlement difference between the current point and any of its nearest neighbor points at all soil depths is arranged in order of soil layers from top to bottom, forming a settlement difference sequence between the current point and any of its nearest neighbor points.

[0072] Furthermore, this embodiment quantifies the local variability of soil settlement trends at the current location by analyzing the differences between settlement difference sequences corresponding to different neighboring points. Specifically:

[0073] In this embodiment, for the current point, the mean of the differences between the settlement difference sequences corresponding to all its nearest neighbor points is used as the quantification result of the local difference in the settlement trend of the soil layer under the current point.

[0074] It should be noted that there are many methods to measure the differences between sequences. In this embodiment, the DTW distance between the settlement difference sequences corresponding to all its nearest neighbor points is taken as the difference between the settlement difference sequences corresponding to all its nearest neighbor points. In practical applications, as other implementation methods, implementers may also use other methods such as Euclidean distance to measure the differences between sequences in combination with specific circumstances. This embodiment does not impose any special restrictions on the selection of methods for measuring the differences between sequences.

[0075] Furthermore, based on the quantification results of the local differences in the settlement trend of the soil layer at the current location, the settlement difference degree, and the time dispersion, this embodiment generates a comprehensive evaluation value of the settlement state of any soil layer at the current location. Specifically:

[0076] In this embodiment, the comprehensive evaluation value of the settlement state of any soil layer at the current point is positively correlated with the quantitative result of the local difference in the settlement trend of the soil layer at the current point and the degree of settlement difference, and negatively correlated with the time dispersion.

[0077] It should be understood that a positive correlation means that the dependent variable increases as the independent variable increases, and the dependent variable decreases as the independent variable decreases. The specific relationship can be additive or multiplicative, etc., and is determined by the actual application. This application does not impose any special restrictions. A negative correlation means that the dependent variable decreases as the independent variable increases, and the dependent variable increases as the independent variable decreases. The relationship can be subtractive or divisive, etc., and is determined by the actual application.

[0078] Preferably, as one implementation method, in this embodiment, the comprehensive evaluation value of the settlement state of the i-th soil layer at the current location is... The expression is: In the formula, The quantitative results represent the local differences in the soil settlement trend at the current location; This represents the average settlement difference between the current location and all its nearest neighbor locations at the i-th soil layer. This represents the time dispersion of the i-th soil layer at the current location; A preset constant greater than 0 is used to prevent the denominator from being 0. In this embodiment, The value is 0.01. In practical applications, as other implementation methods, implementers can also set it according to specific circumstances. This embodiment does not impose any special restrictions. This represents the normalization function.

[0079] It should be noted that the normalization function used in this embodiment is the maximum-minimum value normalization function. Normalization is performed. In practical applications, as other implementation methods, implementers may also adopt other normalization methods according to specific circumstances. This embodiment does not impose any special restrictions.

[0080] Based on the comprehensive settlement status assessment value, it can be understood that the comprehensive settlement status assessment value is used to comprehensively quantify the risk level of abnormal settlement at the current location in a specific soil layer and the possibility of it acting as a load center. It reflects the comprehensive abnormal state of the soil layer in the spatiotemporal dimensions. Its calculation logic is to construct a mathematical model that is positively correlated with the local difference quantification result and the settlement difference degree, and negatively correlated with the time dispersion. This calculation is jointly affected by three core factors: the local difference quantification result, the settlement difference degree, and the time dispersion degree. When the local difference quantification result and the settlement difference degree are larger, and the time dispersion degree is smaller, the calculated comprehensive settlement status assessment value is higher. The larger the value, the greater the difference between the current point and its neighboring points, and the consistency of the differences among the neighboring points. At the same time, the jump time is highly regular, which is very consistent with the characteristics of an abnormal load center. This means that the current point is in the core area of ​​the geological anomaly and the prediction smoothing strategy needs to be significantly adjusted to adapt to rapid settlement. Conversely, when the local difference quantification result and the settlement difference are smaller, and the time dispersion is larger, the calculated comprehensive settlement state assessment value is smaller. This means that the current point is highly synchronized with its neighboring points and the jump time is chaotic. This does not conform to the characteristics of an anomaly center and means that the point is in a normal or marginal area. A more gradual conventional prediction strategy can be adopted.

[0081] Thus, this embodiment constructs a settlement difference sequence at different depths and analyzes its local differences. By combining the single-layer settlement difference degree and time dispersion, a comprehensive assessment value of settlement state is generated. This comprehensively quantifies the abnormal state and load center characteristics of the soil layer in the spatiotemporal dimension. This method can accurately capture the essential difference between the core area and the edge area of ​​geological anomalies, providing an accurate and reliable comprehensive evaluation basis for subsequent adaptive adjustment prediction strategies.

[0082] Step S5: Based on the comprehensive settlement status assessment value, a smoothing algorithm is used to predict the settlement of multi-layer soft soil foundations.

[0083] Based on the above steps, a comprehensive assessment value of the soil layer's settlement state was obtained. Furthermore, a smoothing algorithm was employed to predict the settlement of multi-layered soft soil foundations. Specifically:

[0084] In this embodiment, the settlement of any soil layer at the current point during all sampling times within a preset time period before the current time is used as the input of the smoothing algorithm. The comprehensive evaluation value of the settlement state of any soil layer at the current point is used as the smoothing coefficient in the smoothing algorithm. The predicted settlement value of any soil layer at the current point is output. The predicted settlement value of each soil layer at the current point is obtained by traversing all soil layers at all points in the multi-layer soft soil foundation area. The smoothing algorithm in this embodiment adopts the exponential moving average (EMA) algorithm.

[0085] Furthermore, the prediction results are input into 3D visualization software (ParaView) to generate visualization results for display and feedback.

[0086] To clarify, using the comprehensive settlement assessment value as a smoothing coefficient is to adaptively adjust the sensitivity of the Exponential Moving Average (EMA) algorithm to the latest data based on the severity of the geological anomaly. Specifically, a larger comprehensive settlement assessment value indicates that the current location is in the core area of ​​a geological anomaly, with settlement exhibiting a rapid and regular jumping trend. In this case, to capture this rapid dynamic change in real time and prevent prediction lag, the latest settlement observation data needs to be given higher weight, i.e., the smoothing coefficient needs to be increased to reduce the interference of historical data, allowing the prediction curve to quickly match the actual drastic settlement changes. Conversely, a smaller comprehensive settlement assessment value indicates that the current location is in a geologically stable state, with gentle and synchronous settlement changes. In this case, to filter out accidental measurement noise and maintain the smoothness and stability of the prediction curve, the smoothing coefficient needs to be reduced to enhance the smoothing effect of historical data, thereby avoiding oscillations in the prediction value caused by individual minor fluctuations. Ultimately, this achieves a balance between the adaptability and accuracy of the prediction algorithm under different geological conditions.

[0087] Thus, this embodiment utilizes the aforementioned comprehensive settlement state evaluation value as a smoothing coefficient to adaptively adjust the exponential moving average algorithm, thereby achieving dynamic control of data response sensitivity under different geological conditions. Specifically, it rapidly responds to drastic settlement changes in the abnormal core area, effectively filters measurement noise in the stable area, and ultimately accurately predicts the settlement of each soil layer. The prediction results are then visually displayed through three-dimensional visualization technology, providing decision support for settlement monitoring of multi-layer soft soil foundations that combines real-time performance and stability.

[0088] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, specific embodiments of this specification have been described above. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0089] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0090] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them; modifications to the technical solutions described in the foregoing embodiments, or equivalent substitutions of some of the technical features, do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for predicting settlement of multi-layer soft soil foundations, characterized in that, The method includes the following steps: Monitoring points were set up in the multi-layered soft soil foundation area, and the settlement of the current point at different soil depths over time was obtained using a layered settlement gauge. Before the current moment, for any soil layer at the current location, analyze the real-time fluctuation of settlement to identify the settlement jump moment; based on the degree of disorder in the distribution of time intervals between all adjacent settlement jump moments, construct a time dispersion to characterize the settlement jump time distribution law of the soil layer itself. At the moment of settlement jump, the difference in settlement between the current point and its neighboring points at any soil layer is compared to construct the settlement difference degree between the current point and its neighboring points at any soil layer, which is used to reflect whether the current point and its neighboring points have undergone a sudden change at the same time. The settlement difference between the current point and any of its nearest neighbor points at all soil depths is used to form a settlement difference sequence. By analyzing the differences between the settlement difference sequences corresponding to different nearest neighbor points, the local differences in the settlement trend of the soil layer at the current point are quantified. Combined with the settlement difference and the time dispersion, a comprehensive evaluation value of the settlement state of any soil layer at the current point is generated. Based on the comprehensive assessment value of the settlement state, a smoothing algorithm is used to predict the settlement of multi-layer soft soil foundations. The comprehensive evaluation value of the settlement state of any soil layer at the current location further includes: Comprehensive assessment value of settlement state of the i-th soil layer at the current location The expression is: In the formula, The quantitative results represent the local differences in the soil settlement trend at the current location; This represents the average settlement difference between the current location and all its nearest neighbor locations at the i-th soil layer. This represents the time dispersion of the i-th soil layer at the current location; This is a preset constant greater than 0; Represents the normalization function; The prediction of settlement of multi-layered soft soil foundations includes: The settlement of any soil layer at the current location within a preset time period prior to the current time is used as the input to the smoothing algorithm. The comprehensive evaluation value of the settlement state of any soil layer at the current location is used as the smoothing coefficient in the smoothing algorithm. The predicted settlement value of any soil layer at the current location is output. The predicted settlement value of each soil layer at the current location is obtained by traversing all soil layers at all locations in the multi-layer soft soil foundation area.

2. The method for predicting settlement of multi-layer soft soil foundations as described in claim 1, characterized in that, The method for identifying the moment of settlement jump includes: Before the current moment, for any soil layer at the current point, calculate the backward difference value of settlement at all sampling moments within the preset time period, and use it as the input of the threshold segmentation algorithm to output the segmentation threshold. The sampling moment corresponding to the backward difference value of settlement that is greater than the segmentation threshold is taken as the settlement jump moment.

3. The method for predicting settlement of multi-layer soft soil foundations as described in claim 1, characterized in that, The method for constructing the time discreteness is as follows: Divide all time intervals between all adjacent settlement jump moments of any soil layer at the current location within a preset time period before the current moment into multiple equal-width intervals. Count the frequency of time intervals falling into each interval. Take the ratio of the frequency of each interval to the total number of all time intervals as the probability value of each interval. The probability values ​​of all intervals constitute a probability distribution sequence. Calculate the information entropy of the probability distribution sequence as the time dispersion of any soil layer at the current location at the current moment.

4. The method for predicting settlement of multi-layer soft soil foundations as described in claim 1, characterized in that, The nearest neighbor points of each point are the first preset number of points in the ascending order of the distances from each point to all other points within the multi-layer soft soil foundation area.

5. The method for predicting settlement of multi-layer soft soil foundations as described in claim 1, characterized in that, The process of constructing the settlement difference between the current point and its nearest neighbor points at any soil layer is as follows: The settlement amounts at all settlement jump moments below any soil layer at each point are compiled into a settlement sequence; the difference between the settlement sequence of the current point and its nearest neighbor points below any soil layer is taken as the settlement difference degree between the current point and its nearest neighbor points at any soil layer.

6. The method for predicting settlement of multi-layer soft soil foundations as described in claim 1, characterized in that, The quantification process for the local differences in soil settlement trend at the current location is as follows: For the current location, the mean of the differences between the settlement difference sequences corresponding to all its nearest neighbor locations is used as the quantitative result of the local difference in the settlement trend of the soil layer under the current location.

7. The method for predicting settlement of multi-layer soft soil foundations as described in claim 6, characterized in that, The settlement difference is arranged in a settlement difference sequence according to the soil layers from top to bottom.

8. The method for predicting settlement of multi-layer soft soil foundations as described in claim 1, characterized in that, The comprehensive evaluation value of the settlement state of any soil layer at the current location is positively correlated with the quantitative result of the local difference in the settlement trend of the soil layer at the current location and the degree of settlement difference, and negatively correlated with the time dispersion.

Images