A method and system for formulating a spatio-temporal combined early warning index for the deformation of a concrete dam

Abstract

The present invention belongs to the technical field of concrete deformation early warning, and particularly relates to a method and system for formulating a spatio-temporal combined early warning index for concrete dam deformation. The method includes: performing cluster analysis on the deformation measurement points of the concrete dam; calculating and obtaining the extreme values of the structural deformation response according to adverse working conditions, and expanding the deformation extreme value samples; obtaining the multi-measurement point deformation joint probability distribution function with strong correlation according to the structural deformation response extreme values, and formulating the deformation early warning index; constructing the spatio-temporal combined early warning index for concrete dam deformation according to the failure probability. The system includes an analysis module, an extreme value acquisition module, a formulation module, and a construction module. The spatio-temporal combined deformation early warning index established by the present invention is relatively close to the deformation early warning index formulated by the structural analysis method, which can effectively improve the accuracy of concrete dam deformation early warning, reduce the false alarm rate of abnormal deformation early warning, and at the same time improve the robustness and effectiveness of concrete dam abnormal deformation early warning.

Description

Technical Field

[0001] The present invention belongs to the technical field of concrete deformation early warning, and particularly relates to a method and system for formulating a spatio-temporal combined early warning index for the deformation of a concrete dam. Background Technique

[0002] In order to accurately grasp the spatio-temporal evolution law of the deformation of a concrete dam and avoid the lack of monitoring data caused by the failure or malfunction of individual measuring points during the long-term operation of the dam, multiple deformation measuring points are usually buried at some important structural parts or sections of the dam to accurately obtain the monitoring data reflecting the change of the structural state. After obtaining the deformation monitoring data of the corresponding measuring points of the concrete dam, an early warning judgment interval can be established based on the confidence interval estimation method to conduct early warning analysis on a single measuring point or multiple measuring points. However, the early warning result of a single measuring point is easily interfered by various internal and external unknown factors, and the accuracy is relatively low. Especially when conducting separate early warning on multiple measuring points, the phenomenon of inconsistent early warning results may occur. Therefore, this paper studies the use of multi-measuring-point deformation monitoring data for spatio-temporal combined early warning, and establishes a spatio-temporal deformation combined early warning index for a concrete dam in a multi-dimensional probability space. Summary of the Invention

[0003] In view of the above problems, the present invention provides a method for formulating a spatio-temporal combined early warning index for the deformation of a concrete dam, and the method includes:

[0004] Performing cluster analysis on the deformation measuring points of the concrete dam;

[0005] Calculating and obtaining the extreme value of the structural deformation response according to the adverse working conditions, and expanding the sample of the deformation extreme value;

[0006] Obtaining a multi-measuring-point deformation joint probability distribution function with strong correlation according to the extreme value of the structural deformation response, and formulating a deformation early warning index;

[0007] Constructing a spatio-temporal combined early warning index for the deformation of the concrete dam according to the failure probability.

[0008] Preferably, the spatio-temporal cluster analysis of the deformation measuring points of the concrete dam adopts a spatio-temporal cluster analysis method based on a self-organizing mapping neural network, and includes the following steps:

[0009] [[ID=**31]]Converting the non-linear correlation relationship between the deformations of the deformation measuring points into a geometric relationship;

[0010] Classifying the deformation measuring points according to the correlation of the deformation measuring points to form a deformation measuring point set.

[0011] Preferably, the method for obtaining the extreme value of the deformation response includes calculating and obtaining by using a statistical prediction model and calculating and obtaining by using a hybrid prediction model.

[0012] Preferably, the steps of calculating and obtaining the extreme value of the deformation response by using the hybrid prediction model include: taking the upstream reservoir water level and the extreme temperature drop of the concrete dam as control conditions, establishing a hybrid prediction model between the monitored effect quantity and the load combination, so as to obtain the extreme value of the deformation of the concrete dam response.

[0013] Preferably, the monitored effect quantity includes the deformation and stress of the concrete dam.

[0014] Preferably, the steps of obtaining the joint probability distribution function of multiple measurement points with strong correlation include the following steps:

[0015] Determine the marginal distribution of the deformation measurement points;

[0016] Construct the joint distribution of the deformation measurement points;

[0017] Determine the joint distribution function and joint probability density function of the deformation measurement points according to the marginal distribution and the Copula function.

[0018] Preferably, the methods for determining the marginal distribution of the deformation measurement points include the hypothesis testing method and the kernel density estimation method.

[0019] Preferably, the method for constructing the joint distribution of the deformation measurement points includes constructing by using the t-Copula function.

[0020] Preferably, the method for formulating the deformation warning index is the typical small probability method.

[0021] Preferably, the steps of constructing the spatio-temporal joint warning index for the deformation of the concrete dam according to the failure probability include the following steps:

[0022] Obtain the possibility of deformation occurrence and the degree of safety risk of the deformation measurement points according to the deformation warning index;

[0023] Classify the warning index of the deformation measurement points according to the possibility of deformation occurrence and the degree of safety risk of the deformation measurement points;

[0024] Conduct hierarchical warning on the concrete dam according to multiple levels of the warning index.

[0025] The present invention also proposes a system for formulating the spatio-temporal joint warning index for the deformation of a concrete dam, and the system includes an analysis module, an extreme value acquisition module, a formulation module, and a construction module;

[0026] The analysis module is used for performing cluster analysis on the deformation measurement points of the concrete dam;

[0027] The extreme value acquisition module is used for calculating and obtaining the extreme value of the structural deformation response according to the adverse working conditions, and expanding the extreme value sample of the deformation;

[0028] The proposed module is used to obtain the joint probability distribution function of multi-measurement point deformations with strong correlation based on the extreme value of the structural deformation response, and to propose a deformation warning index.

[0029] The construction module is used to construct a spatio-temporal joint warning index for the deformation of the concrete dam according to the failure probability.

[0030] Preferably, the analysis module is used to perform cluster analysis on the deformation measurement points of the concrete dam, including: the analysis module is used to perform cluster analysis on the deformation measurement points of the concrete dam based on the spatio-temporal cluster analysis method of the self-organizing mapping neural network.

[0031] Preferably, the analysis module is used to perform cluster analysis on the deformation measurement points of the concrete dam based on the spatio-temporal cluster analysis method of the self-organizing mapping neural network, including:

[0032] The analysis module is used to convert the non-linear correlation relationship between the deformations of the deformation measurement points into a geometric relationship;

[0033] Classify the deformation measurement points according to the correlation of the deformation measurement points to form a set of deformation measurement points.

[0034] Preferably, the extreme value acquisition module is used to calculate and obtain the extreme value of the structural deformation response, including: the extreme value acquisition module is used to calculate and obtain the extreme value of the structural deformation response through a statistical prediction model or the extreme value acquisition module is used to calculate and obtain the extreme value of the structural deformation response through a hybrid prediction model.

[0035] Preferably, the extreme value acquisition module is used to calculate and obtain the extreme value of the structural deformation response through a hybrid prediction model, including: the extreme value acquisition module takes the upstream reservoir water level and the extreme temperature drop of the concrete dam as control conditions, establishes a hybrid prediction model between the monitoring effect quantity and the load combination, so as to obtain the extreme value of the deformation of the concrete dam response;

[0036] The monitoring effect quantity includes the deformation quantity and stress of the concrete dam.

[0037] Preferably, the proposed module is used to obtain the joint probability distribution function of multi-measurement points with strong correlation, including:

[0038] The proposed module is used to determine the marginal distribution of the deformation measurement points;

[0039] Construct the joint distribution of the deformation measurement points;

[0040] Determine the joint distribution function and joint probability density function of the deformation measurement points according to the marginal distribution and the Copula function.

[0041] Preferably, the construction module is used to construct a spatio-temporal joint warning index for the deformation of the concrete dam, including:

[0042] The building block is used to obtain the deformation occurrence probability and safety risk degree of the deformation measuring points according to the deformation warning index;

[0043] Formulate multi-level warning indexes for the deformation measuring points according to the deformation occurrence probability and safety risk degree of the deformation measuring points;

[0044] Perform hierarchical warning on the concrete dam according to the multi-level warning indexes.

[0045] The present invention has the following beneficial effects:

[0046] (1) By using the self-organizing mapping neural network SOM clustering method to perform spatio-temporal clustering on the deformation measuring points of the concrete dam, and calculating the extreme value samples of the deformation of the concrete dam under relatively unfavorable load combinations through the hybrid prediction model, the representativeness of the warning index formulation samples is improved, so that they include the extreme values of the deformation under relatively unfavorable load combinations. With the help of the Copula theory, the joint probability distribution function of the deformation measuring point groups with strong correlation is solved, and the spatio-temporal deformation joint warning index in the multi-dimensional probability space is established, which can realize the spatio-temporal joint warning of the deformation of multiple measuring points of the concrete dam;

[0047] (2) The spatio-temporal joint deformation warning index established by the present invention is relatively close to the deformation warning index formulated by the structural analysis method, which can effectively improve the accuracy of the deformation warning of the concrete dam, reduce the false alarm rate of the deformation anomaly warning, and at the same time improve the robustness and effectiveness of the deformation anomaly warning of the concrete dam;

[0048] (3) The present invention not only relates to the extreme values of the deformation under the unfavorable load combinations encountered during the operation period of the concrete dam, but also includes the extreme values of the deformation under the relatively unfavorable load combinations that may occur in the future. The deformation warning index formulated accordingly has good adaptability to the warning and evaluation of the state evolution of the dam during the operation period;

[0049] (4) The present invention effectively overcomes the problems existing in the formulation of the deformation warning index for a single measuring point of the concrete dam, such as poor representativeness of the extreme value samples of the deformation, susceptibility to various unknown factors, poor suitability of the overall distribution function, inconsistent warning results, etc., making the warning results more scientific and reasonable.

[0050] Other features and advantages of the present invention will be described in the subsequent specification, and part of them will become obvious from the specification or be understood by implementing the present invention. The objectives and other advantages of the present invention can be achieved and obtained through the structures pointed out in the specification and the drawings. Description of the Drawings

[0051] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0052] Figure 1 Shows the flow chart of a method for formulating spatio-temporal combined warning indicators for the deformation of a concrete dam in an embodiment of the present invention;

[0053] Figure 2 Shows the spatio-temporal clustering map of deformation measurement points of a concrete arch dam in an embodiment of the present invention;

[0054] Figure 3 Shows the spatio-temporal clustering map of deformation measurement points of a concrete gravity dam in an embodiment of the present invention;

[0055] Figure 4 Shows the schematic diagram of the joint probability density of the deformation of a concrete dam in an embodiment of the present invention;

[0056] Figure 5 Shows the layout plan of the horizontal displacement measurement points of the tension wire on the dam crest in an embodiment of the present invention;

[0057] Figure 6 Shows the process line of the horizontal displacement monitoring data of the EX1 - EX12 measurement points of the tension wire on the dam crest in an embodiment of the present invention;

[0058] Figure 7 Shows the spatio-temporal clustering map of the horizontal displacement measurement points on the dam crest in an embodiment of the present invention;

[0059] Figure 8 Shows the finite element model of the entire dam section of a concrete dam in an embodiment of the present invention;

[0060] Figure 9 Shows the calculation result of the water pressure component of a concrete dam under the design flood level condition in an embodiment of the present invention;

[0061] Figure 10 Shows the cumulative probability density curves of the empirical distribution function and kernel density estimation of the deformation of the EX3 measurement point in an embodiment of the present invention;

[0062] Figure 11 Shows the cumulative probability density curves of the empirical distribution function and kernel density estimation of the deformation of the EX4 measurement point in an embodiment of the present invention;

[0063] Figure 12 Shows the binary frequency distribution histogram of the deformation of the EX3 and EX4 measurement points in an embodiment of the present invention;

[0064] Figure 13The joint probability density diagram of the cumulative distribution of deformation extreme values of the EX3 and EX4 measuring points in an embodiment of the present invention is shown;

[0065] Figure 14 The joint probability distribution diagram of the deformation extreme values of the EX3 and EX4 measuring points in the embodiment of the present invention is shown;

[0066] Figure 15 The joint probability distribution contour map of the deformation extreme values of the EX3 and EX4 measuring points in the embodiment of the present invention is shown;

[0067] Figure 16 A scatter plot of extreme value samples of downstream horizontal displacement of measuring points EX3 and EX4 in an embodiment of the present invention is shown;

[0068] Figure 17 A diagram showing a system for formulating spatiotemporal combined early warning indicators for concrete dam deformation according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0069] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0070] In order to develop a spatiotemporal joint early warning indicator for concrete dam deformation based on SOM-Copula, this paper first performs cluster analysis on the concrete dam deformation measurement points through a spatiotemporal clustering algorithm based on the self-organizing map neural network (SOM). Then, a statistical prediction model or a hybrid prediction model is used to calculate the extreme values of the structural deformation response under multiple groups of less favorable load conditions. This is used to expand the existing deformation extreme value samples and improve the representativeness of the sample space. Finally, the multivariate Copula theory is used to obtain the joint probability distribution function of multiple measurement points with strong correlation, and the typical small probability method is used to formulate the deformation early warning indicator. Finally, a spatiotemporal joint early warning indicator for concrete dam deformation based on SOM-Copula is constructed.

[0071] like Figure 1 As shown, the present invention proposes a method for formulating a spatiotemporal combined early warning indicator for concrete dam deformation, the method comprising:

[0072] (1) Cluster analysis of concrete dam deformation measurement points

[0073] In order to fully combine the deformation monitoring data of multiple measuring points and formulate the joint early warning index for the spatio-temporal deformation of concrete dams in a multi-dimensional probability space, it is first necessary to conduct spatio-temporal clustering analysis on the deformation measuring points of concrete dams. In this paper, a spatio-temporal clustering analysis method based on the self-organizing mapping neural network SOM is used to conduct spatio-temporal clustering analysis on the deformation measuring points of concrete dams, transforming the complex non-linear correlation relationship between the deformations of each measuring point into a simple geometric relationship, and dividing the deformation measuring points with strong correlation into the same category. Figure 2 Figure 2 is the spatio-temporal clustering diagram of the deformation measuring points of a certain concrete arch dam. The deformations of the measuring points in each category in the figure have a high degree of correlation; Figure 3 Figure 3 is the spatio-temporal clustering diagram of the deformation measuring points of a certain concrete gravity dam based on SOM. For a concrete gravity dam, due to the setting of its transverse joints, the deformations of each dam section are relatively independent. The deformations of most measuring points are consistent with the clustering results according to the dam sections, but there are also some measuring points that do not satisfy this rule. Through the spatio-temporal clustering method of the deformation measuring points of a concrete gravity dam based on SOM, the deformation measuring points of the concrete dam can be divided into several categories. The set of deformation measuring points in the same category containing p measuring points can be expressed as:

[0074] Y = {Y1, Y2, …, Y p} (Formula 1)

[0075] Among them, Y represents the set of deformation measuring points in the same category containing p measuring points, and Y p is the pth measuring point.

[0076] (2) Calculate and obtain the extreme values of the structural deformation response according to the adverse working conditions, and expand the samples of the extreme deformation values

[0077] In the deformation data during the operation period of the concrete dam, according to different dam types and actual operation conditions, select the monitoring effect quantity or each load component in its mathematical model when including the adverse load combination (generally select the load combination at the moment when the annual maximum deformation occurs) as the typical monitoring effect quantity. The sample space of the extreme deformation values of the ith deformation measuring point of the concrete dam containing the adverse load combination is:

[0078]

[0079] Among them: Y i represents the sample space of the extreme deformation values of the ith deformation measuring point containing the adverse load combination, represents the extreme deformation value under the mth adverse load combination, generally the annual maximum deformation.

[0080] Concrete dams generally do not experience unfavorable load combinations during their operation. Consequently, measured deformation monitoring data do not include corresponding deformation monitoring values for these unfavorable load combinations. Consequently, deformation early warning indicators developed based on these data are only extreme values under current load conditions and cannot accurately reflect the extreme operational behavior of concrete dams. Only when long-term operational monitoring data actually encounter unfavorable load combinations can the extreme deformation values at each measuring point be accurately estimated, providing a representative data sample for the development of early warning indicators. Based on this, this paper, based on existing long-term deformation monitoring data, uses the reservoir water level and extreme temperature drop upstream of the concrete dam as control conditions. Through structural calculations, a hybrid prediction model is established that links monitoring effect quantities (such as deformation and stress) with load combinations. This model then derives the extreme deformation values of the concrete dam under unfavorable load combinations. This expands the sample space containing measured extreme deformation values for unfavorable load combinations, improves the representativeness of the sample space, and makes the proposed early warning indicators more scientific and reasonable.

[0081] According to the relevant specifications and requirements for concrete dam operation, the upstream water level of the dam is generally strictly controlled and is not allowed to exceed the check flood level. Therefore, several water levels between the historical highest reservoir water level and the check flood level are selected as control conditions. The higher reservoir water level and the extreme temperature drop are calculated as the extreme value samples of dam deformation under the less favorable load combination conditions. Assuming that l water level control conditions between the historical highest reservoir water level and the check flood level are selected, the extreme values of concrete dam deformation under the control conditions are added to the measured deformation sample space containing the unfavorable load combination. Then, the extreme value sample space of deformation under the less favorable load combination that is representative of the i-th measuring point is:

[0082]

[0083] in: Represents the extreme deformation value under the mth unfavorable load combination, generally the maximum deformation value in a year. Represents the extreme deformation value under the lth most unfavorable load combination condition.

[0084] (3) Obtain a joint probability distribution function of deformation at multiple measuring points with strong correlation based on the extreme value of the structural deformation response, and formulate deformation warning indicators

[0085] For a set of deformation measurement points of the same type Y={Y1,Y2,…,Y p}, first determine the marginal distributions of each deformation measurement point according to the hypothesis testing method or kernel density estimation method. Considering that the deformation warning index proposed for concrete dams pays more attention to the tail characteristics of its distribution, a multivariate normal Copula function or a multivariate t-Copula function can be selected to construct the joint distribution of the deformation measurement points. Taking the set of highly correlated deformation measurement points of the same type containing two measurement points Y1 and Y2 as an example, determine their respective marginal distribution functions as F1(y1) and F2(y2), and select the bivariate normal Copula function as the connection function. Then their expressions are respectively:

[0086]

[0087]

[0088] where: C Ga (F1(y1), F2(y2)) represents the bivariate joint distribution function solved by the bivariate normal Copula function, s and t are both integral variables, and ρ represents the linear correlation coefficient between the measurement points Y1 and Y2; Φ -1 represents the inverse function of the standard normal distribution function, and c(F1(y1), F2(y2); ρ) is the bivariate joint probability density function solved by the bivariate normal Copula function.

[0089] The marginal distribution functions F1(y1) and F2(y2) of the deformation samples of the measurement points Y1 and Y2 do not contain unknown parameters. Estimate the parameter ρ in the Copula function by the semi-parametric estimation method. According to the Copula theory, the expressions of the joint distribution function and the joint probability density function of the measurement points Y1 and Y2 are respectively:

[0090]

[0091]

[0092] where, H(y1, y2) is the joint distribution function of the deformations of the measurement points Y1 and Y2, and c(y1, y2) is the joint probability density function of the deformations of the measurement points Y1 and Y2, and ρ represents the linear correlation coefficient between the measurement points Y1 and Y2.

[0093] For the set of highly correlated deformation measurement points of the same type containing multiple measurement points, the process of deriving its multivariate joint distribution function and multivariate joint probability density function with the help of the multivariate Copula function is similar to the process of deriving the bivariate joint distribution function and the bivariate joint probability density function, and will not be elaborated here.

[0094] (4) Construct the spatio-temporal joint warning index for the deformation of the concrete dam according to the failure probability

[0095] The deformation behavior of concrete dams is affected by various random factors such as material parameters and external loads. The deformation warning indicators formulated based on the measured deformation data of concrete dams also have a certain degree of randomness. The extreme values of the deformations at measuring points Y1 and Y2 follow the joint distribution function and joint probability density function in Formulas 6 and 7. For a given significance level α, there is a corresponding upper α quantile in the probability density function. Then, the deformation value corresponding to the upper α quantile of measuring point Y1 is y 1，1-α , and the deformation value corresponding to the upper α quantile of measuring point Y2 is y 2，1-a . The planar region formed by these four points is the normal domain of the deformations at measuring points Y1 and Y2 under the given significance level α, denoted as D 1-α . When the deformation values of measuring points Y1 and Y2 at a certain moment are (y1, y2), then the probability that (y′1, y'2) ∈ D 1-α is:

[0096]

[0097] When formulating the deformation warning indicators for two measuring points using the typical small probability method, its normal domain D 1-α is a finite plane; when formulating the deformation warning indicators for three or more measuring points, its normal domain D 1-α is a finite space of high dimension.

[0098] According to the likelihood of deformation occurrence and the degree of safety risk, the deformation warning indicators of concrete dams are divided into first-level warning indicators and second-level warning indicators based on the significance level. The classified deformations of concrete dams are: normal samples, warning samples, and abnormal samples. Figure 4 is a schematic diagram of the joint probability density of concrete dam deformation. Assume that the value range of the deformation value of a certain measuring point is D. When the significance level is α = 5% and the confidence level is 95.00%, the deformation values of measuring points Y1 and Y2 fall within their corresponding normal domain D 95.00 , then it is considered that the deformation behavior of measuring points Y1 and Y2 of the concrete dam is in a normal state; when the significance level is α = 1% and the confidence level is 99.00%, the deformation values of measuring points Y1 and Y2 exceed D 95.00 normal domain, but still within their corresponding basically normal domain D 99.00 , then it is considered that the deformation behavior of measuring points Y1 and Y2 of the concrete dam is in a warning state. When the deformation values of measuring points Y1 and Y2 exceed the basically normal domain D 99.00 , and are within D - D 99.00 , then it is considered that the deformation behavior of measuring points Y1 and Y2 of the concrete dam is in an abnormal state. The spatio-temporal joint multi-level warning indicators for concrete dam deformation are shown in Table 1.

[0099] Table 1 Spatio-temporal joint multi-level warning indicators for concrete dam deformation

[0100]

[0101] Based on measured deformation data from concrete dams during operation, a multidimensional probability density function reflecting their overall variation characteristics was constructed to develop a spatiotemporal joint early warning indicator for concrete dam deformation. This paper analyzes deformation monitoring data obtained from the tensioning line monitoring system of the concrete dam at the Mianhuatan Hydropower Station and constructs a spatiotemporal joint early warning indicator for deformation of a typical dam section. The results are compared with those from traditional early warning indicator development methods to verify the feasibility and reliability of the proposed SOM-Copula-based method for developing a spatiotemporal joint early warning indicator for concrete dam deformation.

[0102] 1. Project Overview

[0103] The Mianhuatan Hydropower Station is located in Yongding County, Fujian Province. The dam site is at Fuzhiting in the middle of the Mianhuatan Canyon section of the main stream of the Tingjiang River, about 21km away from Yongding County. The project is mainly for power generation, and also has comprehensive benefits such as flood control, shipping, and aquaculture. The Mianhuatan Hydropower Station hub project is a Class I hub project, mainly consisting of a roller-compacted concrete gravity dam, a Huyangli auxiliary dam, an open spillway on the dam top, a bottom outlet for water discharge, a left bank water transfer structure, and an underground power plant. The maximum dam height of the roller-compacted concrete gravity dam is 113.00m, and the dam top elevation is 179.00m. The normal water level of the reservoir is 173.00m, and the regulating storage capacity is 1.122 billion m 3 The verified flood level is 177.80m, and the corresponding total reservoir capacity is 2.035 billion m 3 .

[0104] To ensure the safe operation of the hydropower station dam and underground powerhouse, the hub project has set up relatively comprehensive monitoring projects such as deformation, seepage, and environmental quantities. Deformation monitoring includes vertical lines, inverted vertical lines, and tension lines. The horizontal displacement of the dam top is monitored using the tension line method. Figure 5 This is the arrangement diagram of the horizontal displacement measuring points of the dam top tension line. Figure 6 This is a process diagram of the horizontal displacement monitoring data of measuring points EX1 to EX12 on the dam top tension line from January 1, 2003 to December 31, 2008.

[0105] 2. Spatiotemporal clustering of concrete dam deformation measurement points

[0106] The SOM spatiotemporal clustering analysis method was used to cluster the deformation measurement points of the Mianhuatan concrete gravity dam. The results of some measurement points were consistent with the results of clustering the deformation measurement points according to dam sections. Figure 7 This is the spatiotemporal clustering diagram of the horizontal displacement measurement points of the Mianhuatan concrete gravity dam.

[0107] Taking the 2# typical dam section and the 5# typical dam section as examples, this paper constructs the spatiotemporal joint early warning indicators for the corresponding dam sections based on the horizontal displacement monitoring data of the dam crest tensioning line EX3 and EX4 measuring points corresponding to the 2# typical dam section and the dam crest tensioning line EX8 and EX9 measuring points corresponding to the 5# typical dam section.

[0108] 3. Expansion of samples of extreme deformation values of concrete dams

[0109] (1) Sample space under unfavorable load combinations

[0110] For concrete gravity dams, when the upstream reservoir water level is high, the dam body and foundation are generally in an unfavorable state of strength and stability. Therefore, the historical annual maximum deformation values at each measuring point and the extreme deformation values under less favorable load combinations are selected to form a representative sample space of extreme deformation values. Based on the actual operation of the concrete gravity dam at the Mianhuatan Hydropower Station, the upstream water level exhibits a clear annual periodicity, and the horizontal displacement of the dam crest tensioning line is primarily affected by the water level. Therefore, the annual maximum horizontal displacement values at each measuring point on the dam crest tensioning line constitute the sample space of extreme deformation values under unfavorable load combinations. Table 2 shows the statistical annual maximum horizontal displacement values at measuring points on the dam crest tensioning line for typical dam sections.

[0111] Table 2 Statistics of annual maximum horizontal displacements of measuring points on the dam crest tension line of typical dam sections

[0112]

[0113] Analysis of measured deformation monitoring data for the Mianhuatan Hydropower Station's concrete gravity dam reveals that upstream reservoir water levels generally remain high from April and May until the beginning of the following year. Flood peaks typically occur in June and July, when temperatures are also high. However, water pressure and temperature have opposite effects on dam deformation. The combined effects of high water levels and temperatures may result in smaller dam crest displacements during high flood seasons than at other times. Consequently, the established sample space represents deformations under different load combinations and excludes extreme deformations under unfavorable load combinations. Furthermore, since its operation, the Mianhuatan concrete dam has not yet experienced high water level operating conditions, such as the design flood level or the verification flood level. Therefore, the extreme deformation values in the existing monitoring data represent only those under current load conditions and do not include those under less favorable load combinations, such as the design flood level or the verification flood level. Therefore, the existing extreme deformation sample needs to be expanded to improve its representativeness.

[0114] (2) Extreme deformation under less favorable load combinations

[0115] To improve the representativeness of the sample space of the horizontal displacement of the tension wire at the top of the Mianhuatan concrete dam, the extreme values of the water pressure component under relatively unfavorable load combinations are obtained through structural calculations. The extreme values of the temperature component and the aging component are calculated through the results of the statistical prediction model. According to the principle of the most unfavorable load, the extreme values of the deformation under relatively unfavorable load combination conditions are obtained by accumulation.

[0116] 1) Finite element model

[0117] During the process of building the finite element model, the modeling scope of the bedrock and the slope: vertically, 68.00 m below the lowest elevation of the dam body is selected; parallel to the dam axis direction, 50.00 m is extended from the dam heads on both the left and right banks to both banks respectively; parallel to the riverbed direction, 150.00 m above the upstream dam heel and 75.00 m below the downstream dam toe are selected. A finite element model of the bedrock and the slope with a rectangular plane outer boundary is established within the above range.

[0118] The Mianhuatan dam body is divided into 6 dam sections, named 1# - 6# dam sections from left to right in sequence. Among them, the 1#, 2#, 5#, and 6# dam sections are water retaining dam sections, and the 3# and 4# dam sections are overflow dam sections. The finite element model of the entire dam section of the Mianhuatan dam is mainly composed of eight-node hexahedron isoparametric elements and a small number of pentahedron and tetrahedron elements, with a total of 140,554 elements and 152,790 nodes. The finite element model of the entire dam section of the dam can be seen in Figure 8 .

[0119] 2) Calculation parameters

[0120] To obtain the extreme values of the deformation of the Mianhuatan concrete gravity dam under relatively unfavorable load combinations, 9 water levels (173.50 m, 174.00 m, 174.50 m, 175.00 m, 175.50 m, 176.00 m, 176.50 m, 177.00 m, and 177.50 m) between the historical highest reservoir water level (September 11, 2001, 173.00 m) and the check flood level (177.80 m) are selected as control conditions, and the extreme values of the deformation under each relatively unfavorable load combination are calculated respectively using the hybrid prediction model. At the same time, to verify the rationality of the spatio-temporal joint warning index of the concrete dam based on SOM-Copula, its proposed results are compared and analyzed with the warning indexes proposed using the typical small probability method and the structural analysis method. The material parameters involved in the structural calculation process and their values are shown in Table 3, where the concrete density and the anti-sliding stability parameters of the dam foundation surface take the design values, the elastic moduli of the dam body and the dam foundation take the inversion values obtained by the conventional inversion method, and the safety factor of the anti-sliding stability critical state of the dam foundation surface takes K f = 3.

[0121] Table 3 Calculation parameters of the dam body and the dam foundation materials

[0122]

[0123] 3) Finite element calculation results of typical dam sections

[0124] The finite element model of the entire concrete dam of Mianhuatan Hydropower Station is used to calculate the dam deformation. Since the obtained deformation is the calculated value relative to the initial zero displacement state, in order to obtain its relative deformation value relative to the water pressure component at the initial water level, the water pressure component on the reference date also needs to be considered. Figure 9 Table 4 shows the relative water pressure components of the horizontal displacement of the tension wire measuring points of the 2# and 5# typical dam sections under some water level control conditions, which are the calculation results of the water pressure components of the concrete dam of Mianhuatan Hydropower Station under the checking water level condition. According to the established mixed prediction model of the horizontal displacement of the tension wire on the dam crest, the temperature component and the aging component are separated. Considering the influence of the temperature component of the horizontal displacement on the reference date, it is determined that the extreme values of the relative temperature components of the EX3 and EX4 measuring points of the 2# typical dam section under the extreme temperature drop control condition are 0.12 mm and 0.10 mm respectively, and the extreme values of the relative temperature components of the EX8 and EX9 measuring points of the 5# typical dam section under the extreme temperature drop control condition are 0.22 mm and 0.25 mm respectively; the horizontal displacements of the EX3, EX4, EX8 and EX9 measuring points of the 2# and 5# typical dam sections are less affected by aging, and the aging components of each measuring point have tended to be stable. Therefore, the cumulative aging component from the initial monitoring date to the last day of the monitoring data sequence is selected as the extreme value of the aging component. The extreme values of the aging components of the horizontal displacements of the EX3 and EX4 measuring points of the 2# typical dam section are calculated to be -1.26 mm and -1.27 mm respectively; the extreme values of the aging components of the horizontal displacements of the EX8 and EX9 measuring points of the 5# typical dam section are 0.35 mm and -0.34 mm respectively. According to the principle of the most unfavorable load, by adding up the relative water pressure components, the extreme values of the temperature components, the extreme values of the aging components and the deformation monitoring reference value (January 1, 2003) of the EX3, EX4, EX8 and EX9 measuring points of the above 2# and 5# typical dam sections, the extreme values of the deformation of each measuring point under the relatively unfavorable load control conditions can be obtained, as shown in Table 5 in detail.

[0125] Table 4 Calculation table of relative water pressure components of the tension wire measuring points of the 2# and 5# typical dam sections under some water level control conditions

[0126]

[0127]

[0128] Table 5 Table of each component and extreme value of horizontal displacement at the tension wire measuring points of the 2# and 5# typical dam sections under some control conditions

[0129]

[0130] 4) Extreme value samples of deformation

[0131] The horizontal displacement samples under unfavorable load combinations and the extreme value samples of horizontal displacement under relatively unfavorable load combinations are combined to jointly constitute the sample space for formulating the warning index of the horizontal displacement of the measuring points of the tension wire on the dam crest. The sample space for formulating the warning index of the horizontal displacement of the measuring points of the tension wire on the dam crest of the typical dam section of the Mianhuatan concrete gravity dam is shown in Table 6.

[0132] Table 6 Sample space for formulating the warning index of the horizontal displacement of the measuring points of the tension wire on the dam crest of the typical dam section

[0133]

[0134]

[0135] 4. Joint distribution probability density function of the deformations of multiple measuring points of concrete dams

[0136] The distribution types of the extreme value samples of the deformation effect quantities of concrete dams under relatively unfavorable load combinations are diverse. Generally, the hypothesis testing method or the kernel density estimation method can be used to fit their distribution types. Since the distribution types of the sample data of the extreme values of the horizontal displacements of the measuring points EX3, EX4, EX8, and EX9 of the tension wire on the dam crest under relatively unfavorable load combinations are unknown, therefore, in this paper, the kernel density estimation method is used to fit the distribution types of the extreme value samples of the deformations of each measuring point. Figure 10 is the empirical distribution function of the deformation of measuring point EX3 and the cumulative probability density curve of kernel density estimation. Figure 11 is the empirical distribution function of the deformation of measuring point EX4 and the cumulative probability density curve of kernel density estimation.

[0137] Analysis Figure 10 and Figure 11 show that: The cumulative probability curves of the deformations of measuring points EX3 and EX4 obtained by kernel density estimation fit well with the empirical distribution probability cumulative curves. Therefore, the cumulative probability density curves obtained by kernel density estimation can be used to characterize the distribution of the overall samples.

[0138] The Copula function is the connection function between the marginal distributions of each random variable and the joint distribution of each variable. The selection of the Copula function mainly depends on the distribution types of each random variable. Therefore, the most suitable Copula function can be determined by analyzing the joint distribution characteristics of the random variables. Figure 12 is the binary frequency distribution histogram of the deformations of measuring points EX3 and EX4.

[0139] When formulating the spatio-temporal joint warning index using the deformation monitoring data of concrete dams, the tail characteristics of the distribution of deformation effect quantities are extremely important. Therefore, in this paper, the normal Copula function with a thicker tail and better able to reflect the tail correlation characteristics between variables is selected to construct the joint distribution functions of the horizontal displacements of the measuring points of the tension wires in the typical dam sections 2# and 5# respectively. Based on the sample space of the extreme deformation values of the measuring points EX3 and EX4 of the tension wire on the top of the Mianhuatan concrete dam, the estimated value of the linear correlation parameter ρ in the bivariate normal Copula function obtained through parameter estimation is:

[0140]

[0141] Among them, is the estimated value of the linear correlation parameter ρ.

[0142] Substituting into the bivariate normal Copula function, the estimated bivariate normal Copula function can be obtained as:

[0143]

[0144]

[0145] The obtained bivariate normal Copula function depicts the relationship between the extreme deformation distribution of the measuring point EX3, the extreme deformation distribution of the measuring point EX4, and the joint distribution of the extreme deformations of the measuring points EX3 and EX4. In order to intuitively reflect the correlation relationship among the three, Figure 13 the joint probability density diagram of the cumulative distribution of the extreme deformations of the measuring points EX3 and EX4 is drawn, Figure 14 the joint probability distribution diagram of the extreme deformations of the measuring points EX3 and EX4 is drawn, Figure 15 the joint probability distribution contour diagram of the extreme deformations of the measuring points EX3 and EX4 is drawn.

[0146] Figure 13 , Figure 14 and Figure 15 are to construct the joint distribution of the extreme deformations of the measuring points EX3 and EX4 of the tension wire on the top of the Mianhuatan concrete dam using the Copula function. Analyzing Figure 13 , Figure 14 and Figure 15 shows that: there is a strong correlation between the deformations of the measuring points EX3 and EX4. Especially, the tail deformations of the two measuring points show obvious overall synchronism, and the annual maximum values of the downstream deformations are generated simultaneously under the action of the external environment. The measured maximum deformation values of the measuring points EX3 and EX4 are 2.83 mm and 3.18 mm respectively, which also indicates that there is a large safety margin for the deformation at the top of the dam.

[0147] 5. Formulation of the spatio-temporal joint warning index for the deformation of multiple measuring points of concrete dams

[0148] After obtaining the joint distribution function of the extreme values of the deformations at measuring points EX3 and EX4 through the bivariate Copula theory, according to the small probability theory, the deformation critical values corresponding to the distribution function at significance levels of 5% and 1% (corresponding assurance rates of 95.00% and 99.00%) are taken as the first-level warning index and the second-level warning index respectively. Since the value range of the bivariate distribution function is a surface area, the spatio-temporal joint warning index of the concrete dam crest deformation constructed in this paper is the critical value of the surface area. The first-level warning index is defined as the critical value of the deformation value range (i.e., D 99.00 -D 95.00 ) when the assurance rate is 95.00% - 99.00%; the second-level warning index is defined as the critical value of the deformation value range (i.e., D - D 99.00 ) when the assurance rate is 99.00% - 100.00%. When the deformations (y′1, y'2) of measuring points EX3 and EX4 ∈ D 95.00 , the deformation states of measuring points EX3 and EX4 at the crest of the Mianhuatan Concrete Dam are normal; when the deformations (y′1, y'2) of measuring points EX3 and EX4 ∈ D 99.00 -D 95.00 , a first-level warning is given to the deformation states of measuring points EX3 and EX4 at the crest of the Mianhuatan Concrete Dam, the warning situation is analyzed, the warning sources are explored, and potential safety hazards are eliminated; when the deformations (y′1, y'2) of measuring points EX3 and EX4 ∈ D - D 99.00 , when a second-level warning is given to the deformation states of measuring points EX3 and EX4 at the crest of the Mianhuatan Concrete Dam, the warning sources are immediately investigated, and control decisions are made according to the investigation results.

[0149] To verify the effectiveness of the spatio-temporal joint warning index of the concrete dam deformation based on SOM-Copula, the measured horizontal displacement of the tension wire at the crest of the Mianhuatan Concrete Dam will be analyzed for warning according to the calculated spatio-temporal joint multi-level warning index, and its warning results will be compared and analyzed with the warning results of the single-measurement-point warning index established based on the typical small probability method. Figure 16 The green sample data in

[0150] Table 7 Warning indexes at different significance levels

[0151]

[0152] Table 8 Comparison table of warning results

[0153]

[0154] AnalysisFigure 16 As can be seen from Table 7 and Table 8: 1) When the deformation value of the EX3 measurement point exceeds the warning index determined by the typical small probability method due to a large outlier error, but the deformation value of the EX4 measurement point does not exceed (or vice versa), the typical small probability method will produce false alarms. However, the space-time joint warning index determination method will combine the data distribution types and their correlations of multiple measurement points and will not give warnings to such measurement points. This also proves that the space-time joint warning index for the deformation of concrete dams based on SOM-Copula has stronger robustness and higher accuracy; 2) For the warning index determined by the typical small probability method, at the significance level α = 5%, the warning results include the normal deformation values under normal operating conditions, resulting in a relatively large anomaly and a high false alarm rate for the normal data samples in the monitoring data. At the significance level α = 1%, although the typical small probability method will not produce false alarm warnings for the existing measured deformation monitoring data, it will produce false alarm warnings for the normal deformation values under the design conditions. This is mainly because the deformation warning index determined by the typical small probability method does not include the deformation extreme value samples under relatively unfavorable load conditions, making the determined warning index too small and greatly reducing the accuracy and reliability of the warning results; 3) The space-time joint warning index for the deformation of concrete dams based on SOM-Copula is significantly larger than the deformation warning index determined by the typical small probability method and is relatively close to the deformation warning index determined by the structural analysis method, which proves the rationality of the method for determining the space-time joint warning index for the deformation of concrete dams based on SOM-Copula; 4) Based on the measured deformation monitoring data, the measured horizontal displacement of the tension wire at the top of the Mianhuatan concrete dam downstream is basically maintained between 0 and 3 mm, all within the 95.00% confidence interval. The determination result based on the determined warning index is consistent with the actual state of the dam.

[0155] As Figure 17 shown, the present invention also proposes a system for determining a space-time joint warning index for the deformation of a concrete dam. The system includes an analysis module, an extreme value acquisition module, a determination module, and a construction module;

[0156] The analysis module is used for performing cluster analysis on the deformation measurement points of the concrete dam; the analysis module is used for performing cluster analysis on the deformation measurement points of the concrete dam, including: the analysis module is used for performing cluster analysis on the deformation measurement points of the concrete dam based on the SOM-based space-time cluster analysis method; the analysis module is used for performing cluster analysis on the deformation measurement points of the concrete dam based on the SOM-based space-time cluster analysis method, including: the analysis module is used for converting the non-linear correlation relationship between the deformations of the deformation measurement points into a geometric relationship; classifying the deformation measurement points according to the correlation of the deformation measurement points to form a deformation measurement point set;

[0157] The extreme value acquisition module is used to calculate and obtain the extreme values of the structural deformation response according to adverse working conditions, and expand the extreme value samples of the deformation; the extreme value acquisition module is used to calculate and obtain the extreme values of the structural deformation response, including: the extreme value acquisition module is used to calculate and obtain the extreme values of the structural deformation response through a statistical prediction model or the extreme value acquisition module is used to calculate and obtain the extreme values of the structural deformation response through a hybrid prediction model; the extreme value acquisition module is used to calculate and obtain the extreme values of the structural deformation response through the hybrid prediction model, including: the extreme value acquisition module takes the upstream reservoir water level and the extreme temperature drop of the concrete dam as control conditions, establishes a hybrid prediction model between the monitored effect quantity and the load combination, so as to obtain the extreme values of the deformation of the concrete dam response; the monitored effect quantity includes the deformation and stress of the concrete dam.

[0158] The formulation module is used to obtain the multi-point deformation joint probability distribution function with strong correlation according to the extreme values of the structural deformation response, and formulate the deformation warning index; the formulation module is used to obtain the multi-point joint probability distribution function with strong correlation, including: the formulation module is used to determine the marginal distribution of the deformation measurement points; construct the joint distribution of the deformation measurement points; determine the joint distribution function and joint probability density function of the deformation measurement points according to the marginal distribution and the Copula function.

[0159] The construction module is used to construct the spatio-temporal joint warning index of the concrete dam deformation according to the failure probability; the construction module is used to construct the spatio-temporal joint warning index of the concrete dam deformation, including: the construction module is used to obtain the deformation occurrence possibility and safety risk degree of the deformation measurement points according to the deformation warning index; formulate the multi-level warning index of the deformation measurement points according to the deformation occurrence possibility and safety risk degree of the deformation measurement points; conduct hierarchical warning on the concrete dam according to the multi-level warning index.

[0160] In summary, the method for formulating the spatio-temporal joint warning index of the concrete dam deformation based on SOM-Copula can effectively relate the distribution laws of the deformations of the measurement points with strong correlation, fully consider the deformation responses under various adverse loads, and establish the spatio-temporal deformation joint warning index in the multi-dimensional probability space. The spatio-temporal joint warning index of the concrete dam deformation based on SOM-Copula effectively overcomes the problems existing in the formulation of the single-point deformation warning index of the concrete dam, such as poor representativeness of the extreme value samples of the deformation, susceptibility to various factors, poor suitability of the overall distribution function, inconsistent warning results, etc. It not only reduces the false alarm rate of the abnormal deformation warning of the concrete dam, but also improves the effectiveness of the abnormal deformation warning of the concrete dam.

[0161] Those of ordinary skill in the art should understand that: Although the present invention has been described in detail with reference to the foregoing embodiments, it is still possible to modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for formulating a spatio-temporal combined early warning index for the deformation of a concrete dam, characterized in that, The method includes: Performing cluster analysis on the deformation measurement points of the concrete dam; including: performing spatio-temporal cluster analysis on the deformation measurement points of the concrete dam based on the spatio-temporal cluster analysis method of the self-organizing mapping neural network; Calculating and obtaining the extreme values of the structural deformation response according to adverse working conditions, and expanding the extreme value samples of the deformation; Obtaining the joint probability distribution function of the multi-measurement point deformation with strong correlation according to the extreme values of the structural deformation response, and formulating the deformation warning index; wherein, the method for obtaining the joint probability distribution function of the multi-measurement point with strong correlation includes the following steps: determining the marginal distribution of the deformation measurement points; constructing the joint distribution of the deformation measurement points; determining the joint distribution function and joint probability density function of the deformation measurement points according to the marginal distribution and the Copula function; Constructing the spatio-temporal joint warning index of the concrete dam deformation according to the failure probability.

2. A method for formulating the spatio-temporal joint warning index of the concrete dam deformation according to claim 1, characterized in that Performing spatio-temporal cluster analysis on the deformation measurement points of the concrete dam by using the spatio-temporal cluster analysis method based on the self-organizing mapping neural network, including the following steps: Converting the non-linear correlation relationship between the deformations of the deformation measurement points into a geometric relationship; Classifying the deformation measurement points according to the correlation of the deformation measurement points to form a deformation measurement point set.

3. A method for formulating the spatio-temporal joint warning index of the concrete dam deformation according to claim 1, characterized in that The method for obtaining the extreme value of the deformation response includes calculating and obtaining by using a statistical prediction model and calculating and obtaining by using a hybrid prediction model.

4. A method for formulating the spatio-temporal joint warning index of the concrete dam deformation according to claim 3, characterized in that The steps of calculating and obtaining the extreme value of the deformation response by using the hybrid prediction model include: taking the upstream reservoir water level and the extreme temperature drop of the concrete dam as control conditions, establishing a hybrid prediction model between the monitored effect quantity and the load combination, so as to obtain the extreme value of the deformation of the concrete dam response.

5. A method for formulating the spatio-temporal joint warning index of the concrete dam deformation according to claim 4, characterized in that The monitored effect quantity includes the deformation and stress of the concrete dam.

6. A method for formulating the spatio-temporal joint warning index of the concrete dam deformation according to claim 1, characterized in that The method for determining the marginal distribution of the deformation measurement points includes the hypothesis testing method and the kernel density estimation method.

7. A method for formulating the spatio-temporal joint warning index of the concrete dam deformation according to claim 1, characterized in that The method for constructing the joint distribution of the deformation measurement points includes constructing by using the t-Copula function.

8. A method for formulating the spatio-temporal joint warning index of the concrete dam deformation according to claim 1, characterized in that The method for formulating the deformation warning index is the typical small probability method.

9. A method for formulating the spatio-temporal joint warning index of the concrete dam deformation according to claim 1, characterized in that Constructing the spatio-temporal joint warning index of the concrete dam deformation according to the failure probability includes the following steps: Obtaining the deformation occurrence possibility and the safety risk degree of the deformation measurement points according to the deformation warning index; Classify the early warning indicators of the deformation measurement points according to the possibility of deformation occurrence and the degree of safety risk of the deformation measurement points; Conduct hierarchical early warning on the concrete dam according to multiple levels of the early warning indicators.

10. A system for formulating spatio-temporal combined early warning indicators for the deformation of concrete dams, characterized in that, The system includes an analysis module, an extreme value acquisition module, a formulation module, and a construction module; The analysis module is used to conduct cluster analysis on the deformation measurement points of the concrete dam; including: the analysis module is used to conduct cluster analysis on the deformation measurement points of the concrete dam based on the spatio-temporal cluster analysis method of the self-organizing mapping neural network; The extreme value acquisition module is used to calculate and obtain the extreme value of the structural deformation response according to adverse working conditions, and expand the extreme value sample of the deformation; The formulation module is used to obtain the joint probability distribution function of the deformation of multiple measurement points with strong correlation according to the extreme value of the structural deformation response, and formulate the deformation early warning indicators; wherein, the formulation module is used to obtain the joint probability distribution function of multiple measurement points with strong correlation, including: the formulation module is used to determine the marginal distribution of the deformation measurement points; construct the joint distribution of the deformation measurement points; determine the joint distribution function and joint probability density function of the deformation measurement points according to the marginal distribution and the Copula function; The construction module is used to construct the spatio-temporal joint early warning indicator of the concrete dam deformation according to the failure probability.

11. A system for formulating the spatio-temporal joint early warning indicator of the concrete dam deformation according to claim 10, wherein, The analysis module is used to conduct cluster analysis on the deformation measurement points of the concrete dam based on the spatio-temporal cluster analysis method of the self-organizing mapping neural network, including: The analysis module is used to convert the non-linear correlation relationship between the deformations of the deformation measurement points into a geometric relationship; Classify the deformation measurement points according to the correlation of the deformation measurement points to form a deformation measurement point set.

12. A system for formulating the spatio-temporal joint early warning indicator of the concrete dam deformation according to claim 10, wherein, The extreme value acquisition module is used to calculate and obtain the extreme value of the structural deformation response, including: the extreme value acquisition module is used to calculate and obtain the extreme value of the structural deformation response through a statistical prediction model or the extreme value acquisition module is used to calculate and obtain the extreme value of the structural deformation response through a hybrid prediction model.

13. A system for formulating the spatio-temporal joint early warning indicator of the concrete dam deformation according to claim 12, wherein, The extreme value acquisition module is used to calculate and obtain the extreme value of the structural deformation response through a hybrid prediction model, including: The extreme value acquisition module takes the upstream reservoir water level and the extreme temperature drop of the concrete dam as control conditions, establishes a hybrid prediction model between the monitored effect quantity and the load combination, so as to obtain the extreme value of the deformation of the concrete dam response; The monitored effect quantity includes the deformation quantity and stress of the concrete dam.

14. A system for formulating the spatio-temporal joint early warning indicator of the concrete dam deformation according to claim 10, wherein, The construction module is used to construct the spatio-temporal joint early warning indicator of the concrete dam deformation, including: The construction module is used to obtain the possibility of deformation occurrence and the degree of safety risk of the deformation measurement points according to the deformation early warning indicators; Formulate multi-level early warning indicators of the deformation measurement points according to the possibility of deformation occurrence and the degree of safety risk of the deformation measurement points; Conduct hierarchical early warning on the concrete dam according to multiple levels of the early warning indicators.

Images