Integrated building deformation state monitoring management method and system

Abstract

The application provides an integrated building deformation state monitoring management method and system, relates to the technical field of data management, and comprises the following steps: determining a monitoring device array based on a basic plan of a target integrated building and a foundation pit support design drawing; acquiring a monitoring data sequence array, and performing deformation state trend prediction; performing trend consistency division based on a deformation state trend prediction factor array; determining M first monitoring bandwidths and M second monitoring bandwidths according to M deformation state trend consistency prediction factors, constructing M monitoring double channels, and performing asynchronous monitoring management on M divided monitoring device sub-arrays by using the M monitoring double channels. The application solves the technical problem that, in the prior art, due to the contradiction between a uniform monitoring strategy and non-uniform deformation of a building structure, monitoring resource allocation is unreasonable, and then the efficiency of building safety management is affected, realizes comprehensive monitoring of the deformation state of a building, and improves the efficiency of building management.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data management, and particularly relates to an integrated building deformation state monitoring management method and system. BACKGROUND

[0002] In the field of deformation state monitoring of integrated buildings, an automatic monitoring method based on a sensor array is generally adopted, and a unified and homogeneous management strategy is usually implemented for all monitoring points, i.e., the same sampling frequency, data transmission period and alarm threshold are configured for the entire monitoring device array, which meets the basic data acquisition requirements to a certain extent. However, the deformation response of a building structure under actual load and environmental action has significant spatial heterogeneity and temporal dynamics, and the deformation rates and risk levels of different regions are quite different. There is a fundamental contradiction between the uniform monitoring strategy and the objective reality of the non-uniform deformation of the building structure, which leads to the fact that the monitoring resources cannot be optimally configured, the excessive monitoring of stable regions causes resource waste, and the monitoring of key risk regions may be insufficient, thereby reducing the early warning sensitivity and management efficiency of the overall deformation monitoring, and affecting the real-time and accuracy of building safety management.

[0003] In summary, in the prior art, due to the contradiction between the uniform monitoring strategy and the non-uniform deformation of the building structure, the monitoring resources are not reasonably allocated, and the real-time and accuracy of building safety management are affected. SUMMARY

[0004] The purpose of the present application is to provide an integrated building deformation state monitoring management method and system, so as to solve the technical problem in the prior art that due to the contradiction between the uniform monitoring strategy and the non-uniform deformation of the building structure, the monitoring resources are not reasonably allocated, and the real-time and accuracy of building safety management are affected.

[0005] In view of the above problems, the present application provides an integrated building deformation state monitoring management method and system.

[0006] In a first aspect, the application provides an integrated building deformation state monitoring management method, which is implemented by an integrated building deformation state monitoring management system. The integrated building deformation state monitoring management method comprises: determining a monitoring device array based on a target integrated building's base plan and foundation pit support design drawing; traversing to obtain a monitoring data sequence array of the monitoring device array within a preset monitoring window, performing deformation state trend prediction based on the monitoring data sequence array, and determining a deformation state trend prediction factor array; performing trend consistency division on the monitoring device array based on the deformation state trend prediction factor array, obtaining M divided monitoring device sub-arrays and M deformation state trend consistency prediction factors; determining M first monitoring bandwidths and M second monitoring bandwidths based on the M deformation state trend consistency prediction factors, constructing M monitoring double channels based on the M first monitoring bandwidths and the M second monitoring bandwidths, and performing asynchronous monitoring management on the M divided monitoring device sub-arrays by using the M monitoring double channels.

[0007] Optionally, deformation state feature analysis is performed on the monitoring data sequence array to obtain a deformation state feature sequence array; deformation state trend iteration is performed on the deformation state feature sequence array respectively to determine a deformation state trend iteration feature array; deformation state trend prediction is performed based on the deformation state trend iteration feature array to determine the deformation state trend prediction factor array.

[0008] Optionally, a first deformation state feature sequence is extracted from the deformation state feature sequence array; deformation state trend iteration is performed on a first deformation state feature and a second deformation state feature in the first deformation state feature sequence to obtain a first-stage deformation state trend iteration feature; deformation state trend iteration is performed on a third deformation state feature in the first deformation state feature sequence based on the first-stage deformation state trend iteration feature, and so on, to obtain a first deformation state trend iteration feature, which is added to a deformation state trend iteration feature array.

[0009] Optionally, fine-grained approximation analysis is performed on the first deformation state feature and the second deformation state feature by using an inner product mapping formula to determine a first fine-grained sub-feature similarity set; the first fine-grained sub-feature similarity set is normalized, and a first-stage deformation trend iteration matrix is constructed based on the processing result; deformation state trend iteration is performed on the second deformation state feature based on the first-stage deformation trend iteration matrix to obtain a first-stage deformation state trend iteration feature.

[0010] Optionally, M deformation state trend predictors are randomly extracted from the deformation state trend predictor array to obtain M deformation state trend predictor starting points; the M deformation state trend predictor starting points are subjected to trend consistency division in the deformation state trend predictor array to obtain M divided deformation state trend consistency predictor sets; the M divided deformation state trend consistency predictor sets are used to perform mapping division on the monitoring device array to obtain M divided monitoring device sub-arrays; mean shift analysis is performed on the M divided deformation state trend consistency predictor sets to obtain the M deformation state trend consistency predictors.

[0011] Optionally, dispersion identification is performed on the M deformation state trend predictor starting points; if the dispersion identification result meets preset requirements, M starting point neighborhoods are constructed according to a preset consistency bandwidth; edge diffusion is performed on the M starting point neighborhoods according to the preset consistency bandwidth to obtain M diffused starting point neighborhoods; it is determined whether the neighborhood density of the M diffused starting point neighborhoods is greater than or equal to the neighborhood density of the M starting point neighborhoods; if yes, the diffusion on the M diffused starting point neighborhoods is continued according to the preset consistency bandwidth until a preset diffusion number is met, and the M diffused starting point neighborhoods obtained by the last diffusion are taken as M initial divided deformation state trend consistency predictor sets; deformation state predictors in the deformation state trend predictor array that are not divided into the M initial divided deformation state trend consistency predictor sets are added into the initial divided deformation state trend consistency predictor set corresponding to the maximum similarity value in the initial divided deformation state trend consistency predictor set corresponding to the maximum similarity value in the M deformation state trend predictor starting points to obtain the M divided deformation state trend consistency predictor sets.

[0012] Optionally, the preset requirements are that the similarity between any two deformation state trend predictor starting points in the M deformation state trend predictor starting points is less than or equal to a preset similarity threshold.

[0013] Optionally, M matched deformation state abnormal logs are obtained based on the M deformation state trend consistency predictors; the M matched deformation state abnormal log sets are extracted with an abnormal interval length as an index to obtain M abnormal interval length sets; the M first monitoring bandwidths and the M second monitoring bandwidths are determined according to the M abnormal interval length sets.

[0014] Optionally, the maximum value in each of the M abnormal interval length sets is taken as a first monitoring bandwidth to obtain M first monitoring bandwidths; and the minimum value in each of the M abnormal interval length sets is taken as a second monitoring bandwidth to obtain M second monitoring bandwidths.

[0015] In a second aspect, the application further provides an integrated building deformation state monitoring management system for implementing the integrated building deformation state monitoring management method of the first aspect, wherein the integrated building deformation state monitoring management system comprises: a monitoring array determination module configured to determine a monitoring device array based on a foundation plan and a foundation pit support design of a target integrated building; a trend prediction module configured to traverse to obtain a monitoring data sequence array of the monitoring device array within a preset monitoring window, perform deformation state trend prediction based on the monitoring data sequence array, and determine a deformation state trend prediction factor array; a consistency division module configured to perform trend consistency division on the monitoring device array based on the deformation state trend prediction factor array to obtain M divided monitoring device sub-arrays and M deformation state trend consistency prediction factors; a dual-channel construction module configured to determine M first monitoring bandwidths and M second monitoring bandwidths based on the M deformation state trend consistency prediction factors, and construct M monitoring dual-channels based on the M first monitoring bandwidths and the M second monitoring bandwidths; and an asynchronous monitoring module configured to perform asynchronous monitoring management on the M divided monitoring device sub-arrays by using the M monitoring dual-channels.

[0016] The one or more technical solutions provided in the application have at least the following beneficial effects:

[0017] By determining a monitoring device array based on a foundation plan and a foundation pit support design of a target integrated building, traversing to obtain a monitoring data sequence array of the monitoring device array within a preset monitoring window, performing deformation state trend prediction based on the monitoring data sequence array, determining a deformation state trend prediction factor array, performing trend consistency division on the monitoring device array based on the deformation state trend prediction factor array to obtain M divided monitoring device sub-arrays and M deformation state trend consistency prediction factors, determining M first monitoring bandwidths and M second monitoring bandwidths based on the M deformation state trend consistency prediction factors, constructing M monitoring dual-channels based on the M first monitoring bandwidths and the M second monitoring bandwidths, and performing asynchronous monitoring management on the M divided monitoring device sub-arrays by using the M monitoring dual-channels, the overall monitoring of the deformation state of a building is achieved, and the safety management efficiency of the building is improved.

[0018] The above description is only a summary of the technical solutions of the present application. In order to make the technical means of the present application more clearly understood and implemented according to the content of the specification, and in order to make the above and other purposes, characteristics and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application are described below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only exemplary, and other drawings can be obtained by the provided drawings without creative labor for those skilled in the art.

[0020] Figure 1 The flowchart of the integrated building deformation state monitoring management method of the present application.

[0021] Figure 2 The structural schematic diagram of the integrated building deformation state monitoring management system of the present application.

[0022] Explanation of reference signs: monitoring array determination module 11, trend prediction module 12, consistency division module 13, double-channel construction module 14, asynchronous monitoring module 15. DETAILED DESCRIPTION

[0023] The present application provides an integrated building deformation state monitoring management method and system, which solves the technical problem that the real-time and accuracy of building safety management are affected due to the contradiction between the uniform monitoring strategy and the non-uniform deformation of building structure in the prior art, resulting in unreasonable allocation of monitoring resources. By designing a suitable monitoring device array, predicting the deformation state trend according to the monitoring data in the preset monitoring window, dividing the prediction factors according to the trend consistency, automatically adjusting the monitoring frequency and range according to the change of deformation trend, and asynchronously monitoring and managing M divided monitoring device sub-arrays, the overall monitoring of the deformation state of the building is realized, and the efficiency of building safety management is improved.

[0024] Below, the technical solutions in the present application will be described clearly and completely with reference to the drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. It should be understood that the present application is not limited by the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative work belong to the scope of protection of the present application. In addition, it should be noted that, for the convenience of description, only parts related to the present application are shown in the drawings, not all.

[0025] Embodiment one, please refer to the attached Figure 1 The present application provides an integrated building deformation state monitoring management method, wherein the integrated building deformation state monitoring management method is executed by an integrated building deformation state monitoring management system, and the integrated building deformation state monitoring management method specifically includes the following steps:

[0026] Based on the target integrated building's foundation plan and foundation pit support design drawing, the monitoring equipment array is determined.

[0027] Specifically, according to the target integrated building's foundation plan and foundation pit support design drawing, it is determined which areas are most prone to deformation or stress concentration, and the key parts are selected for the array layout of monitoring equipment, including the load-bearing walls of the building, the foundation columns, the support points of the foundation pit, etc. The layout of monitoring equipment should not be too dense or dispersed, but should be reasonably planned according to the specific needs of the building structure. Different types of sensors may need to be set at different locations, for example, strain gauges are used in the support structure of the foundation pit, while displacement meters are used to monitor settlement on the building foundation. Each monitoring equipment should have appropriate measurement range and sensitivity to ensure that it can capture the small deformation of the building during construction and use.

[0028] Through the key information provided by these drawings, the positions of the monitoring points and the types of the monitoring equipment are accurately selected to ensure that the monitoring equipment can cover the most needed parts of the building and the foundation pit. Suitable monitoring equipment is laid out at the selected monitoring points to form a monitoring equipment array, including various types of sensors. The foundation plan is used to describe the layout and design of the building foundation part, mainly including the size of the building, the position information of each support part. The foundation pit support design drawing shows the layout and design of the foundation pit support structure, usually including the specific form of the support structure, the depth, the position of the support points, etc.

[0029] Exemplarily, three observation reference points of the building are determined through the base plan and the foundation pit support design drawing. The observation reference points are measurement control points arranged outside the construction influence range and in a stable geological condition area. The coordinates thereof are fixed and unchangeable, serving as an absolute reference system for measuring displacement of all deformation points. Monitoring equipment is arranged at the key positions of the slope top of the foundation pit and the main body of the building. The slope top of the foundation pit is usually a key area of deformation, and therefore, 24 deformation monitoring points are arranged to monitor horizontal displacement and settlement. The main body of the building is selected to have 36 points for settlement observation, so as to ensure that the health condition of the main body of the building can be comprehensively reflected. At the 24 monitoring points of the slope of the foundation pit, deformation monitoring equipment is arranged to mainly monitor horizontal displacement and settlement, and to monitor displacement, settlement and other deformations that may occur in the construction process of the foundation pit, especially the stability of the slope of the foundation pit. At the 36 settlement observation points of the main body of the building, settlement monitoring equipment is arranged to record the settlement amount of the building in real time, so as to determine whether the building has deformation problems such as sinking and tilting. The 24 monitoring points of the slope of the foundation pit are monitored for horizontal displacement, and the monitoring cycle is planned to be 15 months, and the total number of monitoring times is 115. Through periodic monitoring, it can be found in time whether the slope of the foundation pit has abnormal deformation, so as to ensure the safety of the construction. The main body of the building is observed for settlement, and it is expected that the total number of monitoring times will be 116 within 15 months. Through monitoring of the settlement change of the main body of the building, the stability of the building in the use process can be ensured.

[0030] In the actual monitoring process, when the monitoring data is abnormal, such as excessive deformation of the foundation pit or abnormal settlement of the building, some monitoring points or equipment may need to be adjusted or the monitoring frequency may need to be increased. For example, if an abnormal displacement occurs at a slope of the foundation pit, the number of monitoring times at the point is increased or an additional monitoring point is arranged. Through the array arrangement of the monitoring equipment at the key positions of the slope of the foundation pit and the main body of the building, the deformation, settlement and other conditions of the building and the foundation pit can be comprehensively monitored. Based on the reasonable planning of the base plan and the foundation pit support design drawing, the layout of the monitoring equipment is scientific, and the key risk areas of the building and the foundation pit are covered. Through periodic monitoring, abnormal changes of the building and the foundation pit can be found in time, potential risks can be early warned, and the safety in the construction and use process of the building can be ensured.

[0031] The monitoring data sequence array of the monitoring equipment array in a preset monitoring window is obtained through traversal, and a deformation state trend prediction factor array is determined according to the monitoring data sequence array.

[0032] Further, the application further comprises the following steps: traversing the deformation state feature sequence array to obtain a deformation state feature sequence array; performing intra-sequence deformation state trend iteration on the deformation state feature sequence array respectively to determine a deformation state trend iteration feature array; performing deformation state trend prediction according to the deformation state trend iteration feature array to determine the deformation state trend prediction factor array.

[0033] Further, the application further comprises the following steps: extracting a first deformation state feature sequence from the deformation state feature sequence array; performing deformation state trend iteration on the first deformation state feature sequence to obtain a first-stage deformation state trend iteration feature; performing deformation state trend iteration on the third deformation state feature in the first deformation state feature sequence based on the first-stage deformation state trend iteration feature, and so on to obtain a first deformation state trend iteration feature, and adding the first deformation state trend iteration feature into the deformation state trend iteration feature array.

[0034] Further, the application further comprises the following steps: performing fine-grained approximation analysis on the first deformation state feature and the second deformation state feature by using an inner product mapping formula to determine a first fine-grained sub-feature similarity set; performing normalization processing on the first fine-grained sub-feature similarity set, and constructing a first-stage deformation trend iteration matrix according to the processing result; performing deformation state trend iteration on the second deformation state feature according to the first-stage deformation trend iteration matrix to obtain a first-stage deformation state trend iteration feature.

[0035] Specifically, a preset monitoring window, i.e., a monitoring time period for analysis, is determined. A monitoring data sequence array in the preset monitoring window, i.e., a data set arranged in time sequence collected by the monitoring device of each monitoring point in the preset monitoring window, is obtained from the monitoring device array. For example, the settlement values of foundation observation point B07 in the past 5 days are 1.2mm, 1.5mm, 1.9mm, 2.4mm, 3.0mm, etc., and the time sequence of all 63 points constitutes a monitoring data sequence array.

[0036] The deformation state feature analysis is performed on the monitoring data sequence array, and the characteristic parameters of deformation are extracted. For example, by calculating the change rate, fluctuation amplitude and other characteristics, the dynamic law of structural deformation is identified. Assuming that the measurement of the monitoring point is settlement data, the difference between the data of two consecutive days is calculated as the deformation rate, such as from 0.5 mm to 0.8 mm, the deformation rate is 0.3 mm / day; the acceleration, i.e. the change of the change of velocity, is obtained by calculating the change of the deformation rate, which is used to detect whether the deformation is accelerating or slowing down; the abnormal fluctuation of deformation is identified by analyzing the fluctuation amplitude of the data. All the deformation state features extracted from the monitoring data will be arranged in time sequence, and finally form a deformation state feature sequence array.

[0037] Any one of the deformation state feature sequences in the deformation state feature sequence array is extracted as the first deformation state feature sequence. The first deformation state feature sequence does not refer to the first one, but refers to any one of the multiple feature sequences provided by the monitoring device. The first two data points, i.e. the first deformation state feature and the second deformation state feature, are extracted from the first deformation state feature sequence, and the deformation state trend iteration analysis is performed to identify the deformation trend between the two data points.

[0038] The first deformation state feature and the second deformation state feature are analyzed by inner product mapping formula for fine-grained approximation, i.e. calculating the similarity of the sub-features in their corresponding dimensions. Inner product mapping refers to measuring the similarity between two vectors by calculating the inner product of the two vectors. In deformation state feature analysis, inner product is often used to measure the similarity between two deformation state features. The greater the inner product value of two features, the higher their similarity. For example, assuming that the deformation state feature of each day is defined as a 3-dimensional vector: settlement velocity mm / day, horizontal displacement rate mm / day, temperature ℃. To compare the displacement changes of monitoring point 1 and monitoring point 2 at different time points, use inner product to calculate their similarity in a certain time window: assuming that the first deformation state feature is 1.5, 0.2, 25.0, and the second deformation state feature is 1.9, 0.3, 26.5. Calculate the similarity of the three dimensions using cosine similarity: the settlement rate dimension similarity is 0.98, the horizontal displacement dimension similarity is 0.95, and the temperature dimension similarity is 0.99, then the first fine-grained sub-feature similarity set is 0.98, 0.95, 0.99. If the inner product value is large, it means that the two features are similar, otherwise they are quite different.

[0039] The first fine-grained sub-feature similarity set is normalized. In order to make the data scales of all features consistent, it is normalized to a value in the range of [0, 1]. Based on the normalized similarity data, a deformation trend iteration matrix is constructed to define how to fuse the feature information of the previous state into the feature of the current state, reflecting the dynamic evolution trend of the feature sequence. For example, the first fine-grained sub-feature similarity set 0.98, 0.95, 0.99 is normalized using the softmax function to obtain the weight set 0.335, 0.318, 0.347. It can be seen that the temperature change obtains slightly higher attention in this iteration.

[0040] According to the first stage deformation trend iteration matrix, the second deformation state feature is iterated for deformation state trend iteration. The neural network model is used to iteratively predict the deformation state trend, that is, to predict the deformation state of the next time step according to the current deformation state. The relationship between the deformation state feature and the subsequent change is learned through the neural network. The first stage deformation trend iteration matrix and the second deformation state feature are input into a dedicated neural network model. The neural network model acts as a complex trend fusion device. Its internal weights have been pre-trained and can understand the change pattern from the first deformation state feature to the second deformation state feature encoded by the iteration matrix. The model performs nonlinear changes on the second deformation state feature, and the change mode is strongly guided by the first stage deformation trend iteration matrix. Finally, a brand new first stage deformation state trend iteration feature is output. This new feature is a trend-enhanced version of the original second deformation state feature, which emphasizes the continuity and directionality of changes rather than isolated instantaneous states.

[0041] According to the first stage deformation state trend iteration feature, the next data point is analyzed for trend iteration. The same operation is performed on the next feature each time, and the cycle continues until the entire sequence is processed, thereby obtaining the first deformation state trend iteration feature. The first deformation state trend iteration feature is the deformation state trend of the first deformation state feature sequence. The above operation is repeated on other deformation state feature sequences in the deformation state feature sequence array until the deformation state trend iteration feature corresponding to each deformation state feature sequence is obtained. All of them are added to the deformation state trend iteration feature array. The deformation state trend iteration feature array is a set composed of multiple deformation state trend iteration features, including the deformation state trend of each monitoring point in the preset monitoring window.

[0042] Exemplarily, assuming that the settlement data of a certain foundation pit monitoring point in the past 10 days are respectively 12.2 mm, 12.5 mm, 12.9 mm, 13.2 mm, 13.7 mm, 14.1 mm, 14.3 mm, 14.8 mm, 15.1 mm, and 15.6 mm, the deformation state feature analysis is performed, the settlement rates are calculated as 0.3 mm / day, 0.4 mm / day, 0.3 mm / day, 0.5 mm / day, 0.4 mm / day, 0.2 mm / day, 0.5 mm / day, 0.3 mm / day, 0.5 mm / day, and 0.5 mm / day, and a deformation state feature sequence array is formed. Through inner product mapping formula and normalization, the deformation state trend iteration feature is obtained as 0.45 mm / day, which fuses the data of the past 10 days and reflects the possible deformation speed of the monitoring point in a future period of time.

[0043] Based on the deformation state trend iteration feature array, the future deformation state is predicted through a mathematical model, the deformation trend in a future period of time is estimated, and a corresponding prediction factor is generated. The deformation state trend prediction factor array is the result output by the prediction model, and these factors represent the future deformation trend prediction of each monitoring point, including the prediction deformation type, confidence, and prediction amplitude. For example, assuming that the deformation data of foundation pit monitoring point B07 is analyzed. Through the previous deformation state feature sequence and trend iteration calculation, the deformation state trend iteration features of B07 are obtained as I1=1.72, I2=2.12, and I3=2.62, which represent that the deformation trend of the monitoring point has gradually accelerated. The LSTM model is used for time series prediction, and the LSTM model outputs the prediction deformation rates in the future 3 days through training and prediction. Assuming that the model prediction result is 3.4 mm / day, 3.8 mm / day, and 4.0 mm / day. According to the prediction result, the deformation state trend prediction factor is generated as the prediction type is accelerated settlement, the confidence is 0.90, and the prediction amplitude is 3.4−4.0 mm / day, which represents the deformation trend of monitoring point B07 in a future period of time.

[0044] Through deformation state feature analysis and trend iteration, the original monitoring data is compressed into concise trend features, the complexity of the data is reduced, the key mode of deformation is retained, and the prediction is more accurate. Through the prediction of the deformation state trend, the monitoring frequency and mode are adjusted according to the different deformation states of the monitoring points, the resource utilization efficiency is improved, and the individualized monitoring strategy is realized.

[0045] Based on the deformation state trend prediction factor array, the monitoring device array is divided into trend consistency, M divided monitoring device sub-arrays and M deformation state trend consistency prediction factors are obtained.

[0046] Further, the application further comprises the following steps: randomly extracting M deformation state trend prediction factors from the deformation state trend prediction factor array to obtain M deformation state trend prediction factor starting points; performing trend consistency division in the deformation state trend prediction factor array based on the M deformation state trend prediction factor starting points to obtain M divided deformation state trend consistency prediction factor sets; performing mapping division on the monitoring device array according to the M divided deformation state trend consistency prediction factor sets to obtain M divided monitoring device sub-arrays; performing mean shift analysis on the M divided deformation state trend consistency prediction factor sets to obtain the M deformation state trend consistency prediction factors.

[0047] Further, the application further comprises the following steps: performing dispersion identification on the M deformation state trend prediction factor starting points, if the dispersion identification result meets a preset requirement, constructing M starting point neighborhoods according to a preset consistency bandwidth; and performing edge diffusion on the M starting point neighborhoods according to the preset consistency bandwidth to obtain M diffused starting point neighborhoods; judging whether the neighborhood density of the M diffused starting point neighborhoods is greater than or equal to the neighborhood density of the M starting point neighborhoods, if yes, continuing to diffuse the M diffused starting point neighborhoods according to the preset consistency bandwidth until a preset diffusion number of times is met, taking the M diffused starting point neighborhoods obtained by the last diffusion as the M initial divided deformation state trend consistency prediction factor sets; adding the deformation state prediction factors in the deformation state trend prediction factor array that are not divided into the M initial divided deformation state trend consistency prediction factor sets into the initial divided deformation state trend consistency prediction factor set corresponding to the maximum similarity value in the initial divided deformation state trend consistency prediction factor set corresponding to the maximum similarity value in the M deformation state trend prediction factor starting points to obtain the M divided deformation state trend consistency prediction factor sets.

[0048] Further, the application further comprises the following steps: the preset requirement is that the similarity between any two deformation state trend prediction factor starting points in the M deformation state trend prediction factor starting points is less than or equal to a preset similarity threshold.

[0049] Specifically, a part of the trend predictors of deformation state is randomly selected as the starting point of analysis, i.e. M trend predictors of deformation state are extracted as the starting points of M trend predictors of deformation state. M is an integer greater than or equal to 1. Before starting to divide the trend predictors of deformation state, it is ensured that the selected starting points are dispersed in the data space, i.e. each starting point represents a unique trend, avoiding all starting points being concentrated in the same group, thereby affecting the quality of subsequent grouping. By calculating the similarity between the trend predictors of deformation state, if the dispersion identification result meets the preset requirement, the neighborhood of the M starting points is constructed according to the preset consistency bandwidth. For example, it is assumed that X1(1.0) and X2(3.5) are randomly selected, the Euclidean distance between X1 and X2 is calculated to be 2.5, which is less than the preset threshold 5.0, so the dispersion identification result meets the preset requirement, and X1 and X2 are taken as two starting points.

[0050] It is ensured that the selection of the starting points can represent different trend patterns, avoiding all starting points being concentrated in similar areas, thereby affecting the subsequent division effect. The preset requirement is that the similarity between any two starting points of the M trend predictors of deformation state is less than or equal to the preset similarity threshold. The preset similarity threshold is a standard for measuring the similarity between two trend predictors of deformation state. If the similarity between certain starting points exceeds the preset similarity threshold, it means that these starting points are too similar in trend and cannot be used as independent starting points.

[0051] When the M deformation state trend predictor starting points meet the preset requirements, the neighborhood of the starting points is further divided according to the preset consistency bandwidth, that is, the M starting points are taken as the center and the preset consistency bandwidth is taken as the radius to construct the M starting point neighborhoods. The preset consistency bandwidth is used to control the tolerance of the trend consistency division, and defines the maximum similarity difference that can be tolerated between certain deformation state predictors in the neighborhood. The greater the bandwidth, the wider the range of the neighborhood, and the more points can be accommodated; the smaller the bandwidth, the smaller the range of the neighborhood. In other words, all points with a distance less than or equal to the preset consistency bandwidth from the starting point are found and added to the starting point neighborhood. For example, all points with a distance less than or equal to 1.0 from X1 are found, such as F1 with a distance of 0.3 < 1.0 from X1, F2 with a distance of 0.5 < 1.0 from X1, F3 with a distance of 2.5 > 1.0 from X1, F4 with a distance of 2.8 > 1.0 from X1, F5 with a distance of 2.2 > 1.0 from X1, F6 with a distance of 0.53 < 1.0 from X1, and F7 with a distance of 0.45 < 1.0 from X1. The initial neighborhood of X1 is {F1, F2, F6, F7} and the neighborhood density is 4.

[0052] The M starting point neighborhoods are edge-diffused according to the preset consistency bandwidth, gradually expanding the range of the neighborhood and including more predictors to obtain M diffused starting point neighborhoods. The neighborhood density of the M diffused starting point neighborhoods and the neighborhood density of the M starting point neighborhoods are calculated, that is, the number of deformation state trend predictors included in the M diffused starting point neighborhoods and the M starting point neighborhoods. If the neighborhood density of the M diffused starting point neighborhoods is greater than or equal to the neighborhood density of the M starting point neighborhoods, that is, the neighborhood density after diffusion is greater, it means that the diffusion is effective, and the diffusion continues until the preset diffusion number is met. Each diffusion gradually expands the range of the neighborhood and increases the included deformation state trend predictors. The preset diffusion number refers to the maximum number of execution of the neighborhood diffusion operation. Each diffusion gradually increases the range of the neighborhood and thus adds more similar predictors.

[0053] The neighborhood of the M diffusion starting points obtained in the last diffusion is taken as the M initial partition deformation state trend consistency predictor set. The deformation state predictors in the deformation state trend predictor array that are not partitioned into the M initial partition deformation state trend consistency predictor set are added to the initial partition deformation state trend consistency predictor set corresponding to the maximum similarity of the M deformation state trend predictor starting points, that is, according to the rule of the closest similarity, they are distributed into an existing neighborhood. Each unpartitioned predictor is added to the most similar neighborhood set, ensuring that all predictors have a home, thereby obtaining the M partition deformation state trend consistency predictor sets.

[0054] Exemplarily, it is assumed that two starting points S1(3.0, 0.2) and S2(0.5, 0.1) are randomly selected. The Euclidean distance between the two is much larger than the preset similarity threshold, and the dispersion meets the requirement. A circle with S1 as the center and a preset consistency bandwidth 1.0 as the radius is drawn to find the points in the neighborhood, forming the starting point neighborhood 1 {S1, A, B}. Similarly, a circle with S2 as the center is drawn to form the starting point neighborhood 2 {S2, C, D, E}. The radius of neighborhood 1 is expanded to 2.0 to obtain the diffusion neighborhood 1, which newly contains point F and the density increases from 3 to 4. The density increases, and the diffusion is effective. The radius of neighborhood 2 is expanded to 2.0 to obtain the diffusion neighborhood 2, which newly contains point G. The density increases from 4 to 5. The density increases, and the diffusion is effective. The radius of the diffusion neighborhood 1 is expanded to 3.0 to obtain a new diffusion neighborhood, and no new point falls into it. The density remains 4. The density does not increase, and the preset diffusion number (2 times) has been reached, so the diffusion is stopped. Similarly, the density of the diffusion neighborhood 2 does not increase after diffusion, and the diffusion is stopped. The neighborhoods 1 and 2 obtained in the last diffusion are taken as the initial partition set. At this time, there are remaining points P and Q. For each unpartitioned deformation state trend predictor, the similarity with the M initial partition points is calculated, the most similar initial partition point is selected, and it is added to the corresponding partition set. For example, the similarity of P with the diffusion starting point neighborhood M1 is 0.75, the similarity of P with the diffusion starting point neighborhood M2 is 0.81, and the similarity of P with the diffusion starting point neighborhood M3 is 0.74. Therefore, the similarity of P with the diffusion starting point neighborhood M2 is the highest, and P should be added to the diffusion starting point neighborhood M2. The similarity of Q with the diffusion starting point neighborhood M1 is 0.60, the similarity of P with the diffusion starting point neighborhood M2 is 0.75, and the similarity of P with the diffusion starting point neighborhood M3 is 0.62. Therefore, the similarity of P with the diffusion starting point neighborhood M2 is the highest, and P should be added to the diffusion starting point neighborhood M2.

[0055] According to the M set of deformation state trend consistency prediction factors, the monitoring device array is mapped and divided, and the devices in the monitoring device array are mapped into corresponding sub-arrays to obtain M divided monitoring device sub-arrays. For each set of divided deformation state trend consistency prediction factors, mean shift analysis is performed. The core of the mean shift algorithm is to find the area with higher data density by moving data points. Based on the position of each factor and the density of other factors around it, the position of each factor is continuously adjusted until all factors converge to the area with the highest density. Mean shift is a density-based clustering algorithm that identifies the clustering center of data points by iteratively moving data points to the position of the density peak.

[0056] The density of the neighborhood around each prediction factor is calculated. If the position of a certain factor belongs to an area with lower density, it will drift towards an area with higher density along the density gradient. The mean shift algorithm repeatedly updates the position of each prediction factor until the position of the factor converges. The converged factor position is the density peak region it represents. After mean shift analysis, all prediction factors will converge to areas with higher density, thereby obtaining M deformation state trend consistency prediction factors. Part of the data of M divided monitoring device sub-arrays and M deformation state trend consistency prediction factors obtained in an experimental example is shown in Table 1:

[0057] Table 1 Deformation state trend consistency prediction factor and monitoring device division table

[0058] Monitoring point ID Monitoring device ID Deformation state trend predictor Partition deformation state trend consistency predictor Partition monitoring device subarray ID J1 D1 0.32 C1 S1 J1 D2 0.29 C1 S1 J2 D5 0.42 C2 S2 J2 D6 0.45 C2 S2 J3 D9 0.25 C1 S1 J4 D15 0.24 C1 S1 J5 D21 0.48 C2 S2

[0059] By randomly extracting deformation state trend prediction factors and performing trend consistency division, and then refining the division results through mean shift analysis, the subtle changes in deformation trend can be accurately captured, which not only improves the accuracy of monitoring, but also enhances the adaptability and early warning capability in the face of complex deformation conditions.

[0060] According to the M deformation state trend consistency prediction factors, M first monitoring bandwidths and M second monitoring bandwidths are determined, and M monitoring double channels are constructed based on the M first monitoring bandwidths and the M second monitoring bandwidths.

[0061] Further, the application further includes the following steps: matching deformation state abnormal logs based on the M deformation state trend consistency prediction factors to obtain a set of M matched deformation state abnormal logs; extracting the set of M matched deformation state abnormal logs with an abnormal interval length as an index to obtain a set of M abnormal interval lengths; and determining the M first monitoring bandwidths and the M second monitoring bandwidths according to the set of M abnormal interval lengths.

[0062] Further, the application further comprises the following steps: taking the maximum value in each of the M sets of abnormal interval durations as a first monitoring bandwidth, thereby obtaining M first monitoring bandwidths; and taking the minimum value in each of the M sets of abnormal interval durations as a second monitoring bandwidth, thereby obtaining M second monitoring bandwidths.

[0063] Specifically, according to the M deformation state trend consistency predictors, abnormal logs related to these factors are searched in the monitoring data, and M sets of matched deformation state abnormal logs containing historical abnormal events of each monitoring point are obtained by comparing the similarity of the predictors and the historical abnormal logs. The abnormal logs contain information such as timestamp, deformation amplitude, and speed of the deformation state, which are used to track the time point and trend of the abnormal occurrence.

[0064] The abnormal interval duration is the time interval between two events of abnormal deformation state. According to the abnormal interval duration, the M sets of matched deformation state abnormal logs are extracted, and the time interval between two abnormal events in each matched log is extracted, thereby obtaining M sets of abnormal interval durations. For each monitoring point, a set of historical abnormal interval durations is obtained.

[0065] The maximum value of each abnormal interval duration in the M sets of abnormal interval durations is taken as a first monitoring bandwidth, thereby obtaining M first monitoring bandwidths. Similarly, the minimum value of each abnormal interval duration in the M sets of abnormal interval durations is taken as a second monitoring bandwidth, thereby obtaining M second monitoring bandwidths. The first monitoring bandwidth refers to the maximum value of the abnormal interval duration of the monitoring point in all abnormal events, which reflects the required inspection frequency of the monitoring device in the slowest change, i.e., regular patrol inspection of the monitoring point within the longest abnormal interval. The second monitoring bandwidth refers to the minimum value of the abnormal interval duration of the monitoring point in all abnormal events, which represents the need for high-frequency monitoring of the monitoring device within a shorter time interval in order to capture possible rapid changes.

[0066] Illustratively, assuming that the abnormal logs of a foundation pit monitoring point are processed, the abnormal interval of abnormal event 1 is 3 days; the abnormal interval of abnormal event 2 is 5 days; the abnormal interval of abnormal event 3 is 2 days; and the abnormal interval of abnormal event 4 is 7 days. For this monitoring point, the maximum abnormal interval is 7 days, and therefore the first monitoring bandwidth is 7 days; the minimum abnormal interval is 2 days, and therefore the second monitoring bandwidth is 2 days. By taking the maximum and minimum values of the abnormal interval duration as the monitoring bandwidths, respectively, appropriate patrol frequencies can be customized for different types of monitoring points. For example, for a slowly changing monitoring point, a longer first monitoring bandwidth can save resources; and for a rapidly changing monitoring point, a shorter second monitoring bandwidth can timely capture abnormalities.

[0067] According to the deformation state and historical data of each monitoring device, a monitoring double channel is constructed for each device, and a first monitoring bandwidth and a second monitoring bandwidth are allocated to each device, that is, M monitoring double channels are constructed according to M first monitoring bandwidths and M second monitoring bandwidths. Using the first monitoring bandwidth, the monitoring is performed at a regular frequency according to the historical deformation trend of the monitoring point; when the device detects abnormal changes, the second monitoring bandwidth is switched to for higher frequency monitoring. The monitoring device will perform monitoring at different frequencies within different time windows. By allocating different monitoring channels and bandwidths to each device, the monitoring frequency is flexibly adjusted. In normal cases, the monitoring device operates at a lower frequency, saving resources; while in the case of abnormal deformation trend, it can quickly respond and monitor at a higher frequency to ensure that potential risks are captured in time.

[0068] The M monitoring double channels are used to perform asynchronous monitoring management on the M divided monitoring device subarrays.

[0069] Specifically, the M monitoring double channels are used to perform asynchronous monitoring management on the M divided monitoring device subarrays. The monitoring devices of each subarray perform data collection according to their set monitoring channels, but the monitoring frequencies of different devices can be asynchronous. For example, device A of subarray 1 will perform regular monitoring every 10 days, but will switch to high-frequency monitoring every 2 days when an abnormality occurs; at the same time, device B of subarray 2 may perform regular monitoring every 5 days, and if an abnormal change occurs, it will switch to high-frequency monitoring every 1 day.

[0070] Each monitoring device is asynchronously scheduled to ensure that each device independently performs the monitoring task according to its set frequency without waiting for the monitoring results of other devices. For example, device A will perform regular monitoring on the 10th day, and if the deformation trend accelerates, it will switch to high-frequency monitoring on the 12th day; device B will perform regular monitoring on the 5th day, and if an abnormality is found, it will immediately switch to high-frequency monitoring on the 6th day. Although the monitoring of each device is asynchronous, the monitoring data of all devices is integrated for unified analysis and response. If the monitoring result of a device indicates that the deformation state enters a dangerous zone, high-frequency monitoring is automatically triggered through a preset threshold or prediction factor, and synchronous warning and intervention are performed on the entire monitoring array.

[0071] Through asynchronous monitoring management, resource waste caused by synchronous monitoring of all devices is avoided. Each device performs the monitoring task at an appropriate time point, thereby achieving efficient use of resources. When deformation abnormalities occur, the deformation changes are quickly responded to, and the high-frequency monitoring mode is automatically switched to, ensuring that important data is not missed. At the same time, other devices still independently perform tasks according to their regular monitoring periods, ensuring the flexibility and real-time performance of the whole.

[0072] In summary, the integrated building deformation state monitoring management method provided by the present application has the following beneficial effects:

[0073] By means of the base plan and the foundation pit support design map of the target integrated building, the monitoring device array is determined; the monitoring data sequence array of the monitoring device array in a preset monitoring window is obtained through traversal, the deformation state trend prediction is performed according to the monitoring data sequence array, and the deformation state trend prediction factor array is determined; the monitoring device array is divided in terms of trend consistency based on the deformation state trend prediction factor array, and M divided monitoring device sub-arrays and M deformation state trend consistency prediction factors are obtained; the M first monitoring bandwidths and the M second monitoring bandwidths are determined according to the M deformation state trend consistency prediction factors, the M monitoring double channels are constructed based on the M first monitoring bandwidths and the M second monitoring bandwidths, and the M divided monitoring device sub-arrays are monitored and managed asynchronously by means of the M monitoring double channels. That is to say, the appropriate monitoring device array is determined by means of the design map, the deformation state trend prediction is performed according to the monitoring data in the preset monitoring window, the prediction factors are divided in terms of trend consistency, the monitoring frequency and range are automatically adjusted according to the deformation trend change, the M divided monitoring device sub-arrays are monitored and managed asynchronously, the comprehensive monitoring of the building deformation state is realized, and the building safety management efficiency is improved.

[0074] In the embodiment two, based on the same inventive concept as the integrated building deformation state monitoring management method in the aforementioned embodiment one, the present application further provides an integrated building deformation state monitoring management system. Please refer to the accompanying drawings Figure 2 The integrated building deformation state monitoring management system comprises:

[0075] The monitoring array determination module 11 is configured to determine the monitoring device array based on the base plan and the foundation pit support design map of the target integrated building; the trend prediction module 12 is configured to obtain the monitoring data sequence array of the monitoring device array in a preset monitoring window through traversal, perform the deformation state trend prediction according to the monitoring data sequence array, and determine the deformation state trend prediction factor array; the consistency division module 13 is configured to divide the monitoring device array in terms of trend consistency based on the deformation state trend prediction factor array, and obtain M divided monitoring device sub-arrays and M deformation state trend consistency prediction factors; the double channel construction module 14 is configured to determine the M first monitoring bandwidths and the M second monitoring bandwidths according to the M deformation state trend consistency prediction factors, and construct the M monitoring double channels based on the M first monitoring bandwidths and the M second monitoring bandwidths; and the asynchronous monitoring module 15 is configured to monitor and manage the M divided monitoring device sub-arrays asynchronously by means of the M monitoring double channels.

[0076] Further, the trend prediction module 12 in the integrated building deformation state monitoring management system is further configured to: traverse the monitoring data sequence array to perform deformation state feature analysis, and obtain a deformation state feature sequence array; perform intra-sequence deformation state trend iteration on the deformation state feature sequence array respectively, and determine a deformation state trend iteration feature array; and perform deformation state trend prediction based on the deformation state trend iteration feature array, and determine the deformation state trend prediction factor array.

[0077] Further, the trend prediction module 12 in the integrated building deformation state monitoring management system is further configured to: extract a first deformation state feature sequence from the deformation state feature sequence array; perform deformation state trend iteration on a first deformation state feature and a second deformation state feature in the first deformation state feature sequence, and obtain a first-stage deformation state trend iteration feature; perform deformation state trend iteration on a third deformation state feature in the first deformation state feature sequence based on the first-stage deformation state trend iteration feature, and so on, to obtain a first deformation state trend iteration feature, and add the first deformation state trend iteration feature into a deformation state trend iteration feature array.

[0078] Further, the trend prediction module 12 in the integrated building deformation state monitoring management system is further configured to: perform fine-grained approximation analysis on the first deformation state feature and the second deformation state feature by using an inner product mapping formula, to determine a first fine-grained sub-feature similarity set; perform normalization processing on the first fine-grained sub-feature similarity set, and construct a first-stage deformation trend iteration matrix based on the processing result; and perform deformation state trend iteration on the second deformation state feature based on the first-stage deformation trend iteration matrix, to obtain a first-stage deformation state trend iteration feature.

[0079] Further, the consistency division module 13 in the integrated building deformation state monitoring management system is further configured to: randomly extract M deformation state trend prediction factors from the deformation state trend prediction factor array, to obtain M deformation state trend prediction factor starting points; perform trend consistency division in the deformation state trend prediction factor array based on the M deformation state trend prediction factor starting points, to obtain M divided deformation state trend consistency prediction factor sets; perform mapping division on the monitoring device array based on the M divided deformation state trend consistency prediction factor sets, to obtain M divided monitoring device sub-arrays; and perform mean shift analysis on the M divided deformation state trend consistency prediction factor sets, to obtain the M deformation state trend consistency prediction factors.

[0080] Further, the consistency division module 13 in the integrated building deformation state monitoring management system is further used for: dispersively identifying the M deformation state trend predictor starting points, if the dispersively identified result meets preset requirements, constructing M starting point neighborhoods according to a preset consistency bandwidth; and performing edge diffusion on the M starting point neighborhoods according to the preset consistency bandwidth to obtain M diffused starting point neighborhoods; judging whether the neighborhood density of the M diffused starting point neighborhoods is greater than or equal to the neighborhood density of the M starting point neighborhoods, if yes, continuing to diffuse the M diffused starting point neighborhoods according to the preset consistency bandwidth until a preset diffusion number is met, taking the M diffused starting point neighborhoods obtained by the last diffusion as M initial division deformation state trend consistency predictor sets; adding the deformation state predictors in the deformation state trend predictor array that are not divided into the M initial division deformation state trend consistency predictor sets into the initial division deformation state trend consistency predictor set corresponding to the maximum similarity value in the M deformation state trend predictor starting points, to obtain M division deformation state trend consistency predictor sets.

[0081] Further, the consistency division module 13 in the integrated building deformation state monitoring management system is further used for: the preset requirements are that the similarity between any two deformation state trend predictor starting points in the M deformation state trend predictor starting points is less than or equal to a preset similarity threshold.

[0082] Further, the dual-channel construction module 14 in the integrated building deformation state monitoring management system is further used for: based on the M deformation state trend consistency predictors, matching deformation state abnormal logs are obtained to obtain M matched deformation state abnormal log sets; taking an abnormal interval length as an index, the M matched deformation state abnormal log sets are extracted to obtain M abnormal interval length sets; and according to the M abnormal interval length sets, the M first monitoring bandwidths and the M second monitoring bandwidths are determined.

[0083] Further, the dual-channel construction module 14 in the integrated building deformation state monitoring management system is further used for: respectively taking the maximum value in the M abnormal interval length sets as a first monitoring bandwidth to obtain M first monitoring bandwidths; and respectively taking the minimum value in the M abnormal interval length sets as a second monitoring bandwidth to obtain M second monitoring bandwidths.

[0084] Each of the embodiments in the specification is described in a progressive manner, and each embodiment focuses on the difference from other embodiments. The foregoing Figure 1The integrated building deformation state monitoring management method and the specific examples in the embodiment one are also applicable to the integrated building deformation state monitoring management system in the embodiment, and through the foregoing detailed description of the integrated building deformation state monitoring management method, those skilled in the art can clearly know the integrated building deformation state monitoring management system in the embodiment, so as to avoid the description of the integrated building deformation state monitoring management system in the embodiment again.

[0085] The above description of disclosed embodiments enables one skilled in the art to make or use the application. Numerous modifications to these embodiments would be apparent to those skilled in the art, and the generic principles defined herein can be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0086] Obviously, for those skilled in the art, some improvements and modifications can be made to the application without departing from the principles of the application, and these improvements and modifications also fall within the protection scope of the application.

Claims

1. An integrated building deformation state monitoring management method, characterized by, The method comprises the following steps: Based on the foundation plan and foundation pit support design of the target integrated building, determine the monitoring device array; Iterate to obtain the monitoring data sequence array of the monitoring device array within the preset monitoring window; Iterate the monitoring data sequence array to perform deformation state feature analysis and obtain a deformation state feature sequence array; Extract a first deformation state feature sequence from the deformation state feature sequence array; Perform fine-grained approximate analysis on the first and second deformation state features of the first deformation state feature sequence using the inner product mapping formula to determine a first fine-grained sub-feature similarity set; Normalize the first fine-grained sub-feature similarity set and construct a first-stage deformation trend iteration matrix based on the processing result; Perform deformation state trend iteration on the second deformation state feature based on the first-stage deformation trend iteration matrix to obtain a first-stage deformation state trend iteration feature; Based on the first-stage deformation state trend iteration feature, perform deformation state trend iteration on the third deformation state feature in the first deformation state feature sequence, and so on, to obtain a first deformation state trend iteration feature, and add the first deformation state trend iteration feature to the deformation state trend iteration feature array; Perform deformation state trend prediction based on the deformation state trend iteration feature array to determine the deformation state trend prediction factor array; Based on the deformation state trend prediction factor array, perform trend consistency division on the monitoring device array to obtain M divided monitoring device sub-arrays and M deformation state trend consistency prediction factors; Determine M first monitoring bandwidths and M second monitoring bandwidths based on the M deformation state trend consistency prediction factors, and construct M monitoring double channels based on the M first monitoring bandwidths and the M second monitoring bandwidths, wherein the first monitoring bandwidth refers to the maximum value of the abnormal interval duration of the monitoring point in all abnormal events, reflecting the required inspection frequency of the monitoring device in the slowest change case, and the second monitoring bandwidth refers to the minimum value of the abnormal interval duration of the monitoring point in all abnormal events, representing the need for high-frequency monitoring of the monitoring device within a short time interval; Use the M monitoring double channels to perform asynchronous monitoring management on the M divided monitoring device sub-arrays.

2. The integrated building deformation state monitoring management method according to claim 1, wherein Based on the deformation state trend prediction factor array, perform trend consistency division on the monitoring device array to obtain M divided monitoring device sub-arrays and M deformation state trend consistency prediction factors, comprising: Randomly extract M deformation state trend prediction factors from the deformation state trend prediction factor array to obtain M deformation state trend prediction factor starting points; Based on the M deformation state trend prediction factor starting points, perform trend consistency division in the deformation state trend prediction factor array to obtain M divided deformation state trend consistency prediction factor sets; Map divide the monitoring device array based on the M divided deformation state trend consistency prediction factor sets to obtain M divided monitoring device sub-arrays; Perform mean shift analysis on the M sets of partitioned deformation state trend consistency predictors to obtain the M deformation state trend consistency predictors.

3. The integrated building deformation state monitoring management method according to claim 2, wherein Based on the M deformation state trend predictor starting points, perform trend consistency partitioning in the deformation state trend predictor array to obtain M sets of partitioned deformation state trend consistency predictors, including: Perform dispersion identification on the M deformation state trend predictor starting points. If the dispersion identification result meets the preset requirements, construct M starting point neighborhoods according to a preset consistency bandwidth. And according to the preset consistency bandwidth, perform edge diffusion on the M starting point neighborhoods to obtain M diffused starting point neighborhoods. Determine whether the neighborhood density of the M diffused starting point neighborhoods is greater than or equal to the neighborhood density of the M starting point neighborhoods. If so, continue to diffuse the M diffused starting point neighborhoods according to the preset consistency bandwidth until the preset diffusion times are met. Take the M diffused starting point neighborhoods obtained by the last diffusion as the M initial partitioned deformation state trend consistency predictor sets. Add the deformation state predictors in the deformation state trend predictor array that are not partitioned into the M initial partitioned deformation state trend consistency predictor sets into the initial partitioned deformation state trend consistency predictor set corresponding to the maximum similarity value in the M deformation state trend predictor starting points to obtain the M sets of partitioned deformation state trend consistency predictors.

4. The integrated building deformation state monitoring management method according to claim 3, wherein The preset requirement is that the similarity between any two deformation state trend predictor starting points in the M deformation state trend predictor starting points is less than or equal to a preset similarity threshold.

5. The integrated building deformation state monitoring management method according to claim 1, wherein According to the M deformation state trend consistency predictors, determine M first monitoring bandwidths and M second monitoring bandwidths, and construct M monitoring double channels based on the M first monitoring bandwidths and the M second monitoring bandwidths, including: Match the deformation state abnormal logs based on the M deformation state trend consistency predictors to obtain M sets of matched deformation state abnormal logs. Take the abnormal interval duration as an index to extract the M sets of matched deformation state abnormal logs to obtain M sets of abnormal interval durations. Determine the M first monitoring bandwidths and M second monitoring bandwidths according to the M sets of abnormal interval durations.

6. The integrated building deformation state monitoring management method according to claim 5, wherein Take the maximum value in each of the M sets of abnormal interval durations as a first monitoring bandwidth to obtain M first monitoring bandwidths. Take the minimum value in each of the M sets of abnormal interval durations as a second monitoring bandwidth to obtain M second monitoring bandwidths.

7. An integrated building deformation state monitoring management system, characterized by, The integrated building deformation state monitoring management system for implementing the steps of the integrated building deformation state monitoring management method of any one of claims 1-6, comprising: A monitoring array determination module for determining a monitoring device array based on a target integrated building's base plan and foundation pit support design drawing. A trend prediction module for traversing to obtain a monitoring data sequence array of the monitoring device array within a preset monitoring window, performing deformation state trend prediction based on the monitoring data sequence array, and determining a deformation state trend predictor array. A monitoring double channel construction module for determining M first monitoring bandwidths and M second monitoring bandwidths based on the deformation state trend predictor array, and constructing M monitoring double channels based on the M first monitoring bandwidths and the M second monitoring bandwidths. a consistency division module configured to perform trend consistency division on the array of monitoring devices based on the array of deformation state trend prediction factors, to obtain M divided monitoring device sub-arrays and M deformation state trend consistency prediction factors; a dual-channel construction module configured to determine M first monitoring bandwidths and M second monitoring bandwidths according to the M deformation state trend consistency prediction factors, and to construct M monitoring dual-channels based on the M first monitoring bandwidths and the M second monitoring bandwidths, wherein the first monitoring bandwidth refers to the maximum value of abnormal interval duration of the monitoring point in all abnormal events, reflecting the required inspection frequency of the monitoring device in the slowest change, and the second monitoring bandwidth refers to the minimum value of abnormal interval duration of the monitoring point in all abnormal events, representing the high-frequency monitoring required by the monitoring device in a short time interval; an asynchronous monitoring module configured to perform asynchronous monitoring management on the M divided monitoring device sub-arrays by using the M monitoring dual-channels.

Images