An intelligent automatic monitoring system for deformation of ultra-deep foundation pits

Abstract

This application relates to the field of deep foundation pit deformation detection technology, and proposes an intelligent automatic monitoring system for ultra-deep foundation pit deformation. The system includes: collecting horizontal displacement, groundwater level, and support internal forces at different monitoring points on the same fault in the area to be monitored for ultra-deep foundation pit deformation; marking the target acquisition time and target monitoring points, and determining the data anomaly degree, first consistency, and probability of data anomaly at the acquisition time; determining the fusion weight of the monitoring points at the acquisition time; establishing a fault displacement surface at the same acquisition time based on the fusion weight of each monitoring point; and realizing the monitoring of deep foundation pit deformation based on the difference between the fault displacement surfaces at two adjacent acquisition times. This application improves the stability and accuracy of ultra-deep foundation pit deformation monitoring.

Description

Technical Field

[0001] This application relates to the field of deep foundation pit deformation detection technology, specifically to an intelligent automatic monitoring system for ultra-deep foundation pit deformation. Background Technology

[0002] To ensure the safety of the foundation pit itself and the surrounding environment, improve the quality of engineering construction, avoid safety accidents such as foundation pit instability and collapse, and ensure the safety and stability of the surrounding environment, it is necessary to monitor the deformation of ultra-deep foundation pits. The deformation of ultra-deep foundation pits is generally fault displacement, with concentrated and complex local deformation. Considering the ability of interpolation results to capture local details, the IDW (Inverse Distance Weighting) method is typically used to interpolate and fit the deformation data of the fault displacement monitoring points of ultra-deep foundation pits, generating a deformation displacement surface. The deformation monitoring results of ultra-deep foundation pits are then obtained based on this deformation displacement surface.

[0003] Different fault displacement monitoring points on the same fault are easily affected by factors such as physical field anomalies, sensor heterogeneity, and random construction disturbances during data collection. As a result, the instantaneous deformation caused by different influencing factors is often recorded as the true deformation data. However, abnormal data often affects the interpolation results of the IDW inverse distance weighting method, leading to false alarms in ultra-deep foundation pit deformation monitoring and affecting the accuracy of ultra-deep foundation pit deformation monitoring. Summary of the Invention

[0004] This application provides an intelligent automatic monitoring system for the deformation of ultra-deep foundation pits to solve the problem that abnormal data caused by different influencing factors affects the accuracy of ultra-deep foundation pit deformation monitoring. The specific technical solution adopted is as follows:

[0005] One embodiment of this application provides an intelligent automatic monitoring system for deformation of ultra-deep foundation pits, the system comprising the following modules:

[0006] The data acquisition module is used to collect the horizontal displacement, groundwater level, and support internal force at different monitoring points on the same fault in the area to be monitored for deformation of ultra-deep foundation pits at different acquisition times.

[0007] The data anomaly probability acquisition module is used to record any collection time as the target collection time and any monitoring point as the target monitoring point. Based on the differences between the horizontal displacement, groundwater level and support internal force of the target monitoring point at the collection time adjacent to the target collection time, the first consistency of the target monitoring point at the target collection time is determined. Based on the differences between the first consistency of different monitoring points on the same fault at the same collection time and the distance between different monitoring points on the same fault, the data anomaly degree of each monitoring point at the same collection time is determined. Based on the first consistency of the target monitoring point at the target collection time and the collection time adjacent to the target collection time, the data anomaly degree is weighted and summed to determine the data anomaly probability of the target monitoring point at the target collection time.

[0008] The fusion weight determination module is used to determine the fusion weight of each monitoring point on the same fault at each acquisition time based on the changing trend of data anomaly degree and the difference in the probability of data anomaly of all monitoring points on the same fault at adjacent acquisition times.

[0009] The deep foundation pit deformation monitoring module is used to establish a fault displacement surface at the same acquisition time based on the fusion weight of each monitoring point at the same acquisition time, and to realize the monitoring of deep foundation pit deformation based on the difference between the fault displacement surfaces at two adjacent acquisition times.

[0010] Furthermore, the method for determining the first consistency of the target monitoring points at the target acquisition time is as follows:

[0011] The difference between the horizontal displacement of the target monitoring point at the target acquisition time and the previous adjacent acquisition time is recorded as the adjacent horizontal displacement difference of the target monitoring point at the target acquisition time; the difference between the support internal forces of the target monitoring point at the target acquisition time and the previous adjacent acquisition time is recorded as the adjacent support internal force difference of the target monitoring point at the target acquisition time.

[0012] When the difference between adjacent horizontal displacements and the difference between adjacent internal forces of adjacent supports at the target acquisition time are both positive or both negative, the first correlation coefficient of the target monitoring point at the target acquisition time is assigned a value of 1. When the difference between adjacent horizontal displacements and the difference between adjacent internal forces of adjacent supports at the target acquisition time are one positive and one negative, the first correlation coefficient of the target monitoring point at the target acquisition time is assigned a value of -1.

[0013] Based on the groundwater level at the target monitoring point at the target acquisition time and the acquisition time adjacent to the target acquisition time, determine the characteristic value of the adjacent groundwater level at the target monitoring point at the target acquisition time.

[0014] The product of the first correlation coefficient of the target monitoring point at the target acquisition time and the characteristic value of the adjacent groundwater level is denoted as the first consistency of the target monitoring point at the target acquisition time.

[0015] Furthermore, the specific steps for determining the characteristic value of the adjacent groundwater level of the target monitoring point at the target acquisition time based on the groundwater level at the target acquisition time and the acquisition time adjacent to the target acquisition time include:

[0016] The difference between the groundwater level at the target monitoring point at the target acquisition time and the previous adjacent acquisition time is recorded as the adjacent groundwater level difference at the target monitoring point at the target acquisition time. The negative correlation processing result of the adjacent groundwater level difference at the target monitoring point at the target acquisition time is recorded as the adjacent groundwater level characteristic value at the target monitoring point at the target acquisition time.

[0017] Furthermore, the determination of the data anomaly degree of each monitoring point at the same acquisition time based on the difference in the first consistency between different monitoring points on the same fault at the same acquisition time, and the distance between different monitoring points on the same fault, includes the following specific contents:

[0018] The absolute value of the difference between the first consistency of the target monitoring point and other monitoring points at the target acquisition time is denoted as the first absolute difference between the target monitoring point and other monitoring points at the target acquisition time; the negative correlation processing result of the distance between the target monitoring point and other monitoring points is denoted as the distance feature value between the target monitoring point and other monitoring points; the product of the first absolute difference between the target monitoring point and other monitoring points at the target acquisition time and the distance feature value is denoted as the first product between the target monitoring point and other monitoring points at the target acquisition time.

[0019] The data anomaly degree of the target monitoring point at the target acquisition time is determined by the first product of the target monitoring point and all other monitoring points at the target acquisition time.

[0020] Furthermore, the specific steps for determining the data anomaly degree of the target monitoring point at the target acquisition time based on the first product of the target monitoring point and all other monitoring points at the target acquisition time are as follows:

[0021] The normalized value of the sum of the first products of the target monitoring point and all other monitoring points at the target acquisition time is denoted as the data anomaly degree of the target monitoring point at the target acquisition time.

[0022] Furthermore, the method for determining the probability of data anomalies at the target monitoring point at the target acquisition time is as follows:

[0023] The first consistency normalized value of the target monitoring point at the previous adjacent collection time is used as the weight to perform a weighted summation of the data anomalies. The weighted summation result is recorded as the adjacent data anomaly of the target monitoring point at the target collection time.

[0024] The probability of data anomalies at the target monitoring point at the target acquisition time is determined based on the weights of the data anomalies at the target acquisition time and the adjacent data anomalies at the target acquisition time.

[0025] Furthermore, the specific method for determining the probability of data anomalies at the target monitoring point at the target acquisition time based on the weights of the data anomaly degree of the target monitoring point at the target acquisition time and the anomalies of adjacent data at the target acquisition time includes:

[0026] The preset first weight parameter is used as the weight of the data anomaly degree of the target monitoring point at the target acquisition time. The difference between the number 1 and the first weight parameter is used as the weight of the adjacent data anomaly degree of the target monitoring point at the target acquisition time. The weighted sum is recorded as the data anomaly probability of the target monitoring point at the target acquisition time.

[0027] Furthermore, the method for determining the fusion weight of each monitoring point on the same fault at each acquisition time is as follows:

[0028] Linear fitting is performed on the data anomaly degree of the target monitoring point at the target acquisition time and the adjacent acquisition time before the target acquisition time with the corresponding acquisition time. The slope of the fitted line is obtained. The absolute value of the slope of the fitted line and the sum of the number 1 are recorded as the anomaly tendency of the target monitoring point at the target acquisition time.

[0029] Based on the differences and tendencies of the probability of data anomalies at all monitoring points on the same fault at the target acquisition time, the fusion weight of the target monitoring point at the target acquisition time is determined.

[0030] Furthermore, the specific steps for determining the fusion weight of the target monitoring point at the target acquisition time based on the differences and tendencies of data anomalies among all monitoring points on the same fault at the target acquisition time include:

[0031] The mean of the absolute values ​​of the differences between the data anomaly probability of the target monitoring point and all other monitoring points on the same fault at the target acquisition time is denoted as the anomaly difference of the target monitoring point on the same fault at the target acquisition time; the ratio of the anomaly difference of the target monitoring point on the same fault at the target acquisition time to the anomaly tendency is denoted as the anomaly degree of the target monitoring point on the same fault at the target acquisition time.

[0032] The negative correlation processing result of the same fault anomaly degree of the target monitoring point at the target acquisition time is recorded as the fusion weight of the target monitoring point at the target acquisition time.

[0033] Furthermore, the specific steps for establishing the fault displacement surface at the same acquisition time based on the fusion weights of each monitoring point at the same acquisition time are as follows:

[0034] The fusion weights of each monitoring point at the same acquisition time are used as the weights of the horizontal displacement of each monitoring point. The inverse distance weighting method (IDW) is used for interpolation to obtain the fault displacement surface at the same acquisition time.

[0035] The beneficial effects of this application are:

[0036] This application extracts the correlation between different types of data based on the differences in horizontal displacement, groundwater level, and support internal forces at adjacent acquisition times, obtaining the first consistency of the target monitoring point at the target acquisition time. To highlight the impact of anomalies on the target monitoring point, it assigns a smaller weight to the difference in the first consistency between two closely spaced monitoring points at the target acquisition time based on the differences in the correlation between different types of data extracted from different monitoring points on the same fault, thus obtaining the data anomaly degree. The data anomaly degree is then weighted and summed based on the first consistency of the target monitoring point at the target acquisition time and the adjacent acquisition times before the target acquisition time to determine the probability of data anomalies at the target monitoring point at the target acquisition time. Different monitoring points on a fault are subjected to the same geological structure and load. Therefore, anomalies at different monitoring points on the same fault usually have spatial continuity. Thus, based on the changing trend of data anomaly degree and the difference in the probability of data anomalies at adjacent acquisition times of all monitoring points on the same fault, the fusion weight of each monitoring point on the same fault at each acquisition time is determined. Based on the fusion weight of each monitoring point at the same acquisition time, the fault displacement surface at the same acquisition time is established. Based on the difference of the fault displacement surface at two adjacent acquisition times, the deformation monitoring of deep foundation pits is realized, solving the problem of the impact of abnormal data formed by different influencing factors on the accuracy of ultra-deep foundation pit deformation monitoring, and improving the stability and accuracy of ultra-deep foundation pit deformation monitoring. Attached Figure Description

[0037] To more clearly illustrate the technical solutions 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.

[0038] Figure 1 A flowchart illustrating an intelligent automatic monitoring system for deformation of an ultra-deep foundation pit, provided as an embodiment of this application;

[0039] Figure 2This is a schematic diagram of the structure of an intelligent automatic monitoring system for deformation of an ultra-deep foundation pit, provided as an embodiment of this application. Detailed Implementation

[0040] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0041] Please see Figure 1 It shows a flowchart of an intelligent automatic monitoring system for deformation of ultra-deep foundation pits according to an embodiment of this application. Figure 2 The diagram shows a schematic of an intelligent automatic monitoring system for deformation of an ultra-deep foundation pit according to an embodiment of this application. The system includes: a data acquisition module, a data anomaly probability acquisition module, a fusion weight determination module, and a deep foundation pit deformation monitoring module.

[0042] The data acquisition module collects the horizontal displacement, groundwater level, and support internal force at different monitoring points on the same fault in the area to be monitored for deformation of ultra-deep foundation pits at different acquisition times.

[0043] Using a static level, different monitoring points were set up on the same fault in the area to be monitored for deformation of the ultra-deep foundation pit. At each monitoring point, a fixed inclinometer was used to collect the horizontal displacement, a water level gauge was used to collect the groundwater level, and a steel support axial force gauge was used to collect the support internal force.

[0044] Preferably, in one embodiment of this application, when collecting data on the horizontal displacement, groundwater level, and support internal forces of the monitoring points, the data sampling frequency is once per minute, and the number of monitoring points set up on the same fault is 20. In practical applications, as other implementation methods, implementers can decide the sampling frequency of the horizontal displacement, groundwater level, and support internal forces of the monitoring points and the number of monitoring points set up on the same fault according to the actual situation; this application does not impose any special restrictions.

[0045] Thus, the horizontal displacement, groundwater level, and support internal forces at different monitoring points on the same fault in the area to be monitored for the deformation of the ultra-deep foundation pit were obtained at different collection times.

[0046] The data anomaly probability acquisition module designates any acquisition time as the target acquisition time and any monitoring point as the target monitoring point. Based on the differences in horizontal displacement, groundwater level, and support internal forces of the target monitoring point at acquisition times adjacent to the target acquisition time, it determines the first consistency of the target monitoring point at the target acquisition time. Based on the differences in the first consistency of different monitoring points on the same fault at the same acquisition time, and the distance between different monitoring points on the same fault, it determines the data anomaly degree of each monitoring point at the same acquisition time. Based on the first consistency of the target monitoring point at the target acquisition time and the acquisition times adjacent to the target acquisition time, it performs a weighted summation of the data anomaly degree to determine the data anomaly probability of the target monitoring point at the target acquisition time.

[0047] Factors such as physical field anomalies, sensor heterogeneity, and random construction disturbances can affect the horizontal displacement, groundwater level, and support internal forces of monitoring points. This often results in data silos, necessitating analysis of different types of data to determine their correlations. For example, a correlation might be observed where increased deformation at a monitoring point leads to increased horizontal displacement, often resulting in stress release in the support. Therefore, the horizontal displacement and support internal forces at the same monitoring point at the same acquisition time exhibit a negative correlation.

[0048] Any acquisition time is recorded as the target acquisition time. Based on the differences in horizontal displacement, groundwater level, and support internal force of the same monitoring point at acquisition times adjacent to the target acquisition time, the first consistency of the same monitoring point at the target acquisition time is determined.

[0049] For the same monitoring point, the difference in horizontal displacement between the same monitoring point at the target acquisition time and the previous adjacent acquisition time is recorded as the adjacent horizontal displacement difference of the same monitoring point at the target acquisition time; the difference in support internal force between the same monitoring point at the target acquisition time and the previous adjacent acquisition time is recorded as the adjacent support internal force difference of the same monitoring point at the target acquisition time; when both the adjacent horizontal displacement difference and the adjacent support internal force difference of the same monitoring point at the target acquisition time are positive or both are negative, the first correlation coefficient of the same monitoring point at the target acquisition time is assigned a value of 1. When the difference between the internal forces of the monitoring point and the adjacent support is a positive number and a negative number, the first correlation coefficient of the same monitoring point at the target acquisition time is assigned a value of -1; the difference between the groundwater level of the same monitoring point at the target acquisition time and the previous adjacent acquisition time is recorded as the adjacent groundwater level difference of the same monitoring point at the target acquisition time; the negative correlation processing result of the adjacent groundwater level difference of the same monitoring point at the target acquisition time is recorded as the adjacent groundwater level characteristic value of the same monitoring point at the target acquisition time; the product of the first correlation coefficient of the same monitoring point at the target acquisition time and the adjacent groundwater level characteristic value is recorded as the first consistency of the same monitoring point at the target acquisition time.

[0050] It is understood that negative correlation processing is applied to the differences in adjacent groundwater levels, that is, to ensure that the differences in adjacent groundwater levels are negatively correlated with the characteristic values ​​of adjacent groundwater levels. It is understood that the negative correlation in this application refers to the relationship between the independent variable and the dependent variable, where the independent variable is the difference in adjacent groundwater levels and the dependent variable is the characteristic value of adjacent groundwater levels. The negative correlation means that the dependent variable decreases (increases) as the independent variable increases (decreases), and can be an inverse relationship, a subtraction relationship, etc.

[0051] Preferably, as an embodiment of this application, the negative of the difference between adjacent groundwater levels at the target acquisition time is used as the exponent of an exponential function with the natural constant as the base, and the calculation result of the exponential function is recorded as the characteristic value of the adjacent groundwater level at the target acquisition time.

[0052] The first correlation coefficient at the target acquisition time represents the correlation between the horizontal displacement and the internal force of the support at the target acquisition time. The first correlation coefficient for a positive correlation is 1, and the first correlation coefficient for a negative correlation is -1. The first consistency at the target acquisition time is used to detect abnormalities in the relationship between the horizontal displacement and the internal force of the support at the target acquisition time. When the horizontal displacement, groundwater level and internal force of the support at the monitoring point are normal at the target acquisition time, the first correlation coefficient at the target acquisition time is 1. If the characteristic value of the adjacent groundwater level at the target acquisition time is positive and large, then the first consistency of the monitoring point at the target acquisition time is large.

[0053] The same method can be used to obtain the first consistency of the monitoring point at any collection time.

[0054] Furthermore, based on the differences in the first consistency between different monitoring points on the same fault at the same acquisition time, and the distance between different monitoring points on the same fault, the data anomaly degree of each monitoring point at the same acquisition time is determined respectively.

[0055] Let any one monitoring point be designated as the target monitoring point. Let the absolute value of the difference between the target monitoring point and other monitoring points at the target acquisition time be designated as the first absolute difference between the target monitoring point and other monitoring points at the target acquisition time. Let the negative correlation processing result of the distance between the target monitoring point and other monitoring points be designated as the distance feature value between the target monitoring point and other monitoring points. Let the product of the first absolute difference between the target monitoring point and other monitoring points at the target acquisition time and the distance feature value be designated as the first product between the target monitoring point and other monitoring points at the target acquisition time. Let the normalized value of the sum of the first products between the target monitoring point and all other monitoring points at the target acquisition time be designated as the data anomaly degree of the target monitoring point at the target acquisition time.

[0056] It should be noted that this embodiment uses the Z-Score standard normalization method to calculate the normalized value. In practical applications, implementers may use other methods of existing technology, such as the maximum-minimum normalization method or the sigmoid function, to calculate the normalized value, and no limitation is made here.

[0057] Preferably, as an embodiment of this application, the negative of the distance between the target monitoring point and other monitoring points is used as the exponent of an exponential function with the natural constant as the base, and the calculation result of the exponential function is recorded as the distance characteristic value between the target monitoring point and other monitoring points.

[0058] The distance between different monitoring points on the same fault can adjust the weight of the difference in the first consistency between other monitoring points and the target monitoring point at the target acquisition time. This assigns a smaller weight to the difference in the first consistency between two closely spaced monitoring points at the target acquisition time, highlighting the impact of anomalies on the target monitoring point. The greater the anomaly of the data at the target acquisition time, the greater the likelihood that the horizontal displacement, groundwater level, and support internal forces collected at the target acquisition time are affected.

[0059] The same method can be used to obtain the data anomaly degree of any monitoring point at any collection time.

[0060] The probability of data anomalies at the target acquisition time is determined by weighted summation based on the first consistency of the target monitoring point at the target acquisition time and the adjacent acquisition times before the target acquisition time.

[0061] Preferably, as an embodiment of this application, the target monitoring points adjacent to the target acquisition time are... Each acquisition time is recorded as the previous adjacent acquisition time of the target monitoring point at the target acquisition time. The sum of the first consistency of all previous adjacent acquisition times of the target monitoring point at the target acquisition time is recorded as the first consistency sum of the target monitoring point at the target acquisition time. The ratio of the first consistency of the target monitoring point at the previous adjacent acquisition times to the first consistency sum is recorded as the first consistency normalized value of the target monitoring point at the previous adjacent acquisition times.

[0062] The first consistent normalized value of the target monitoring point at the target acquisition time, taken as the weight, is used to weight the data anomaly scores. The weighted sum is recorded as the adjacent data anomaly score of the target monitoring point at the target acquisition time. A preset first weight parameter is used as the weight of the data anomaly score of the target monitoring point at the target acquisition time. The difference between the number 1 and the first weight parameter is used as the weight of the adjacent data anomaly score of the target monitoring point at the target acquisition time. These are then weighted and summed, and the weighted sum is recorded as the data anomaly probability of the target monitoring point at the target acquisition time.

[0063] in, The first preset quantity is 30 in this embodiment; the first weight parameter is a preset parameter, and in this embodiment, the first weight parameter is 0.6. The function of the preset first weight parameter is to adjust the influence weight of the data anomaly degree of the previous adjacent acquisition time and the target acquisition time on the probability of data anomaly at the target acquisition time. The larger the first weight parameter, the greater the influence of the data anomaly degree of the monitoring point on the probability of data anomaly at the target acquisition time.

[0064] The same method can be used to obtain the probability of data anomalies at any monitoring point at any collection time.

[0065] This allows us to determine the probability of data anomalies at all monitoring points at all collection times.

[0066] The fusion weight determination module determines the fusion weight of each monitoring point on the same fault at each acquisition time based on the changing trend of data anomaly degree and the difference in the probability of data anomaly at adjacent acquisition times of all monitoring points on the same fault.

[0067] Different monitoring points on the same fault are subjected to the same geological structure and loads, and anomalies at different monitoring points on the same fault usually exhibit spatial continuity. When the deviation in anomaly degree between a certain monitoring point on the same fault and other different monitoring points is too large, the data collected at that monitoring point is more likely to be abnormal data caused by random interference. This can be used as a basis to distinguish the data collected at the corresponding acquisition time for deep foundation pit deformation and random interference.

[0068] Based on the changing trend of the data anomaly degree of the target monitoring point at the target acquisition time and the adjacent acquisition time before the target acquisition time, the anomaly tendency of the target monitoring point at the target acquisition time is determined.

[0069] Using the target acquisition time and the adjacent acquisition time before the target acquisition time as independent variables, and the data anomaly degree of the target monitoring point at the target acquisition time and the adjacent acquisition time before the target acquisition time as dependent variables, a linear fit is performed to obtain the slope of the fitted line. The absolute value of the slope of the fitted line and the sum of the number 1 are recorded as the anomaly tendency of the target monitoring point at the target acquisition time.

[0070] This embodiment uses the least squares method for linear fitting, which is a well-known technique and will not be elaborated further. In practical applications, as other implementation methods, while achieving the goal of linear fitting, implementers may use other existing methods such as polynomial fitting to fit the straight line; this application does not impose any special limitations.

[0071] The greater the anomaly tendency at the target acquisition time, the more significant the anomaly of the data acquired at the target acquisition time, and the greater the possibility that the anomaly is caused by anomalies such as local soil instability at the monitoring point location.

[0072] Based on the differences and tendencies of the probability of data anomalies at all monitoring points on the same fault at the target acquisition time, the fusion weight of the target monitoring point at the target acquisition time is determined.

[0073] The mean of the absolute values ​​of the differences between the data anomaly probability of the target monitoring point and all other monitoring points on the same fault at the target acquisition time is denoted as the intra-fault anomaly difference of the target monitoring point at the target acquisition time. The ratio of the intra-fault anomaly difference of the target monitoring point at the target acquisition time to the anomaly tendency is denoted as the intra-fault anomaly degree of the target monitoring point at the target acquisition time. The negative correlation processing result of the intra-fault anomaly degree of the target monitoring point at the target acquisition time is denoted as the fusion weight of the target monitoring point at the target acquisition time.

[0074] Preferably, as an embodiment of this application, the negative number of the same fault anomaly degree of the target monitoring point at the target acquisition time is used as the exponent of an exponential function with the natural constant as the base, and the calculation result of the exponential function is recorded as the fusion weight of the target monitoring point at the target acquisition time.

[0075] The greater the difference in anomalies between the target monitoring points and the same fault at the target acquisition time, the greater the possibility that the data collected at the monitoring point location is abnormal data caused by random interference. At the same time, the smaller the anomaly tendency, the smaller the fusion weight of the target monitoring point at the target acquisition time. When using the data collected by the target monitoring point at the target acquisition time to monitor the deformation of deep foundation pits, the weight assigned to the data collected by the target monitoring point at the target acquisition time should be greater.

[0076] The same method can be used to obtain the fusion weight of any monitoring point at any collection time.

[0077] At this point, the fusion weights of all monitoring points at all data collection times are obtained.

[0078] The deep foundation pit deformation monitoring module establishes a fault displacement surface at the same acquisition time based on the fusion weight of each monitoring point at the same acquisition time, and realizes the monitoring of deep foundation pit deformation based on the difference between the fault displacement surfaces at two adjacent acquisition times.

[0079] The fusion weights of each monitoring point at the same acquisition time are used as the weights of the horizontal displacement of each monitoring point. The inverse distance weighting method (IDW) is used for interpolation to obtain the fault displacement surface at the same acquisition time.

[0080] The method of using IDW (Inverse Distance Weighting) to obtain the fault displacement surface based on horizontal displacement is a well-known technique and will not be elaborated further.

[0081] Based on the difference in fault displacement surfaces at two adjacent acquisition times, fault displacement surface difference analysis is performed to achieve effective monitoring of deep foundation pit deformation.

[0082] Among them, the effective monitoring of deep foundation pit deformation based on the analysis of fault displacement surface differences is a well-known technology and will not be elaborated further. Specifically, as an embodiment of this application, the differences in fault displacement surfaces at two adjacent acquisition times are compared. The area consisting of points with a distance of less than or equal to 2 mm between corresponding points on the two fault displacement surfaces is designated as the safe zone; the area consisting of points with a distance of greater than 2 mm and less than 10 mm between corresponding points on the two fault displacement surfaces is designated as the warning zone; and the area consisting of points with a distance of greater than or equal to 10 mm between corresponding points on the two fault displacement surfaces is designated as the high-risk zone.

[0083] This completes the monitoring of deep foundation pit deformation.

[0084] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the principles of this application should be included within the protection scope of this application.

Claims

1. An intelligent automatic monitoring system for deformation of ultra-deep foundation pits, characterized in that, The system includes the following modules: The data acquisition module is used to collect the horizontal displacement, groundwater level, and support internal force at different monitoring points on the same fault in the area to be monitored for deformation of ultra-deep foundation pits at different acquisition times. The data anomaly probability acquisition module is used to record any collection time as the target collection time and any monitoring point as the target monitoring point. Based on the differences between the horizontal displacement, groundwater level and support internal force of the target monitoring point at the collection time adjacent to the target collection time, the first consistency of the target monitoring point at the target collection time is determined. Based on the differences between the first consistency of different monitoring points on the same fault at the same collection time and the distance between different monitoring points on the same fault, the data anomaly degree of each monitoring point at the same collection time is determined. Based on the first consistency of the target monitoring point at the target collection time and the collection time adjacent to the target collection time, the data anomaly degree is weighted and summed to determine the data anomaly probability of the target monitoring point at the target collection time. The fusion weight determination module is used to determine the fusion weight of each monitoring point on the same fault at each acquisition time based on the changing trend of data anomaly degree and the difference in the probability of data anomaly of all monitoring points on the same fault at adjacent acquisition times. The deep foundation pit deformation monitoring module is used to establish a fault displacement surface at the same acquisition time based on the fusion weight of each monitoring point at the same acquisition time, and to realize the monitoring of deep foundation pit deformation based on the difference between the fault displacement surfaces at two adjacent acquisition times.

2. The intelligent automatic monitoring system for deformation of ultra-deep foundation pits according to claim 1, characterized in that, The method for determining the first consistency of the target monitoring points at the target acquisition time is as follows: The difference between the horizontal displacement of the target monitoring point at the target acquisition time and the previous adjacent acquisition time is recorded as the adjacent horizontal displacement difference of the target monitoring point at the target acquisition time. The difference between the supporting internal forces at the target monitoring point at the target acquisition time and the previous adjacent acquisition time is recorded as the difference in adjacent supporting internal forces at the target monitoring point at the target acquisition time. When the difference between adjacent horizontal displacements and the difference between adjacent internal forces of adjacent supports at the target acquisition time are both positive or both negative, the first correlation coefficient of the target monitoring point at the target acquisition time is assigned a value of 1. When the difference between adjacent horizontal displacements and the difference between adjacent internal forces of adjacent supports at the target acquisition time are one positive and one negative, the first correlation coefficient of the target monitoring point at the target acquisition time is assigned a value of -1. Based on the groundwater level at the target monitoring point at the target acquisition time and the acquisition time adjacent to the target acquisition time, determine the characteristic value of the adjacent groundwater level at the target monitoring point at the target acquisition time. The product of the first correlation coefficient of the target monitoring point at the target acquisition time and the characteristic value of the adjacent groundwater level is denoted as the first consistency of the target monitoring point at the target acquisition time.

3. The intelligent automatic monitoring system for deformation of ultra-deep foundation pits according to claim 2, characterized in that, The specific steps for determining the characteristic value of the groundwater level adjacent to the target sampling time based on the groundwater level of the target monitoring point at the target sampling time and the sampling time adjacent to the target sampling time include: The difference between the groundwater level at the target monitoring point at the target acquisition time and the previous adjacent acquisition time is recorded as the adjacent groundwater level difference at the target monitoring point at the target acquisition time. The negative correlation processing result of the adjacent groundwater level difference at the target monitoring point at the target acquisition time is recorded as the adjacent groundwater level characteristic value at the target monitoring point at the target acquisition time.

4. The intelligent automatic monitoring system for deformation of ultra-deep foundation pits according to claim 1, characterized in that, The method of determining the data anomaly degree of each monitoring point at the same acquisition time based on the difference in the first consistency between different monitoring points on the same fault at the same acquisition time, and the distance between different monitoring points on the same fault, includes the following specific contents: The absolute value of the difference between the first consistency of the target monitoring point and other monitoring points at the target acquisition time is recorded as the first absolute difference between the target monitoring point and other monitoring points at the target acquisition time; the negative correlation processing result of the distance between the target monitoring point and other monitoring points is recorded as the distance feature value between the target monitoring point and other monitoring points. The product of the first absolute difference between the target monitoring point and other monitoring points at the target acquisition time and the distance feature value is denoted as the first product of the target monitoring point and other monitoring points at the target acquisition time. The data anomaly degree of the target monitoring point at the target acquisition time is determined by the first product of the target monitoring point and all other monitoring points at the target acquisition time.

5. The intelligent automatic monitoring system for deformation of ultra-deep foundation pits according to claim 4, characterized in that, The specific steps for determining the data anomaly degree of the target monitoring point at the target acquisition time based on the first product of the target monitoring point and all other monitoring points at the target acquisition time are as follows: The normalized value of the sum of the first products of the target monitoring point and all other monitoring points at the target acquisition time is denoted as the data anomaly degree of the target monitoring point at the target acquisition time.

6. The intelligent automatic monitoring system for deformation of ultra-deep foundation pits according to claim 1, characterized in that, The method for determining the probability of data anomalies at the target monitoring point at the target acquisition time is as follows: The first consistency normalized value of the target monitoring point at the previous adjacent collection time is used as the weight to perform a weighted summation of the data anomalies. The weighted summation result is recorded as the adjacent data anomaly of the target monitoring point at the target collection time. The probability of data anomalies at the target monitoring point at the target acquisition time is determined based on the weights of the data anomalies at the target acquisition time and the adjacent data anomalies at the target acquisition time.

7. The intelligent automatic monitoring system for deformation of ultra-deep foundation pits according to claim 6, characterized in that, The method for determining the probability of data anomalies at the target monitoring point at the target acquisition time based on the weights of the data anomaly degree of the target monitoring point at the target acquisition time and the anomalies of adjacent data at the target acquisition time includes the following specific methods: The preset first weight parameter is used as the weight of the data anomaly degree of the target monitoring point at the target acquisition time. The difference between the number 1 and the first weight parameter is used as the weight of the adjacent data anomaly degree of the target monitoring point at the target acquisition time. The weighted sum is recorded as the data anomaly probability of the target monitoring point at the target acquisition time.

8. The intelligent automatic monitoring system for deformation of ultra-deep foundation pits according to claim 1, characterized in that, The method for determining the fusion weight of each monitoring point on the same fault at each acquisition time is as follows: Linear fitting is performed on the data anomaly degree of the target monitoring point at the target acquisition time and the adjacent acquisition time before the target acquisition time with the corresponding acquisition time. The slope of the fitted line is obtained. The absolute value of the slope of the fitted line and the sum of the number 1 are recorded as the anomaly tendency of the target monitoring point at the target acquisition time. Based on the differences and tendencies of the probability of data anomalies at all monitoring points on the same fault at the target acquisition time, the fusion weight of the target monitoring point at the target acquisition time is determined.

9. The intelligent automatic monitoring system for deformation of ultra-deep foundation pits according to claim 8, characterized in that, The specific steps for determining the fusion weight of the target monitoring point at the target acquisition time based on the differences and tendencies of data anomalies among all monitoring points on the same fault at the target acquisition time are as follows: The mean of the absolute values ​​of the differences between the data anomaly probability of the target monitoring point and all other monitoring points on the same fault at the target acquisition time is denoted as the anomaly difference of the target monitoring point on the same fault at the target acquisition time; the ratio of the anomaly difference of the target monitoring point on the same fault at the target acquisition time to the anomaly tendency is denoted as the anomaly degree of the target monitoring point on the same fault at the target acquisition time. The negative correlation processing result of the same fault anomaly degree of the target monitoring point at the target acquisition time is recorded as the fusion weight of the target monitoring point at the target acquisition time.

10. The intelligent automatic monitoring system for deformation of ultra-deep foundation pits according to claim 1, characterized in that, The specific steps for establishing the fault displacement surface at the same acquisition time based on the fusion weight of each monitoring point at the same acquisition time are as follows: The fusion weights of each monitoring point at the same acquisition time are used as the weights of the horizontal displacement of each monitoring point. The inverse distance weighting method (IDW) is used for interpolation to obtain the fault displacement surface at the same acquisition time.

Images