A method for formulating evaluation indicators based on the hierarchical framework of online evaluation of concrete dam operating performance

Abstract

The present application proposes a hierarchical architecture for online evaluation of the operating status of a concrete dam and a method for formulating evaluation indicators, which relate to the technical field of concrete dam structure safety monitoring. The hierarchical model for online evaluation of the operating status of a concrete dam is obtained, wherein the hierarchical model for online evaluation includes a monitoring data layer, a diagnostic method layer, an evaluation indicator layer, a monitoring project layer, a key part layer, and an overall engineering layer. When performing evaluation and analysis based on the hierarchical model, the method includes: collecting the same type of multi-measurement point monitoring data and multi-type multi-measurement point monitoring data of the concrete dam, wherein the same type of multi-measurement point monitoring data includes deformation monitoring data, seepage monitoring data, stress-strain monitoring data, and temperature monitoring data; determining the analysis results of various indicators of the concrete dam based on the same type of multi-measurement point monitoring data; determining the operation scoring results of various indicators of the concrete dam based on the multi-type multi-measurement point monitoring data; and performing overall evaluation and analysis based on the analysis results of various indicators and the operation scoring results of various indicators.

Description

Technical Field

[0001] The present application relates to the technical field of concrete dam structure safety monitoring, and in particular to a method for formulating evaluation indicators based on a hierarchical architecture for online evaluation of concrete dam operating performance. Background Art

[0002] With the enrichment of dam monitoring technology and the increase in the frequency of automated monitoring, monitoring data has not only increased significantly in quantity, but also continued to grow in dimension. The current dam safety evaluation is mainly based on post-operation periodic evaluation, that is, after the dam has been in operation for a period of time (week, month, quarter, year, etc.), the dam condition monitoring data and on-site inspection data during the period are compiled, and statistical models or other models are established at key measurement points of dam deformation, seepage, stress strain and temperature monitoring items, and the trend of data changes at typical measurement points in key areas is analyzed. Combined with the experience and judgment of on-site operation and maintenance personnel and hydraulic technical experts, the dam operation safety is analyzed and evaluated.

[0003] During the operation period of concrete dam, it is affected by multiple factors such as load, structural type, topographic and geological conditions, and physical and mechanical properties of materials. It is difficult to make a timely and comprehensive analysis of the overall operating status of the concrete dam by relying solely on traditional monitoring data analysis methods. There is an increasingly urgent need to deeply explore the correlation between monitoring data and comprehensively analyze multi-dimensional data to jointly reflect the potential structural safety hazards of the project. Summary of the invention

[0004] The present application aims to solve one of the technical problems in the related art at least to some extent.

[0005] To this end, one purpose of the present application is to propose a method for formulating evaluation indicators based on an online evaluation hierarchical architecture for concrete dam operating performance, by obtaining an online evaluation hierarchical model for concrete dam operating performance, wherein the online evaluation hierarchical model includes a monitoring data layer, a diagnostic method layer, an evaluation indicator layer, a monitoring project layer, a key part layer, and an overall engineering layer; when performing an overall evaluation analysis of a concrete dam based on the online evaluation hierarchical model, the following steps are included: collecting the same type of multi-point monitoring data and multi-type multi-point monitoring data of the concrete dam, wherein the same type of multi-point monitoring data includes deformation monitoring Data, seepage monitoring data, stress-strain monitoring data, and temperature monitoring data; based on the same type of multi-measurement point monitoring data, determine the analysis results of various indicators of the concrete dam, wherein the analysis results of various indicators include deformation analysis results, seepage analysis results, stress-strain analysis results, and temperature analysis results; based on multi-type and multi-measurement point monitoring data, determine the operation scoring results of various indicators of the concrete dam, wherein the operation scoring results of various indicators include deformation scoring results, seepage scoring results, and stress-strain scoring results; conduct an overall evaluation and analysis of the concrete dam based on the analysis results of various indicators and the operation scoring results of various indicators.

[0006] The second purpose of the present application is to provide an online evaluation device for the operating performance of a concrete dam.

[0007] The third objective of the present application is to provide an electronic device.

[0008] A fourth object of the present application is to provide a non-transitory computer-readable storage medium.

[0009] A fifth object of the present application is to provide a computer program product.

[0010] To achieve the above-mentioned purpose, the first embodiment of the present application proposes a method for formulating evaluation indicators based on an online hierarchical architecture for judging the operating performance of a concrete dam, comprising: obtaining an online hierarchical model for judging the operating performance of a concrete dam, wherein the online hierarchical model comprises a monitoring data layer, a diagnostic method layer, an evaluation indicator layer, a monitoring project layer, a key part layer, and an overall engineering layer; when performing an overall evaluation and analysis of the concrete dam based on the online hierarchical model, the following steps are included: collecting the same type of multi-measurement point monitoring data and multi-type multi-measurement point monitoring data of the concrete dam, wherein the same type of multi-measurement point monitoring data comprises variable Determine the analysis results of various indicators of the concrete dam based on the same type of multi-point monitoring data, where the analysis results of various indicators include deformation analysis results, seepage analysis results, stress strain analysis results, and temperature analysis results; Determine the operation scoring results of various indicators of the concrete dam based on multi-type and multi-point monitoring data, where the operation scoring results of various indicators include deformation scoring results, seepage scoring results, and stress strain scoring results; Conduct an overall evaluation and analysis of the concrete dam based on the analysis results of various indicators and the operation scoring results of various indicators.

[0011] According to an embodiment of the present application, the overall engineering layer is the target of online evaluation of the operating status of the concrete dam, and is used to comprehensively evaluate the overall operating status of the concrete dam; the key part layer is the focus of online evaluation of the operating status of the concrete dam, and the key parts can be determined based on the calculation analysis in the design stage of the concrete dam, or the cluster analysis of the long-term operation monitoring data, wherein the key part layer analyzes the special evaluation results of the deformation, seepage, stress-strain, and temperature monitoring effect quantities derived from the monitoring project layer, and the key parts are mainly distributed in the upper dam body, the middle dam body, the heel of the riverbed dam section, the toe of the riverbed dam section, the dam foundation, and the defect distribution areas of the long-term concern of the project operation; the monitoring project layer is the associated medium for online evaluation of the operating status of the concrete dam, and according to the currently implemented concrete dam design and safety monitoring related specifications for concrete dams, the deformation, seepage, stress-strain and temperature monitoring effect quantities are comprehensively analyzed; the evaluation index layer It is the process center of online evaluation of concrete dam operation status. Four types of evaluation indicators, namely deformation, seepage, stress-strain and temperature, are set corresponding to the deformation, seepage, stress-strain and temperature monitoring effect quantities of concrete dam operation safety at the monitoring project layer. Among them, under each type of evaluation indicator, it is further divided into two indicator formulation methods: one is the evaluation indicator of the same type of multi-measurement point combination calculation based on structural characteristics; the other is the comprehensive evaluation indicator of multiple types and multiple measurement points based on data-driven partitioning; the diagnostic method layer is the analysis core of online evaluation of concrete dam operation status. Based on the periodic action of reservoir water and temperature loads during the operation period of concrete dam and the changes in the internal structure and materials of concrete dam, the monitoring effect quantity of multiple points of the same type or multiple types and multiple points is analyzed according to the combined calculation of multiple measurement points of the same type and the comprehensive evaluation indicators of multiple types and multiple measurement points proposed by the evaluation indicator layer; the monitoring data layer is the information base for online evaluation of concrete dam operation status, including engineering safety monitoring, patrol inspection defect data and geophysical detection results data.

[0012] According to an embodiment of the present application, based on the same type of multi-measurement point monitoring data, the deformation analysis results in each indicator analysis result of the concrete dam are determined, including: based on the deformation monitoring data in the same type of multi-measurement point monitoring data, the deformation analysis parameters of the concrete dam are determined, wherein the deformation analysis parameters include the vertical distribution of dam foundation deformation, the overall horizontal displacement of the dam body, the coordination degree of the horizontal displacement of the dam body, the overall deflection of the dam section, the overall settlement of the dam body, the micro-deformation between dam layers, the overall opening and closing of the foundation surface, and the overall opening and closing of the structural joints; based on the deformation analysis parameters, the concrete dam is subjected to deformation analysis to obtain the deformation analysis results.

[0013] According to an embodiment of the present application, based on the same type of multi-measurement point monitoring data, the seepage analysis results in the various index analysis results of the concrete dam are determined, including: based on the seepage monitoring data in the same type of multi-measurement point monitoring data, the seepage analysis parameters of the concrete dam are determined, wherein the seepage analysis parameters include the overall uplift pressure reduction coefficient of the dam foundation, the anti-sliding stability of the dam foundation, the seepage pressure gradient of the structural joints, and the seepage amplitude of the key areas; based on the seepage analysis parameters, the seepage analysis of the concrete dam is performed to obtain the seepage analysis results.

[0014] According to an embodiment of the present application, based on the same type of multi-measurement point monitoring data, the stress-strain analysis results in each indicator analysis result of the concrete dam are determined, including: based on the stress-strain monitoring data in the same type of multi-measurement point monitoring data, the stress-strain analysis parameters of the concrete dam are determined, wherein the stress-strain analysis parameters include the vertical distribution of dam foundation stress, the overall horizontal stress of the dam body, the overall stress in the vertical beam direction, and the amplitude of strain measurement points in special parts; based on the stress-strain analysis parameters, stress-strain analysis is performed on the concrete dam to obtain stress-strain analysis results.

[0015] According to an embodiment of the present application, based on the same type of multi-measurement point monitoring data, the temperature analysis results in the analysis results of various indicators of the concrete dam are determined, including: based on the temperature monitoring data in the same type of multi-measurement point monitoring data, the temperature analysis parameters of the concrete dam are determined, wherein the temperature analysis parameters include the temperature distribution of the dam foundation and the temperature gradient of the dam body; based on the same temperature analysis parameters, the temperature analysis of the concrete dam is performed to obtain the temperature analysis results.

[0016] According to an embodiment of the present application, based on multi-type and multi-measurement point monitoring data, the operation score results of each indicator of the concrete dam are determined, including: dividing the key operating parts of the concrete dam into regions, and obtaining the time series measurement data of different types of monitoring instruments in a certain partition from the multi-type and multi-measurement point monitoring data; establishing a graph structure for the time series measurement data in the time and variable dimensions respectively, to obtain a time feature graph and a variable feature graph; inputting the time feature graph and the variable feature graph into the time graph attention network and the variable graph attention network respectively, to obtain a time attention matrix and a variable attention matrix; splicing the time series measurement data, the time attention matrix and the variable attention matrix together and inputting them into a gated convolutional network to obtain a target feature; calculating an anomaly score based on the target feature, comparing the anomaly score with a preset indicator threshold, and obtaining the operation score result of the corresponding indicator of the concrete dam.

[0017] According to one embodiment of the present application, calculating the anomaly score based on the target feature includes: inputting the target feature into a prediction module and a reconstruction module to obtain a prediction value and a reconstruction probability; and calculating the anomaly score based on the prediction value and the reconstruction probability.

[0018] According to one embodiment of the present application, the prediction module is a multi-layer perceptron.

[0019] According to one embodiment of the present application, the reconstruction module includes a discriminator and an autoencoder.

[0020] To achieve the above-mentioned purpose, the second embodiment of the present application proposes an online evaluation device for the operating performance of a concrete dam, including: a data acquisition module, used to collect the same type of multi-point monitoring data and multi-type multi-point monitoring data of the concrete dam, wherein the same type of multi-point monitoring data includes deformation monitoring data, seepage monitoring data, stress-strain monitoring data, and temperature monitoring data; a first determination module, used to determine the analysis results of various indicators of the concrete dam based on the same type of multi-point monitoring data, wherein the analysis results of various indicators include deformation analysis results, seepage analysis results, stress-strain analysis results, and temperature analysis results; a second determination module, used to determine the operation scoring results of various indicators of the concrete dam based on the multi-type multi-point monitoring data, wherein the operation scoring results of various indicators include deformation scoring results, seepage scoring results, and stress-strain scoring results; an evaluation and analysis module, used to perform an overall evaluation and analysis of the concrete dam based on the analysis results of various indicators and the operation scoring results of various indicators.

[0021] To achieve the above-mentioned purpose, the third aspect embodiment of the present application proposes an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to implement the method for formulating evaluation indicators based on the hierarchical architecture for online evaluation of the operating performance of concrete dams as described in the first aspect embodiment of the present application.

[0022] To achieve the above-mentioned purpose, the fourth aspect embodiment of the present application proposes a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to implement the method for formulating evaluation indicators based on the hierarchical architecture for online evaluation of the operating performance of concrete dams as described in the first aspect embodiment of the present application.

[0023] To achieve the above-mentioned purpose, the fifth aspect embodiment of the present application proposes a computer program product, including a computer program, which, when executed by a processor, implements the method for formulating evaluation indicators based on the hierarchical architecture for online evaluation of the operating performance of concrete dams as described in the first aspect embodiment of the present application.

[0024] This application achieves at least the following beneficial effects: This application has formulated evaluation indicators that reflect the structural characteristics of the dam, including the combination calculation of multiple measurement points of the same type for deformation, the combination calculation of multiple measurement points of the same type for seepage, the combination calculation of multiple measurement points of the same type for stress and strain, and the combination calculation of multiple measurement points of the same type for temperature. At the same time, based on the data-driven concept, it has also formulated multi-type multi-measurement point partition comprehensive evaluation indicators that can be applied to monitoring projects such as deformation, seepage, and stress and strain. Combined with the formulation of online evaluation indicators, each specific online diagnostic method sub-item is further refined to lay a solid foundation for the subsequent quantitative derivation of the hierarchical model for online evaluation of the operating performance of concrete dams, and the realization of rapid intelligent evaluation of the operating performance of concrete dams that integrates structural analysis and data-driven. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0026] Figure 1 It is a schematic diagram of an exemplary implementation of a method for evaluating the operating performance of a concrete dam shown in an embodiment of the present application.

[0027] Figure 2 It is a schematic diagram of an online evaluation framework for the operating performance of a concrete dam shown in this application.

[0028] Figure 3 It is a schematic diagram of a framework of a multi-index graph attention network model for key parts of a concrete dam shown in an embodiment of the present application.

[0029] Figure 4 It is an architecture diagram of F-GAT shown in one embodiment of the present application.

[0030] Figure 5 This is a diagram of a generative adversarial network framework shown in one embodiment of the present application.

[0031] Figure 6 It is a schematic diagram of a device for evaluating the operating performance of a concrete dam shown in one embodiment of the present application. DETAILED DESCRIPTION

[0032] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0033] Figure 1 is a schematic diagram of an exemplary implementation of a method for formulating evaluation indicators based on a hierarchical architecture for online evaluation of concrete dam operating performance shown in the present application, such as Figure 1As shown, the method for formulating evaluation indicators based on the hierarchical architecture of online evaluation of concrete dam operating performance includes the following steps:

[0034] S101, obtaining an online evaluation hierarchical model for the operating performance of a concrete dam, wherein the online evaluation hierarchical model includes a monitoring data layer, a diagnostic method layer, an evaluation index layer, a monitoring project layer, a key part layer, and an overall engineering layer.

[0035] Figure 2 is a schematic diagram of an online evaluation framework for the operation performance of a concrete dam shown in this application, such as Figure 2 As shown in the figure, the hierarchical architecture of online evaluation of concrete dam operation performance is divided into the following from top to bottom: Figure 2 As shown:

[0036] (1) Overall engineering layer

[0037] The overall engineering layer is the ultimate goal of online evaluation of the operating performance of concrete dams, that is, to conduct a comprehensive assessment of the overall operating performance of concrete dams.

[0038] (2) Key parts layer

[0039] The key parts layer is the focus of the online evaluation of the operating performance of concrete dams. The key parts can be determined based on the calculation analysis in the dam design stage or the cluster analysis of long-term operation monitoring data. The key parts layer analyzes the special evaluation results of the monitoring effect quantities such as deformation, seepage, stress strain, and temperature derived from the monitoring project layer. The key parts are mainly distributed in the upper dam body, the middle dam body, the heel of the riverbed dam section, the toe of the riverbed dam section, the dam foundation, and the distribution of defects that have been paid attention to in the long-term operation of the project.

[0040] (3) Monitoring project layer

[0041] The monitoring project layer is the associated medium for online evaluation of the operating performance of concrete dams. In accordance with the current specifications for concrete dam design and safety monitoring, a comprehensive analysis of monitoring effect quantities such as deformation, seepage, stress-strain inspections and geophysical detection is conducted to lay the foundation for the performance evaluation of key parts of concrete dams.

[0042] (4) Evaluation index layer

[0043] The evaluation index layer is the process center of the online evaluation of the operation status of concrete dams. Four types of evaluation indicators, such as deformation, seepage, stress-strain and temperature, are set corresponding to the monitoring items of the upper layer of concrete dam operation safety deformation, seepage, stress-strain and temperature. Under each type of evaluation index, there are further two index formulation methods: one is the evaluation index of the same type of multi-measurement point combination calculation based on structural characteristics; the other is the comprehensive evaluation index of multiple types and multiple measurement points based on data-driven.

[0044] The same type of multi-measurement point combination calculation evaluation indicators under the deformation evaluation index are subdivided into vertical distribution of dam foundation deformation, overall horizontal displacement of dam body, coordination of dam body horizontal displacement, overall deflection of dam section, overall settlement of dam body, micro-deformation between dam layers, overall opening and closing of foundation surface, overall opening and closing of structural joints, etc. In addition, there are multi-type and multi-measurement point data comprehensive evaluation indicators used for deformation zoning monitoring.

[0045] The same type of multi-measurement point combination calculation evaluation indicators under the seepage evaluation index are subdivided into reduction of overall uplift pressure of dam foundation, anti-sliding stability of dam foundation, seepage pressure gradient of dam foundation structural joints, seepage amplitude in key areas, seepage zoning monitoring, etc. In addition, there are multi-type multi-measurement point data comprehensive evaluation indicators used for seepage zoning monitoring.

[0046] The same type of multi-measurement point combination calculation evaluation indicators under the stress-strain evaluation indicators are subdivided into vertical distribution of dam foundation stress, overall horizontal stress of the dam body, overall stress in the vertical beam direction, amplitude of strain measurement points in special parts, etc. In addition, there are multi-type multi-measurement point data comprehensive evaluation indicators used for stress-strain zoning monitoring.

[0047] Temperature includes temperature gradient of dam foundation, temperature gradient of dam body and other heterogeneous evaluation indicators.

[0048] (5) Diagnostic method layer

[0049] The diagnostic method layer is the core of the analysis of the intelligent evaluation system for the operation performance of concrete dams. It focuses on the periodic effects of reservoir water, temperature and other loads during the operation of concrete dams and the changes in their internal structure and materials. According to the combined calculation of multiple measuring points of the same type and the comprehensive evaluation indicators of multiple measuring points of multiple types proposed in the evaluation index layer, the monitoring effect analysis of multiple points of the same type or multiple types of multiple points is carried out. According to the evaluation indicators under the deformation monitoring project, the vertical distribution of dam foundation deformation, the overall horizontal displacement of the dam body, the coordination degree of the horizontal displacement of the dam body, the overall deflection of the dam section, the overall settlement of the dam body, the micro-deformation between dam layers, and the overall opening and closing degree of the foundation surface are proposed. , overall opening and closing degree of structural joints, deformation zoning monitoring and other diagnostic methods; for the evaluation indicators under the seepage monitoring project, the overall uplift pressure reduction of the dam foundation, the anti-sliding stability of the dam foundation, the seepage pressure gradient of the structural joints, the seepage amplitude of key areas, and the seepage zoning monitoring and other diagnostic methods are proposed; for the evaluation indicators under the stress and strain monitoring project, the vertical distribution of dam foundation stress, the overall horizontal stress of the dam body, the overall stress of the vertical beam, the amplitude of strain measurement points in special parts, and the stress and strain zoning monitoring and other diagnostic methods are proposed; for the evaluation indicators under the temperature monitoring project, the specific diagnostic methods include the temperature change of the dam foundation and the temperature gradient of the dam body. The calculated result value is time series data, and the corresponding relationship with the evaluation set is established through the time series adaptive analysis method and the Rhineda criterion.

[0050] (6) Monitoring data layer

[0051] The monitoring data layer is the information base for online evaluation of the operating status of concrete dams, including engineering safety monitoring, inspection defect data and geophysical detection results data. Engineering safety monitoring data mainly include monitoring data of various measurement points such as deformation, seepage, stress and strain, such as plumb line, multi-point displacement meter, static level displacement meter, unidirectional or bidirectional joint gauge, piezometer, pressure tube, water measuring weir, unidirectional strain gauge, multi-directional strain gauge group, anchor stress gauge, and steel bar gauge; inspection defect data mainly include cracks, joint misalignment, seepage, calcification precipitation found on the dam top or upstream and downstream dam surface, and corridors inside the dam, as well as cavitation and erosion of flow holes in the dam body, and other defect data with potential risks and further development trends; geophysical detection results data mainly include results data with potential risks and need to be continuously paid attention to, which are found by various technical inspections or measurement methods such as sound waves, images, lasers, and electromagnetic waves.

[0052] S102, collecting the same type of multi-measurement point monitoring data and multi-type multi-measurement point monitoring data of the concrete dam, wherein the same type of multi-measurement point monitoring data includes deformation monitoring data, seepage monitoring data, stress strain monitoring data, and temperature monitoring data.

[0053] S103, based on the same type of multi-point monitoring data, determining various index analysis results of the concrete dam, wherein the various index analysis results include deformation analysis results, seepage analysis results, stress-strain analysis results, and temperature analysis results.

[0054] S104, based on the multi-type and multi-point monitoring data, determining the operation score results of various indicators of the concrete dam, wherein the operation score results of various indicators include deformation score results, seepage score results, and stress-strain score results.

[0055] S105, conducting an overall evaluation and analysis of the concrete dam based on the analysis results of each indicator and the operation scoring results of each indicator.

[0056] This application has formulated evaluation indicators that reflect the structural characteristics of the dam, including the combination calculation of multiple measurement points of the same type for deformation, the combination calculation of multiple measurement points of the same type for seepage, the combination calculation of multiple measurement points of the same type for stress and strain, and the combination calculation of multiple measurement points of the same type for temperature. At the same time, based on the data-driven concept, it has also formulated multi-type multi-measurement point partition comprehensive evaluation indicators that can be applied to monitoring projects such as deformation, seepage, and stress and strain. Combined with the formulation of online evaluation indicators, each specific online diagnostic method sub-item is further refined to lay a solid foundation for the subsequent quantitative derivation of the hierarchical model for online evaluation of the operating performance of concrete dams, and the realization of rapid intelligent evaluation of the operating performance of concrete dams that integrates structural analysis and data-driven.

[0057] Furthermore, in S103, when determining the deformation analysis results in the various indicator analysis results of the concrete dam based on the same type of multi-measurement point monitoring data, it is necessary to determine the deformation analysis parameters of the concrete dam based on the deformation monitoring data in the same type of multi-measurement point monitoring data, wherein the deformation analysis parameters include the vertical distribution of dam foundation deformation, the overall horizontal displacement of the dam body, the coordination degree of the horizontal displacement of the dam body, the overall deflection of the dam section, the overall settlement of the dam body, the micro-deformation between dam layers, the overall opening and closing of the foundation surface, and the overall opening and closing of the structural joints; based on the deformation analysis parameters, a deformation analysis of the concrete dam is performed to obtain the deformation analysis results.

[0058] The following will introduce the analysis process of the vertical distribution of dam foundation deformation, the overall horizontal displacement of the dam body, the coordination of the horizontal displacement of the dam body, the overall deflection of the dam section, the overall settlement of the dam body, the micro-deformation between dam layers, the overall opening and closing of the foundation surface, and the overall opening and closing of the structural joints in the deformation analysis.

[0059] 1. The vertical distribution analysis process of dam foundation deformation is as follows:

[0060] According to existing research, the influence of reservoir water and other loads on the deformation of dam foundation during the operation of 200m to 300m high concrete dams will reach 2 to 3 times the dam height in the vertical direction, which has exceeded the limit of existing monitoring methods. Combined with measured data, the analysis method of vertical distribution of dam foundation deformation is studied to analyze the vertical influence range of concrete dam foundation deformation during operation.

[0061] When selecting instruments for monitoring the vertical distribution of dam foundation deformation, three inverted plumb lines arranged in the vertical direction of the dam section with maximum dam height but at different depths are selected. The length of the first inverted plumb line is usually 10% of the maximum dam height, the length of the second inverted plumb line is usually 30% of the maximum dam height, and the length of the third inverted plumb line is usually 50% of the maximum dam height.

[0062] According to the distribution law of the inverted vertical deformation along different depths, the relevant calculation model function is established through formula (1):

[0063] y=x / (k1+k2x) (1)

[0064] Where y is the deformation corresponding to different depths, x is the vertical depth, and k1 and k2 are unknown coefficients.

[0065] Set the inverted vertical line No. 1 (x1, y1), the inverted vertical line No. 2 (x2, y2), and the inverted vertical line No. 3 (x3, y3), and use nonlinear fitting to calculate k1 and k2, that is, establish the calculation function of the vertical line depth x and the deformation y, and deduce the theoretical maximum deformation y of the dam foundation by deriving the calculation function y. max .

[0066] Calculate the different depths x of the dam foundation by formula (2) i The corresponding deformation y iThe theoretical maximum deformation of the dam foundation y max Ratio of:

[0067] α=y i / y max (2)

[0068] In the formula, when α=98%, the corresponding depth xi is considered to be the deformation depth of the dam foundation.

[0069] Carry out comparative analysis of vertical distribution of dam foundation deformation over time. Specifically: Fit the calculated values ​​x of vertical distribution of dam foundation deformation over time i , x i+1 , …, x i+n , get the current forecast value x i ', and compare it with the latest x calculated based on actual measurements i A comparison is made and the Rhineda criterion is used to determine the size of the difference, so as to quantitatively analyze whether the vertical distribution of the current dam foundation deformation has a sudden change or a larger amplitude than in the past.

[0070] 2. The overall horizontal displacement analysis process of the dam body is as follows:

[0071] When a concrete dam is subjected to loads such as reservoir water, the loads at the upper and lower elevations of the dam body are mainly horizontal forces. This paper studies the calculation method of the overall displacement of the dam body at representative elevations, and uses this method to analyze the overall displacement of the concrete dam at different elevation planes.

[0072] When selecting instruments for monitoring the horizontal displacement of the dam, vertical lines are selected from representative dam sections, such as the riverbed section with the maximum dam height, the overcurrent dam section, and the dam section affected by geological defects on the bank slope. At the same time, surface deformation monitoring points are selected on the horseways behind the dam at various typical elevations to further verify the vertical line monitoring results and improve the monitoring coverage density.

[0073] The calculation idea of ​​the overall horizontal displacement of the dam body is to establish the correlation curve of the current monitoring value of the horizontal displacement of each typical elevation, and analyze the overall plane displacement of a typical elevation through the distribution probability of the correlation coefficient of the multi-period curve.

[0074] Specifically, the calculation method of the overall horizontal displacement of the dam body is as follows:

[0075] According to the spatial plane coordinate measurement value A of the deformation monitoring point on the same elevation of the vertical line of the representative dam section or the surface of the dam road 1n (x 1n ,y 1n ), A 2n (x 2n ,y 2n ), …, A mn (x mn ,y mn ) Establish spatial correlation model t (xmn ,y mn ), which reflects the plane displacement of the arch ring. Among them: x mn and mn They are respectively the spatial plane coordinate monitoring values ​​of the deformation monitoring point on the nth layer of the mth dam section or the surface of the horseway behind the dam.

[0076] Using Euclidean distance, the spatial correlation model of the typical horizontal displacement monitoring data in the past period is calculated by formula (3): t-k With f t-(k-1) Similarity, the similarity r of the spatial correlation curves of two typical elevation horizontal displacements in the past is obtained 1 t .

[0077]

[0078] When conducting comparative analysis of the duration of the overall horizontal displacement of the dam body, the similarity coefficient r of the spatial correlation curve of the typical elevation horizontal displacement duration is fitted. 1 t , r 1 t-1 ,…,r 1 t-k , get the current forecast value r 1 t ', and compare it with the r calculated by the new period actual measurement 1 t By comparison, the Rhineda criterion is used to quantitatively analyze whether the overall deformation of the dam body at present has a sudden change or a larger change amplitude compared with the previous period.

[0079] 3. The coordination analysis process of dam horizontal displacement is as follows:

[0080] The overall deformation behavior of concrete dams after loading is an important basis for judging whether their working conditions are normal. However, due to the influence of geology, structure and temperature, the dam body is prone to tangential asymmetric deformation during operation. This paper studies the calculation method of the displacement coordination degree of the representative elevation plane of the dam to analyze the asymmetric mutation and trend development of the displacement of the left and right bank sections of the dam.

[0081] When selecting instruments for monitoring the horizontal displacement of the dam body, vertical lines are selected from representative dam sections, such as the riverbed section with the maximum dam height, the overcurrent dam section, and the dam section affected by geological defects of the bank slope. At the same time, surface deformation monitoring points are selected on the horseways behind the dam at various typical elevations to further verify the vertical line monitoring results and improve the monitoring coverage density.

[0082] Calculation idea of ​​coordination degree of horizontal displacement of dam body: take the center line of a dam body as X-axis, define the dam body transverse river deformation monitoring data as positive to the left bank and negative to the right bank, and define the dam body downstream deformation monitoring data as positive and upstream as negative. Establish the current displacement monitoring value correlation curves of the left and right bank dam sections respectively, transform the spatial correlation curve of the displacement monitoring value of the right bank dam section to the first quadrant of the Cartesian coordinate system and compare it with the spatial correlation curve of the displacement monitoring value of the left bank dam section, and describe the displacement coordination degree of the left and right bank dam sections of a typical elevation through the probability of correlation coefficient distribution.

[0083] Specifically, the calculation method of the coordination degree of the dam body horizontal displacement is as follows:

[0084] According to the spatial plane coordinates of the representative dam section at the same elevation or the deformation monitoring point on the bridge behind the dam, A 1n (x 1n ,y 1n ), A 2n (x 2n ,y 2n ), …, A mn (x mn ,y mn ) Establish a spatial correlation model of a certain elevation of the left bank of the dam body f Z (x mn ,y mn ) and the spatial correlation model of the right bank dam section f Y (x mn , -y mn ). Where: x mn and mn They are respectively the spatial plane coordinate monitoring values ​​of the n-layer perpendicular line of the m-th dam section or the surface deformation monitoring point of the bridge behind the dam.

[0085] The spatial correlation curve of the displacement monitoring value of the right bank dam section is transformed into coordinates by formula (4): Y (x mn , -y mn )→f Y '(x',y'), to the first quadrant of the Cartesian coordinate system.

[0086]

[0087] The Pearson correlation coefficient is used to calculate the spatial correlation model of the current monitoring data of the left and right bank dam displacements. Z With f Y Correlation is used to describe the coordination degree of displacement between the left and right bank dam sections. 2 t .

[0088] Comparative analysis of the coordination degree of the horizontal displacement of the dam body over time: The coordination degree r of the displacement curves of the left and right bank dam sections 2t , r 2 t-1 ,…,r 2 t-k Fit and build a time series forecasting model. 2 t 'Compared to the latest period calculated based on actual measurements 2 t The PauTa criterion is used to compare and determine the difference, so as to quantitatively analyze whether the displacement coordination degree of the left and right bank dam sections in the current period has a sudden change or a larger change amplitude compared with the previous period.

[0089] 4. The overall deflection analysis process of the dam section is as follows:

[0090] When a concrete dam is subjected to loads such as reservoir water, the upper and lower parts of the dam body are subjected to force together and play a bearing role. This paper studies the calculation method of the overall deflection of the dam section to analyze the deformation of the vertical beams of each dam section.

[0091] When selecting the instrument for monitoring the overall deflection of the dam section, a representative dam section is selected, and vertical lines are laid out in sections from the dam top downward. Vertical lines are used in key monitoring dam sections, and monitoring data from the horseway table behind the dam is used in non-key monitoring dam sections.

[0092] The idea of ​​calculating the overall deflection of the dam section is to calculate the beam deflection deformation by taking a typical dam section as an example, and to establish the deflection distribution curve of the vertical displacement variable along the river from top to bottom of the dam section and the corresponding elevation. The change of the overall deflection of the dam section is analyzed by the distribution probability of the correlation coefficient of the deflection distribution curve over multiple periods.

[0093] Specifically, the calculation method of the overall deflection of the dam section is as follows:

[0094] According to the displacement variable and the corresponding elevation A of the vertical line along the river from top to bottom of a representative dam section 11 (x 11 ,h 11 ), A 12 (x 12 ,h 12 ),…,A 1n (x 1n ,h 1n ) Establish the spatial correlation model of the dam body t (x mn ,h mn ), which reflects the deflection deformation of a representative dam section. mn is the deformation monitoring value along the river at the measuring point of layer n in the m-th dam section, h mn Arrange the elevations of the n-layer measuring points for the m-dam section.

[0095] The spatial correlation model g of the beam deformation monitoring data in the past period is calculated by formula (5):t-k With g t-(k-1) The Pearson correlation coefficient of the previous two adjacent beam deflection deformation spatial correlation curves is obtained. 3 t .

[0096]

[0097] In the formula, Cov(g t-k , g t-(k-1) ) is g t-k , g t-(k-1 The covariance of t-k )、D(g t-(k-1) ) are g t-k , g t-(k-1) The variance of .

[0098] Comparative analysis of the overall deflection duration of the dam section: correlation coefficient r of the spatial correlation curve of the dam body deflection duration deformation 3 t , r 3 t-1 ,…,r 3 t-k , build a time series forecasting model. Through the current forecast value r 3 t 'Compared to the latest period calculated based on actual measurements 3 t A comparison is made and the Rhineda criterion is used to determine the size of the difference, so as to quantitatively analyze whether the current beam deflection deformation has a sudden change or a larger change amplitude than in the past.

[0099] 5. The overall settlement analysis process of the dam is as follows:

[0100] When a concrete dam is subjected to loads such as the dam's own weight and reservoir water, it will undergo settlement and deformation in the vertical direction. This paper studies the calculation method of the overall settlement and deformation of the dam body in order to analyze the vertical settlement and deformation of the dam body.

[0101] When selecting instruments for monitoring dam settlement and deformation, select static level points or artificial level points located on the dam crest, inspection corridors inside the dam at typical elevations, and horseways behind the dam.

[0102] Calculation idea of ​​dam settlement and deformation: establish correlation curves of current monitoring quantities of settlement and deformation at each typical elevation, and analyze the overall settlement of a typical elevation through the distribution probability of the correlation coefficient of multi-period curves.

[0103] Specifically, the calculation method of dam body settlement deformation is as follows:

[0104] According to the current monitoring value of vertical settlement deformation of each dam section at the same elevation and the corresponding elevation A 11 (x11 ,h 11 ), A 12 (x 12 ,h 12 ),…,A 1n (x 1n ,h 1n ) Establish the spatial correlation model of settlement and deformation of each dam section G t (x mn ,h mn ). where x mn is the vertical settlement deformation monitoring value of the measuring point at layer n of the m-th dam section, h mn It is the corresponding elevation of the measuring point on layer n of the m dam section.

[0105] The spatial correlation model G of the past monitoring data of the overall settlement and deformation of the dam body is calculated by formula (6): t-k With G t-(k-1) The Pearson correlation coefficient of the previous two adjacent dam body overall settlement deformation spatial correlation curves was obtained. 4 t .

[0106]

[0107] In the formula, Cov(G t-k , G t-(k-1) ) is G t-k , G t-(k-1) The covariance of D(G t-k )、D(G t-(k-1) ) are G t-k , G t-(k-1) The variance of .

[0108] Comparative analysis of the overall settlement and deformation of the dam body: The correlation coefficient r of the spatial correlation curve of the overall settlement and deformation of the dam body 4 t , r 4 t-1 ,…,r 4 t-k , build a time series forecasting model. Through the current forecast value r 4 t 'Compared to the latest period calculated based on actual measurements 4 t By comparing the results, the Rhineda criterion is used to determine the size of the difference, so as to quantitatively analyze whether the current vertical settlement deformation has a sudden change or a larger change amplitude than in the past.

[0109] 6. The overall opening and closing analysis process of the foundation surface is as follows:

[0110] The arch dam bears the huge thrust of the upstream reservoir water. The anti-sliding stability of the dam is the focus of attention during operation. This paper studies the calculation method of the overall opening and closing degree of the arch dam foundation surface to analyze the anti-sliding stability of the dam foundation.

[0111] When selecting instruments for monitoring the opening and closing degree of the foundation surface, a unidirectional joint meter is selected, which is arranged radially from the upstream surface to the downstream surface in a straddling manner at the interface between the bedrock of the dam foundation and the concrete dam body.

[0112] Calculation idea of ​​the opening and closing degree of the foundation surface: Establish the spatial distribution curve of the current measured values ​​of the one-way seam meter arranged vertically on the foundation surface, and analyze the overall opening and closing of the foundation surface through the distribution probability of the correlation coefficient of the multi-period curve.

[0113] Specifically, the calculation method of the opening and closing degree of the foundation surface is as follows:

[0114] According to the current monitoring quantity x1, x2, ..., x of the same plane joint meter n Establish a spatial correlation model between monitoring data and dam section distribution t (x), which reflects the overall opening and closing degree of the building surface. n It is the current monitoring quantity of the joint meter n.

[0115] The spatial correlation model J of the past monitoring data of the base surface opening and closing degree is established by formula (7). t With J t-1 The Pearson correlation coefficient of the previous two adjacent construction surfaces is obtained, and the correlation coefficient r of the spatial correlation curve of the overall opening and closing degree of the foundation surface is obtained. 5 t .

[0116]

[0117] In the formula, Cov(J t , J t-1 ) is J t , J t-1 The covariance of t )、D(J t-1 ) are J t , J t-1 The variance of .

[0118] Comparative analysis of the duration of the opening and closing degree of the foundation surface: The correlation coefficient r of the spatial correlation curve of the duration of the opening and closing degree of the foundation surface seam 5 t , r 5 t-1 ,…,r 5 t-k Fit and build a time series forecasting model. 5 t 'Compared to the latest period calculated based on actual measurements 5t For comparison, the PauTa criterion is used to determine the size of the difference, so as to quantitatively analyze whether the overall opening and closing degree of the current foundation surface seam has changed significantly compared with previous periods.

[0119] 7. The overall opening and closing analysis process of the structural joint is as follows:

[0120] In order to prevent the volume expansion of large-volume concrete and the possible tensile stress caused by the huge upstream water thrust, which may lead to cracking of the dam body, multiple types of structural joints such as transverse joints, longitudinal joints, and induced joints are set up in the concrete dam. The calculation method of the overall opening and closing degree of the joint surfaces of various structural joints in the dam body is studied to analyze the actual working conditions of the structural joints during operation.

[0121] When selecting an instrument for monitoring the opening and closing degree of a structural joint, a unidirectional joint meter is selected that is radially arranged from the upstream surface to the downstream surface in a straddling manner on the structural joint surface.

[0122] Calculation idea of ​​structural joint opening and closing degree: Establish the spatial distribution curve of the current measured values ​​of the unidirectional joint meter arranged vertically on the structural joint plane, and analyze the overall opening and closing of the joint surface through the distribution probability of the correlation coefficient of the multi-period curve.

[0123] Specifically, the calculation method of the structural joint opening degree is as follows:

[0124] According to the current monitoring quantity x1, x2, ..., x of the same plane joint meter n Establish a spatial correlation model between monitoring data and dam section distribution t (x), reflects the overall opening and closing degree of the structural joint. n It is the current opening and closing degree monitoring value of the joint meter n.

[0125] The spatial correlation model J of the past monitoring data of the structural joint opening and closing degree is obtained by formula (8). t With J t-1 The Pearson correlation coefficient of the previous two adjacent structural joints overall opening and closing degree spatial correlation curves is obtained. 6 t .

[0126]

[0127] In the formula, Cov(J t , J t-1 ) is J t , J t-1 The covariance of t )、D(J t-1 ) are J t , J t-1 The variance of .

[0128] Comparative analysis of the duration of structural joint opening and closing: The correlation coefficient r of the spatial correlation curve of the duration of structural joint opening and closing 6 t , r 6 t-1 ,…,r 6 t-k Fit and build a time series forecasting model. 6 t 'Compared to the latest period calculated based on actual measurements 6 t The PauTa criterion is used to compare and determine the difference, so as to quantitatively analyze whether the overall opening and closing degree of the current structural joint surface has changed significantly compared with the previous period.

[0129] 8. The process of micro-deformation analysis between dam layers is as follows:

[0130] When a concrete dam is subjected to loads such as the dam's own weight, reservoir water, and temperature, the concrete material of the dam body will change over time and deform. Due to the material properties of concrete, the magnitude of deformation between the two typical elevations of the dam body is small, and it is usually not easy to detect and monitor. The calculation method of the concrete layer deformation between the two typical elevations of the dam body is studied to analyze the micro-deformation between the concrete layers of the dam body, and then to discover in advance the potential trend changes of concrete materials due to internal and external factors.

[0131] Selection of instruments for monitoring micro-deformation between dam layers: Select static leveling points or artificial leveling points arranged on the dam crest and inspection corridors inside the dam at typical elevations.

[0132] The idea of ​​calculating the micro-deformation between dam layers is to establish the current height difference correlation curve of the two typical elevation monitoring values ​​above and below the dam body, and analyze the micro-deformation change trend of a certain layer of the dam body through the correlation coefficient of the multi-period curve.

[0133] Calculation method of micro-deformation between dam layers:

[0134] According to the current monitoring value E of the typical elevation level points on the upper part of each dam section i Compared with the current monitoring value E of the lower typical elevation level point j Height difference x ij , and the corresponding dam body layer A1(x 11 , f1), A2(x 12 , f1), …, A 1n (x 1n , f1) Establish a spatial correlation model of micro-deformation of a certain layer of the dam body g t (x mn , f m ). where x mn is the vertical settlement deformation monitoring value of the mth layer and nth dam section measuring point, fm It is the corresponding elevation of the measuring point of the mth layer and nth dam section.

[0135] The spatial correlation model g of the past monitoring data of the overall settlement and deformation of the dam body is calculated by formula (9): t-k With g t-(k-1) The Pearson correlation coefficient of the previous two adjacent dam body overall settlement deformation spatial correlation curves was obtained. 7 t .

[0136]

[0137] In the formula, Cov(g t-k , g t-(k-1) ) is g t-k , g t-(k-1 The covariance of t-k )、D(g t-(k-1) ) are g t-k , g t-(k-1) The variance of .

[0138] Comparative analysis of micro-deformation duration between dam layers: correlation coefficient r of spatial correlation curve of micro-deformation duration between dam layers 7 , r 7 t-1 ,…,r 7 t-k , build a time series forecasting model. Through the current forecast value r 7 t 'Compared to the latest period calculated based on actual measurements 7 t A comparison is made and the Rhineda criterion is used to determine the size of the difference, so as to quantitatively analyze whether the micro-deformation between the layers of the dam body in the current period has a larger variation than in the past.

[0139] Based on the above, the deformation analysis includes vertical distribution of dam foundation deformation, overall horizontal displacement of dam body, coordination degree of horizontal displacement of dam body, overall deflection of dam section, overall settlement of dam body, overall opening and closing degree of foundation surface and overall opening and closing degree of structural joints.

[0140] Furthermore, in S103, when determining the seepage analysis results in the analysis results of various indicators of the concrete dam based on the same type of multi-point monitoring data, it is necessary to determine the seepage analysis parameters of the concrete dam based on the seepage monitoring data in the same type of multi-point monitoring data, wherein the seepage analysis parameters include the overall uplift pressure reduction of the dam foundation, the seepage pressure gradient of the structural joints and the seepage amplitude of the key areas, and the overall anti-sliding stability of the dam foundation; based on the seepage analysis parameters, the seepage analysis of the concrete dam is performed to obtain the seepage analysis results.

[0141] The following will introduce the analysis process of the overall uplift pressure reduction of the dam foundation, the seepage pressure gradient of the structural joints, the seepage amplitude in key areas and the overall anti-sliding stability of the dam foundation in the seepage analysis.

[0142] 1. The analysis process of reducing the overall uplift pressure of the dam foundation is as follows:

[0143] The seepage field formed by the upstream and downstream water level difference of high concrete dam will generate vertical uplift pressure. In order to reduce the uplift pressure of the dam foundation and improve the stability of the dam body, 1 to 2 anti-seepage curtains and 1 to 2 drainage curtain lines are laid along the dam foundation to form a multi-layer anti-seepage system. The calculation method of the overall uplift pressure of the dam foundation is studied to analyze the changes in the anti-seepage of the dam foundation.

[0144] When selecting instruments for monitoring dam foundation uplift pressure, select the downstream of the dam foundation anti-seepage curtain and the downstream of the foundation drainage curtain line, and lay pressure measuring tubes along the dam axis.

[0145] The idea of ​​calculating the overall uplift pressure of the dam foundation is as follows: establish the seepage control curtain lines of the dam foundation respectively, including the downstream of the anti-seepage curtain (the first seepage control curtain line), the downstream of the first drainage curtain line (the second seepage control curtain line), and the downstream of the second drainage curtain line (the third seepage control curtain line). The correlation curves of the uplift pressure reduction coefficients of the pressure measuring tubes of each seepage control section are analyzed through the distribution probability of the correlation coefficient of the multi-period curve, so as to evaluate the overall anti-seepage effect of each layer of the seepage control curtain line of the dam foundation.

[0146] Specifically, the calculation method of the overall uplift pressure of the dam foundation is as follows:

[0147] According to the Hydraulic Design Manual, the seepage pressure strength of the dam foundation seepage control curtain line is calculated by formula (10):

[0148] P=α1rH (10)

[0149] In the formula, α1 is the uplift pressure reduction coefficient, which reflects the anti-seepage effect; r is the bulk density of water, which is generally taken as 9.81 kN / m 3 ; H is the water level difference between the upstream and downstream of the arch dam.

[0150] If the seepage intensity measured by a pressure measuring tube behind the dam foundation anti-seepage curtain (the first seepage control curtain line) at time t is P 1n , so the uplift pressure reduction coefficient of the monitoring point is calculated by formula (11):

[0151] α 1n =P 1n / rH (11)

[0152] Then the uplift pressure reduction coefficient at each measuring point of the dam foundation anti-seepage curtain section is α 11 , α 12 , …, α 1n Establishing spatial correlation model t (αmn ), which reflects the overall anti-seepage situation of the dam foundation anti-seepage curtain (the first seepage control curtain line). Among them: α mn It is the reduction coefficient of the piezometric pipe uplift pressure at the nth dam section of the mth layer of seepage control curtain line.

[0153] The Pearson correlation coefficient is used to calculate the uplift pressure reduction coefficient of the dam foundation anti-seepage curtain section. The spatial correlation model of the past duration monitoring data f t-k With f t-k-1 Correlation, the correlation coefficient r of the spatial correlation curve of the dam foundation uplift pressure reduction coefficient of the two previous adjacent periods is obtained 8 t .

[0154] Comparative analysis of the overall uplift pressure reduction effect of the dam foundation over time: Correlation coefficient r of the spatial correlation curve of the dam foundation uplift pressure reduction coefficient 8 t , r 8 t-1 ,…r 8 t-k Fit and build a time series forecasting model. 8 t 'Compared to the latest period calculated based on actual measurements 8 t By comparison, the difference analysis value is determined by using the PauTa criterion or the Grubbs criterion, so as to quantitatively analyze whether the current dam foundation uplift pressure reduction coefficient has a sudden change or a large change amplitude compared with the previous period, so as to evaluate the overall seepage control effect of the multi-layer anti-seepage measures of the dam foundation.

[0155] 2. The analysis process of structural joint seepage pressure gradient is as follows:

[0156] In order to prevent the dam heel from cracking due to tensile stress caused by huge upstream water thrust, high dams, especially high arch dams, are equipped with structural joints at the dam heel near the upstream dam surface. After water storage, the structural joints are in a wet joint working state. The calculation method of the seepage pressure gradient of the structural joints is studied to analyze the seepage pressure condition of the structural joint surface.

[0157] When selecting the structural joint seepage pressure monitoring instrument, select the piezometers installed from upstream to downstream on the inner side of the annular drainage pipe upstream and downstream of the composite waterstop on the joint surface of each dam section, with a total of 2 piezometers installed on each dam section.

[0158] Calculation idea of ​​structural joint seepage pressure gradient: Utilize the piezometers arranged inside the annular drainage pipes upstream and downstream of the composite waterstop of the current structural joint to calculate the seepage pressure gradient of each dam section respectively, and use the seepage pressure gradient value of each dam section to fit the seepage pressure gradient curve along the dam axis to obtain the seepage pressure gradient curve; analyze the change of seepage pressure gradient on the structural joint surface through the distribution probability of the correlation coefficient of the multi-period seepage pressure gradient curve.

[0159] Specifically, the calculation method of the seepage pressure gradient of the structural joint is as follows:

[0160] According to the Hydraulic Design Manual, the seepage pressure strength is calculated according to formula (12):

[0161] P=α2rH (12)

[0162] Where: α2 is the seepage pressure reduction coefficient, reflecting the anti-seepage effect; r is the bulk density of water, generally taken as 9.81 kN / m 3 ; H is the water level difference between the upstream and downstream of the arch dam.

[0163] If the seepage pressure strengths measured by the piezometers before and after the water stop of the structural joints in the nth dam section at time t are respectively P n1 , P n2 , the corresponding seepage reduction coefficient is calculated by equations (13) and (14):

[0164] α n1 =P n1 / rH (13)

[0165] α n2 =P n2 / rH (14)

[0166] The osmotic pressure gradient is calculated by formula (15):

[0167] f n (α)=(α n1 -α n2 ) / d n12 (15)

[0168] Where dn12 is the distance between the two piezometers corresponding to the nth dam section.

[0169] The osmotic pressure gradients f1, f2, ..., f calculated based on the osmotic pressure values ​​measured by the structural joint osmometer n Establish a spatial correlation model between seepage pressure gradient and dam section distribution t (f), which reflects the overall seepage pressure gradient of the structural fracture. n It is the seepage pressure gradient value of the structural joint surface of dam section n.

[0170] The Pearson correlation coefficient is used to calculate the spatial correlation model F of the past monitoring data of the seepage gradient curve of the structural joint along the dam axis. t-k With F t-k-1 Correlation, the correlation coefficient r of the seepage gradient of two adjacent structural fractures in the past 9 t .

[0171] Comparative analysis of structural fracture seepage pressure gradient over time: Correlation coefficient r of structural fracture seepage pressure gradient path9 t , r 9 t-1 ,…r 9 t-k Fit and build a time series forecasting model. 9 t 'Compared to the latest period calculated based on actual measurements 9 t Compare and use the PauTa criterion or Grubbs criterion to determine the difference, and quantitatively analyze whether the current structural fracture seepage gradient has a sudden change or a large amplitude compared with the previous period, so as to evaluate the overall seepage control effect of the structural fracture. 3. The analysis process of seepage amplitude in key areas is as follows:

[0172] A large number of seepage monitoring instruments are installed in the dam body and foundation to monitor the seepage volume in important areas such as the dam body and foundation. When the absolute value of the seepage volume monitoring value is not large, the amplitude of the seepage measuring points is compared to analyze the seepage change law of the arch dam.

[0173] When selecting seepage monitoring instruments for important areas, seepage monitoring instruments are deployed in important locations such as the dam foundation and dam body.

[0174] Calculation ideas for comparison of seepage measurement amplitudes in important areas: In order to avoid the influence of calculation accuracy caused by gross errors in monitoring measurements caused by occasional anomalies at the measuring points themselves, the average seepage rate of the current or nearby consecutive measurements is compared with the average seepage rate of previous consecutive measurements, so as to quantitatively analyze the measurement amplitudes of important seepage measuring points.

[0175] Comparative calculation method of seepage amplitude in important areas:

[0176] Select the previous n periods of seepage monitoring data and calculate the derivative f' of each period of deformation monitoring data at time t t+1 (x1), f' t+2 (x2),…,f' t+n (x n ), and the average seepage rate of period n is calculated by formula (16):

[0177] f' avg (x n )=(f' t+1 (x1)+...f' t+n (x n )) / n (16)

[0178] In the formula, x n f' is the monitoring data of the important seepage measuring point in the past n periods. The past n periods can be a series of periods before the recent period, or can be n periods under similar working conditions last year or in previous years; avg (xn ) is the average seepage rate of the important seepage measuring point in the past n periods.

[0179] According to the properties of an important seepage measurement point and recent working conditions, select the current or recent m-period seepage monitoring data, and calculate the derivative f' of each period of seepage monitoring data at time t t+1 (x1), f' t+2 (x2),…,f' t+m (x m ), and use formula (17) to calculate the average seepage rate of the current period or the most recent m periods:

[0180] f' avg (x m )=(f' t+1 (x1)+...f' t+m (x m )) / m (17)

[0181] In the formula, x m f' is the monitoring value of the current or most recent m periods at the important seepage measuring point; avg (x m ) is the average seepage rate of the important seepage measuring point at the current time or in the most recent m periods.

[0182] The variation of the seepage measurement values ​​of important measuring points at the current or recent period is calculated by formula (18).

[0183] y t =f' avg (x m ) / f' avg (x n ) (18)

[0184] Comparative analysis of the amplitude of important seepage measurement points over time: Quantitative value y of the fitting measurement point change t ,y t-1 , …, y t-k , build a time series forecasting model. Through the current forecast value y t ' and the seepage amplitude y of the new period measuring point t The difference is determined by comparison and the Rhineda criterion is used to quantify whether the seepage at the current measuring point has changed significantly compared with previous periods.

[0185] 4. The overall anti-sliding stability analysis process of the dam foundation is as follows:

[0186] The seepage field formed by the water level difference between upstream and downstream of a high concrete dam will generate vertical uplift pressure, which will reduce the friction between the concrete dam body and the dam foundation rock mass, and reduce the anti-sliding stability of the dam foundation. The calculation method of the anti-sliding stability of the dam foundation is studied to analyze the anti-sliding stability of the dam foundation.

[0187] Selection of dam foundation anti-sliding stability monitoring instruments: Select the downstream of the dam foundation anti-seepage curtain and the downstream of the foundation drainage curtain line, and lay pressure measuring tubes along the dam axis.

[0188] The idea of ​​calculating the overall anti-sliding stability of the dam foundation is as follows: the sliding conditions of the foundation surface of each section of the gravity dam are calculated, the rigid body limit equilibrium method is used, and the anti-sliding stability coefficient of the dam foundation surface is calculated according to the shear strength formula. The correlation curve is formed by the anti-sliding stability coefficient of the dam foundation of each section, and the change of the overall anti-sliding stability coefficient of the dam foundation is analyzed through the distribution of the correlation coefficient of the multi-period curve, so as to evaluate the overall anti-sliding stability of the dam foundation.

[0189] Calculation method of dam foundation anti-sliding stability:

[0190] According to the concrete dam design drawings, the axial width and cross-sectional area of ​​the target dam section are measured to calculate the volume of a certain dam section i; combined with the design data, the bulk density of the dam concrete is found and determined, and the deadweight W of a certain dam section is calculated by formula (19): i :

[0191] W i =αrS (19)

[0192] In the formula, a is the axial width of the dam section, S is the cross-sectional area of ​​the dam section, and r is the concrete density.

[0193] Obtain the current measured water level data upstream and downstream of the dam, and use the difference between the two to calculate the current water thrust P per unit area;

[0194] P=r w H 2 / 2 (20)

[0195] Where H is the upstream and downstream water level difference; r w The density of water is 9.81 kN / m 3 .

[0196] Select the current measured monitoring data of the piezometers arranged along the upstream and downstream directions of the dam section i, downstream of the anti-seepage curtain of the dam foundation and downstream of the first and second drainage curtain lines of the foundation, and calculate the current average uplift pressure U at the bottom of the dam section. ia :

[0197] U ia =(U i1 +U i2 +U i3 ) / 3 (21)

[0198] Where U i1 is the uplift pressure downstream of the anti-seepage curtain of the dam foundation of a certain dam section; U i2 is the uplift pressure downstream of the first drainage curtain line of the dam foundation of a certain dam section i; U i3 It is the uplift pressure downstream of the second drainage curtain line of a dam foundation.

[0199] The shear strength formula of the design specification is used to calculate the current anti-sliding stability coefficient K' of a dam section i si :

[0200]

[0201] Where f' is the shear friction coefficient of the contact surface between the concrete of a certain dam section and the dam foundation; c' is the shear cohesion of the contact surface between the concrete of a certain dam section and the dam foundation; A is the interface area of ​​the contact surface of a certain dam section and the dam foundation.

[0202] Then the anti-sliding stability coefficient of the dam foundation of the n dam sections is K' s1 , K' s2 , …, K' sn Establishing spatial correlation model t (K' sn ), which reflects the overall anti-sliding stability of the dam foundation.

[0203] The Pearson correlation coefficient is used to calculate the overall anti-sliding stability coefficient of the dam foundation. The spatial correlation model of the past duration calculation quantity f t-k With f t-k-1 Correlation, the correlation coefficient r of the spatial correlation curve of the anti-sliding stability coefficient of the dam foundation in the past two adjacent periods is obtained 10 t .

[0204] Comparative analysis of the overall anti-sliding stability of the dam foundation over time: Correlation coefficient r of the spatial correlation curve of the anti-sliding stability coefficient of the dam foundation 10 t , r 10 t-1 ,…r 10 t-k Fit and build a time series forecasting model. 10 t 'Compared to the latest period calculated based on actual measurements 10 t For comparison, the difference analysis value is determined by using the PauTa criterion or the Grubbs criterion, so as to quantitatively analyze whether the overall anti-sliding stability coefficient of the dam foundation in the current period has a sudden change or a large change compared with the previous period, so as to evaluate the overall anti-sliding stability of the dam foundation.

[0205] Based on the above, the analysis of the overall uplift pressure reduction coefficient of the dam foundation, the seepage pressure gradient of the structural joints and the seepage amplitude of key areas in the seepage analysis is completed.

[0206] Furthermore, in S103, when determining the stress-strain analysis results in the analysis results of various indicators of the concrete dam based on the monitoring data of multiple measuring points of the same type, it is necessary to determine the stress-strain analysis parameters of the concrete dam based on the stress-strain monitoring data in the monitoring data of multiple measuring points of the same type, wherein the stress-strain analysis parameters include the vertical distribution of dam foundation stress, the overall horizontal stress of the dam body, the overall stress in the vertical beam direction, and the amplitude of strain measuring points at special parts; based on the stress-strain analysis parameters, perform stress-strain analysis on the concrete dam to obtain the stress-strain analysis results.

[0207] The following will introduce the analysis process of the vertical distribution of dam foundation stress, the overall horizontal stress of the dam body, the overall stress in the vertical beam and the amplitude of strain measuring points in special parts in stress-strain analysis.

[0208] 1. The vertical distribution analysis process of dam foundation stress is as follows:

[0209] When a concrete dam is subjected to loads such as reservoir water, the stress on the dam body will be further transmitted to the rock mass at the dam foundation. Especially for arch dams, most of the stress on the middle and lower parts of the dam body will be transmitted to the dam foundation. Therefore, studying the calculation method of the vertical distribution of stress at the dam foundation will help to understand the working conditions of the rock mass at the dam foundation.

[0210] Dam foundation vertical stress distribution monitoring instrument: Select the dam foundation of representative dam sections such as the riverbed dam section in the middle of the concrete dam, the left and right bank slope dam sections, and the dam foundation with weak geological structures. Combined with the arrangement of the anti-sliding anchor piles in the dam foundation, arrange anchor stress gauges on the anchor rods at equal intervals from the foundation surface to the deep part of the bedrock of the dam foundation.

[0211] Calculation idea of ​​vertical distribution of dam foundation stress: Taking the dam section as a unit, a spatial distribution curve from top to bottom is established, which includes the current measured value of vertical stress and the vertical depth of the dam foundation. The overall change of the vertical stress of the dam foundation is analyzed by the distribution probability of the correlation coefficient of the multi-period curve.

[0212] Specifically, the calculation model of vertical distribution of dam foundation stress is:

[0213] Obtain the measured values ​​σ of the anchor stress gauges arranged at equal intervals on the same anchor 1 1, σ 1 2,…σ 1 n , and establish a spatial correlation model S between the monitoring data and the dam foundation depth distribution t (σ 1 n ), which reflects the vertical distribution of stress in a certain section of the dam foundation. Among them: σ 1 n Represents the current stress monitoring value of strain gauge n.

[0214] Then, the Pearson correlation coefficient is used to calculate the spatial correlation curve S of the vertical stress of any section of the arch dam foundation between the two adjacent monitoring values ​​in the past period. t With S t-1 The correlation coefficient r of the vertical stress distribution curves of each adjacent two times in the past is obtained. 11 t .

[0215] Then, the correlation coefficient r of the vertical stress distribution curve is 11 t , r 11 t-1 ,…,r 11 t-k Fitting is performed to build the first time series prediction model. The predicted value r of the current first time series prediction model is 11 t 'Correlation coefficient r with the latest stress vertical distribution curve calculated based on actual measurements 11 t A comparison is made, and the PauTa criterion is used to determine the size of the difference. A quantitative analysis is performed to determine whether the vertical distribution of the dam foundation stress at the current stage has a sudden change compared with the previous stage or has a variation amplitude exceeding the first preset threshold, so as to evaluate the overall change of the dam foundation vertical stress.

[0216] The first preset threshold value may be set based on the PauTa criterion.

[0217] 2. The overall horizontal stress analysis process of the dam body is as follows:

[0218] When a concrete dam is subjected to loads such as reservoir water, the upper part of the dam body is subjected to a large water thrust. Therefore, studying the calculation method of the overall horizontal stress of the dam at a typical elevation will meet the needs of analyzing the overall distribution of horizontal stress in the concrete dam.

[0219] Dam body overall horizontal stress monitoring instrument: Select typical dam sections of the riverbed and left and right banks, and arrange five-way strain gauge groups or seven-way and nine-way strain gauge groups in sections from the dam top to the horizontal sections corresponding to each typical elevation. At the same time, arrange stress-free gauges. No matter whether it is a five-way, seven-way or nine-way strain gauge group, the horizontal radially arranged strain gauges are selected as the dam body overall horizontal stress monitoring unit.

[0220] The idea of ​​calculating the overall horizontal stress of the dam body is as follows: Taking the horizontal section corresponding to the typical elevation as the unit, a spatial distribution curve of the current measured values ​​of the horizontal stress in the same layer is established, and the overall horizontal stress change of the typical horizontal section is analyzed through the distribution probability of the correlation coefficient of the multi-period curve.

[0221] Specifically, the calculation model of the overall horizontal stress of the dam body is:

[0222] Obtain the uniaxial stress σ converted from the horizontal strain gauge measurement value of the same plane strain gauge group 2 1, σ 2 2,…σ 2 n , and establish a spatial correlation model A between the monitoring data and the dam section distribution t (σ 2 n ), which reflects the overall stress of a typical horizontal section. Among them: σ 2 n Represents the current stress monitoring value of strain gauge n.

[0223] Then, the Pearson correlation coefficient is used to calculate the spatial correlation curve A of the two adjacent monitoring values ​​in the horizontal period of any layer of the arch dam. t With A t-1 The correlation coefficient r of the spatial correlation curve of the overall horizontal stress between the two adjacent periods is obtained. 12 t .

[0224] Then, the correlation coefficient r of the horizontal stress spatial correlation curve is 12 t , r 12 t-1 ,…,r 12 t-k Fitting is performed to build the second time series prediction model. The predicted value r of the current second time series prediction model is 12 t 'The correlation coefficient r with the overall horizontal stress spatial correlation curve calculated based on the latest actual measurement 12 t A comparison is made and the PauTa criterion is used to determine the size of the difference. A quantitative analysis is performed to determine whether the overall horizontal stress of the current period has a sudden change compared with the previous period or has a change amplitude exceeding the second preset threshold, so as to evaluate the change in the overall horizontal stress of the dam body.

[0225] The second preset threshold may be set based on the PauTa criterion.

[0226] 3. The overall stress analysis process of the vertical beam is as follows:

[0227] According to the stress and load distribution of the dam structure, when the concrete dam is subjected to loads such as reservoir water, the middle and lower parts of the dam body are mainly subjected to stress in the beam direction. Therefore, the calculation method of the overall stress in the vertical beam direction of the arch dam is studied to realize the overall stress analysis of the vertical beam section of the typical dam section.

[0228] Vertical beam stress monitoring instrument: Select representative dam sections such as the riverbed section in the middle of the concrete dam, the left and right bank slope sections, and the weak geological structure of the dam foundation. From the dam top to each key horizontal section, five-way strain gauge groups, seven-way strain gauge groups, and nine-way strain gauge groups are arranged in sections. At the same time, stress-free gauges are arranged. No matter whether it is a five-way, seven-way, or nine-way strain gauge group, the vertically arranged strain gauges are selected as the vertical beam stress monitoring unit.

[0229] Calculation idea of ​​vertical beam stress: Taking the vertical section of the dam section as a unit, establish a spatial distribution curve from top to bottom that includes the current measured values ​​of vertical stress and elevation, and analyze the overall stress change of the beam section through the distribution probability of the correlation coefficient of the multi-period curve.

[0230] Specifically, the overall stress calculation model of the beam is:

[0231] Obtain the uniaxial stress σ converted from the vertical strain gauge measurement value of the same section strain gauge group 3 1, σ 3 2,…σ 3 n , and establish a spatial correlation model B between monitoring quantity and elevation distribution t (σ 3 n ), which reflects the overall stress of a beam section. Among them: σ 3 n Represents the current stress monitoring value of strain gauge n.

[0232] Then, the Pearson correlation coefficient is used to calculate the spatial correlation curve B of the two adjacent monitoring values ​​in the vertical beam section of any dam section. t With B t-1 The correlation coefficient r of the spatial correlation curve of the overall stress of each adjacent vertical beam is obtained. 13 t .

[0233] Then, the correlation coefficient r of the vertical beam overall stress spatial correlation curve is 13 t , r 13 t-1 ,…,r 13 t-k Perform fitting and build the third time series prediction model.

[0234] The predicted value r of the current third time series forecasting model 13 t 'The correlation coefficient r with the vertical beam overall stress spatial correlation curve calculated based on the latest actual measurement 13 tThe PauTa criterion is used to judge the difference and quantitatively analyze whether the overall stress in the beam direction has a sudden change compared with the previous period or has a change amplitude exceeding the third preset threshold, so as to evaluate the overall stress change of the beam section.

[0235] The third preset threshold value may be set based on the PauTa criterion.

[0236] 4. The amplitude analysis process of strain measurement points in special parts is as follows:

[0237] A large number of strain monitoring instruments are installed in the dam body and foundation of the concrete dam to monitor and discover stress concentration in the dam body from a microscopic perspective. It is expected that the parts prone to local cracking will be monitored and analyzed. When the absolute value of the strain monitoring is not large, the change rate of the strain measuring points will be studied and analyzed. This will help to timely discover and grasp the strain change law of special parts of the dam body, and provide support for the early detection of microscopic defects in the dam body.

[0238] Important strain monitoring instruments: refers to strain gauges deployed at stress concentration locations such as the heel and toe of the dam, where local cracks are expected to occur or where concrete defects have been found.

[0239] Calculation idea of ​​strain rate at measuring points: In order to prevent gross errors in monitoring values ​​caused by occasional abnormalities at the measuring points themselves, the ratio of the average value of the current or recent consecutive measurements to the average value of the previous consecutive measurements is used to analyze the rate of change of the values ​​at important strain measuring points.

[0240] Specifically, the strain rate calculation model of the measuring point is:

[0241] Get the historical average stress monitoring value corresponding to the preset key parts. Specifically, select the historical key parts strain monitoring values ​​ε1, ε2, ...ε n , calculate the average monitoring amount ε for n periods navg . Among them: ε n It indicates the monitoring data of the strain measuring points at key parts in the past n periods. The past n periods can be multiple consecutive periods before the recent period, or can be n periods under similar working conditions last year or in previous years; ε navg It represents the average value of strain measurement points in key areas monitored over the past n periods.

[0242] Then, the stress monitoring value of each period in the current period corresponding to the key part is obtained. Assuming that the current period includes m periods, the strain monitoring values ​​ε1', ε2', ... ε corresponding to the key part in period m are selected. m ', calculate the average monitoring amount ε of period m mavg '. Among them: ε m Indicates the current or recent mth monitoring value of the strain measurement point at the key part; ε mavg' represents the average value of the strain measurement points in key areas monitored in the last m periods.

[0243] Then, multiple strain rates corresponding to the key parts are obtained based on the average stress monitoring value of the historical period and the stress monitoring value of each period in the current period. t =ε m ' / ε navg Calculate the current strain rate of important measuring points, or use y n =ε mavg ' / ε navg Calculate the strain rate of important measuring points in multiple periods.

[0244] Then, for the strain rate y t ,y t-1 ,…y t-k Fitting is performed to construct the fourth time series prediction model, and then the predicted value y of the fourth time series prediction model is obtained. t ' and the strain rate y calculated according to the latest measurement t The PauTa criterion is used to judge the difference and quantitatively analyze whether the current strain rate has a sudden change compared with the previous period or has a variation amplitude exceeding the fourth preset threshold, so as to evaluate the change rate of the measured values ​​of important strain measuring points.

[0245] The fourth preset threshold may be set based on the PauTa criterion.

[0246] Based on the above, the vertical distribution of dam foundation stress, the overall horizontal stress of the dam body, the overall stress in the vertical beam and the amplitude of strain measuring points in special parts in the stress-strain analysis are completed.

[0247] Furthermore, in S103, when determining the temperature analysis results in the analysis results of various indicators of the concrete dam based on the same type of multi-measurement point monitoring data, it is necessary to determine the temperature analysis parameters of the concrete dam based on the temperature monitoring data in the same type of multi-measurement point monitoring data, wherein the temperature analysis parameters include the temperature distribution of the dam foundation and the temperature gradient of the dam body; based on the same temperature analysis parameters, perform temperature analysis on the concrete dam to obtain the temperature analysis results.

[0248] The following will introduce the analysis process of dam foundation temperature distribution and dam body temperature gradient in temperature analysis.

[0249] 1. The dam foundation temperature distribution analysis process is as follows:

[0250] The temperature of the deep and shallow parts of the rock mass at the dam foundation interacts with the temperature of the dam body, and further affects the operating conditions of the dam body through temperature loads. Therefore, using thermometers arranged vertically on the dam foundation to monitor the base temperature of the dam will help to understand the temperature distribution of the dam foundation.

[0251] Dam foundation temperature monitoring instrument: consistent with the representative temperature monitoring dam section of the dam body, thermometers are arranged vertically from top to bottom from the interface between the dam body and the dam foundation to the deep rock mass of the dam foundation to monitor the changes in the dam foundation temperature during operation.

[0252] Calculation idea of ​​dam foundation temperature: Taking the vertical section of the dam section as a unit, establish a spatial distribution curve from top to bottom that includes the current measured values ​​of vertical temperature and elevation, and analyze the overall temperature change of the vertical section of the dam foundation through the distribution probability of the correlation coefficient of the multi-period curve.

[0253] Specifically, the dam foundation temperature calculation model is:

[0254] According to the thermometer values ​​T1, T2, ... T of the same section n Establish a spatial correlation model between monitoring data and elevation distribution t , which reflects the vertical temperature distribution of a temperature monitoring section of the dam foundation. Where: T n —Thermometer n Current temperature monitoring value.

[0255] The Pearson correlation coefficient is used to calculate the spatial correlation curve T of the past monitoring data of a temperature monitoring section of the dam foundation. t With T t-1 Correlation, the correlation coefficient r of the spatial correlation curve of a temperature monitoring section of the adjacent two dam foundations is obtained 14 t .

[0256] Comparative analysis of dam foundation temperature calculation duration: Correlation coefficient r of spatial correlation curve of a temperature monitoring section of the dam foundation 14 t , r 14 t-1 ,…,r 14 t-k Fit and build a time series forecasting model. 14 t 'Compared to the latest period calculated based on actual measurements 14 t The PauTa criterion is used to compare and determine the difference, so as to quantitatively analyze whether the current dam foundation temperature has changed significantly compared with previous periods.

[0257] 2. The dam body temperature gradient analysis process is as follows:

[0258] During the operation period of a concrete dam, in addition to upstream water loads, temperature loads are also the main influencing factors of the dam's operating conditions. Therefore, using thermometers originally used for concrete temperature control during the dam construction period to conduct long-term operating temperature monitoring of the dam will help to understand the temperature distribution of the arch dam.

[0259] Dam temperature monitoring instrument: In each dam section, along a measuring line on the same plane, thermometers with the same spacing are arranged from upstream to downstream, with a total of 2 to 7 thermometers arranged according to the width of the dam body.

[0260] The idea of ​​calculating the temperature gradient of the dam body is as follows: thermometers with the same spacing are arranged from upstream to downstream along a measuring line on the same plane of each dam section, and the temperature gradient of the current elevation of each dam section is calculated respectively. The temperature gradient curve is fitted along the dam axis using the temperature gradient value of each dam section to obtain the temperature gradient curve; the temperature gradient change of the current elevation of the dam section is analyzed by the distribution probability of the correlation coefficient of the temperature gradient curve over multiple periods.

[0261] Specifically, the dam body temperature gradient calculation model:

[0262] If the measured temperatures of the thermometers at a certain elevation m at the nth dam section from upstream to downstream at time t are T nm1 、T nm2 ……T nmi , temperature gradient g(T)=(T nm1 -T nmi ) / (i-1)d nm , d nm It is the average distance between two plane thermometers at a certain elevation m in the nth dam section.

[0263] The temperature gradient g calculated based on the temperature values ​​measured by the thermometer at the same elevation 1m , g 2m , ... g nm Establish the spatial correlation model G between temperature gradient and dam section distribution t (g), which reflects the overall temperature gradient of a certain elevation plane of the dam body. nm —Temperature gradient value of a plane at a certain elevation m in dam section n.

[0264] The Pearson correlation coefficient is used to calculate the spatial correlation model G of the temperature gradient curve of the past monitoring data along the dam axis at a certain elevation m t-k With G t-k-1 Correlation, to obtain the correlation coefficient r of the two adjacent temperature gradients in the past 15 t .

[0265] Comparative analysis of dam body temperature gradient over time: Correlation coefficient r of dam body temperature gradient 15 t , r 15 t-1 ,…,r 15 tk is fitted to build a time series forecasting model. 15 t 'Compared to the latest period calculated based on actual measurements15 t By comparing the results, the PauTa criterion or Grubbs criterion is used to determine the difference, and quantitatively analyze whether the current temperature gradient has a sudden change or a large amplitude compared with the previous period, so as to evaluate the temperature distribution of the dam body.

[0266] Furthermore, in S104, when determining any one of the deformation scoring results, seepage scoring results, and stress-strain scoring results of the concrete dam based on the multi-type and multi-measurement point monitoring data, it is necessary to divide the key operating parts of the concrete dam into regions, and obtain the time series measurement data of different types of monitoring instruments in a certain partition from the multi-type and multi-measurement point monitoring data; establish a graph structure for the time series measurement data in the time and variable dimensions respectively, and obtain a time feature graph and a variable feature graph; input the time feature graph and the variable feature graph into the time graph attention network and the variable graph attention network respectively, and obtain a time attention matrix and a variable attention matrix; concatenate the time series measurement data, the time attention matrix, and the variable attention matrix and input them into the gated convolutional network to obtain the target feature; calculate the anomaly score according to the target feature, compare the anomaly score with the preset indicator threshold, and obtain the operation score result of the corresponding indicator of the concrete dam.

[0267] The calculation of the anomaly score according to the target feature includes: inputting the target feature into a prediction module and a reconstruction module to obtain a prediction value and a reconstruction probability; and calculating the anomaly score according to the prediction value and the reconstruction probability.

[0268] Among them, the prediction module is a multi-layer perceptron.

[0269] Among them, the reconstruction module includes a discriminator and an autoencoder.

[0270] S104 will be specifically introduced below in conjunction with the model diagram.

[0271] Figure 3 This is a schematic diagram of a framework of a multi-index graph attention network model for key parts of a concrete dam shown in this application, such as Figure 3 As shown in the figure, the graph structure is first constructed using the multivariate time series measurement data of multiple types of monitoring instruments. The graph structure specifically includes a graph structure learning layer, a graph attention layer, and a joint network layer. Among them, the graph structure learning layer uses an embedding vector to characterize the inherent attributes of each node (i.e., monitoring point), and constructs graph structures on the time dimension and variable dimension respectively, capturing the time dimension and variable dimension dependencies of multivariate time series data. The graph attention layer further learns and characterizes this relationship, and captures the linear and nonlinear dependencies within and between indicators through an improved attention mechanism. Finally, the joint network layer combines the advantages of the prediction model and the reconstruction model, uses the prediction error and reconstruction probability as the anomaly score to detect anomalies, and locates the indicators that are most likely to cause abnormal dam operation through the attention score.

[0272] The monitoring data collected by multiple types of monitoring instruments for a certain monitoring effect in a certain area is multivariate time series data, which consists of n groups of univariate time series data, which are interrelated through linear and nonlinear relationships. When an abnormality occurs in the operation of a certain part of the concrete dam, it is manifested as changes in all or part of the monitoring data in the time dimension. Therefore, for the monitoring data in the same time window, two graph structures are used to model the correlation between the time and variable dimensions respectively. The details are as follows:

[0273] (1) Variable feature graph. Before constructing the variable feature graph, it is necessary to randomly initialize a representation vector for each variable to reduce the accuracy loss of the model due to different data types and value ranges. In the latent representation space composed of embedded vectors, vectors with closer distances are more similar, and the correlation between their corresponding variables (i.e., the correlation between different types of concrete dam safety monitoring instruments) is stronger. The graph attention network requires data with an explicit graph structure as input, so the N variables in the input time window are used as N nodes to construct the graph structure. First, the similarity relationship between each node is calculated based on the initial representation vector of the node, and then the largest node pairs are selected and connected with edges to obtain a sparse directed graph structure. The specific steps are as follows:

[0274] First, we set an embedding vector for each variable Where d represents the dimension of the vector, and different values ​​are set according to the actual situation. The initial value of the vector is given randomly, and the specific value is adjusted continuously through back propagation during the training process of the model. The variable correlation can be calculated from the embedded vector. For variables i and j, the correlation is calculated by (23):

[0275]

[0276] Where, is f(v i ,v j ) Any similarity calculation method is used. If a directed graph is to be constructed, an asymmetric similarity calculation method is required, and cosine similarity is used to calculate the similarity. After calculating the similarity of all variables in pairs, for any variable i, select the first K neighbor variables j with the greatest similarity to it, and connect i and j with an edge in the spatial graph, corresponding to the column where i is located in the spatial adjacency matrix A, expressed as formula (24):

[0277]

[0278] The value of K can be given by the user according to the specific situation to adjust the sparsity of the graph. In the case of prior information about the graph structure, the adjacency matrix A can be directly given by the user. In some special scenarios, the adjacency matrix can be set to a matrix with all values ​​1, that is, to build a fully connected graph.

[0279] (2) Time feature graph. In order to explicitly model the temporal information between data, a time feature graph within the input time window is constructed using steps similar to those used to construct the variable feature graph. The time graph differs from the variable graph in the setting and composition of the embedding vector. In the variable graph, the embedding vector is used to reduce the impact of different variable data types and value ranges on model accuracy. In the time feature graph, in addition to assuming a corresponding embedding vector for each time point in the sliding window, positional encoding needs to be introduced to represent the temporal position difference of the data. The embedding vector of each node is u i The initial value of is also randomly given, and then the final representation is obtained through training and learning.

[0280] For the time series in the input sliding window, a position code is given to the vector at any timestamp j. Calculated by formula (25):

[0281]

[0282] In the formula, d is the dimension of the position encoding, which needs to be the same as the variable dimension of the current position, that is, d = N, and d also needs to be a multiple of 2. Unlike the above variable feature graph in which each variable is regarded as a node, in order to construct an explicit time graph structure, the data at each moment needs to be regarded as a node in the graph. The specific form of the position encoding is calculated by formula (26):

[0283]

[0284] After assigning position encoding to each time point, the time graph is constructed by calculating the similarity between the embedding vectors of different time points. Calculated by formula (27):

[0285]

[0286] The pairwise similarity between different time points indicates the similarity between different timestamps. When constructing the time graph, similar to the above construction of the variable graph, for any timestamp data i, select the K neighboring time points j with the greatest similarity to it, connect them with edges, and the constructed time adjacency matrix is ​​calculated by formula (28):

[0287]

[0288] In the formula, The value of K is also specified by the user and is generally the same as the sparsity of the spatial adjacency matrix.

[0289] After completing the construction of the graph structure data, the graph attention network is used to learn the information of the time dimension and the variable dimension. After learning, each node in the graph structure will get a final representation vector, which contains the current node information and its adjacent node information. The graph attention layer is composed of a parallel variable graph attention network and a time graph attention network, which are used to simultaneously capture the dependencies of different dimensions in the data. At the same time, by improving the attention mechanism in the variable graph attention layer, the correlation of indicators in the data can be explicitly captured. The details are as follows:

[0290] (1) Variable Graph Attention Network (F-GAT). The constructed variable feature graph is used as the input of F-GAT, and the information in the graph is further mined through the attention mechanism. First, assume that the feature representation of F-GAT in the first layer is H l , the initial input formula (29):

[0291]

[0292] In the formula, is the input sequence of length ω at timestamp t, is the learnable transformation matrix of the input data, || is the concatenation operation, and V is the matrix composed of node representation vectors.

[0293] The architecture of F-GAT is as follows Figure 4 As shown in Figure 1, it consists of three modules: multi-head attention, intra-indicator attention, and inter-indicator attention. The multi-head attention module mainly models the variable dependency between multivariate time series, while the intra-indicator and inter-indicator attention are used to capture the indicator correlation between different time series. Specifically, the intra-indicator correlation is the correlation of all measuring points under the same type of monitoring instruments. For example, the concrete dam deformation monitoring project includes multiple types of monitoring instruments such as vertical lines, surface deformation observations, static levels, and joint meters. The scope of the indicators refers to all measuring points included in the vertical line type of monitoring instrument. The inter-indicator correlation is the correlation of different types of monitoring instruments. For example, the concrete dam deformation monitoring project includes multiple types of monitoring instruments such as vertical lines, surface deformation observations, static levels, and joint meters. The scope of the indicators is multiple types of monitoring instruments such as vertical line measuring points, surface deformation observation points, static leveling points, and joint meters.

[0294] The multi-head attention module updates the feature representation of each node by aggregating the neighbor node information of the target node, and is calculated by formula (30):

[0295]

[0296] In the formula, is the feature representation of node i at the l+1th layer, || is the concatenation operation, S represents the number of attention heads, represents the attention score of the sth attention head of node i and node j in layer l, is the learnable weight matrix of the sth attention head in the sth layer, is the feature representation of node j at the,th layer, is the set of neighbor nodes of node i in the adjacency matrix A representing the variable feature graph. The attention score is calculated by equations (31), (32) and (33):

[0297]

[0298] In the formula, a T is a learnable bias vector, || is a concatenation operation, and LeakyReLU is a nonlinear activation function.

[0299] Traditional graph attention networks fail to consider the correlation between indicators in multivariate time series, and therefore lose important information in some variable dimensions. Neighbor nodes with different dependencies have different influences on the central node. This section improves the effectiveness of the model in modeling inter-sequence variable dependencies by adding two relational attention modules, namely, intra-indicator and inter-indicator attention. The adjacency matrix of the intra-indicator attention graph and the inter-indicator attention graph is defined by equations (34) and (35):

[0300]

[0301] In the formula, and is a candidate set, i.e. represents the node with the same index as node i, indicates a node with different monitoring indicators from node i. It should be noted that when or When , you need to use the TopK operation to select the indexes of the first K largest cosine similarities to build the adjacency matrix.

[0302] Then, two relational attention modules are used to explicitly capture the multi-metric correlations between different time series. The features of the intra-metric attention module are calculated using equations (36), (37), and (38):

[0303]

[0304] In the formula, is the feature representation of node i at layer l+1, is the set of neighbor nodes within the index of node i, is the attention score of node i and node j in layer l, and is the weight matrix of the lth layer, is the bias vector of the ,th layer. Similarly, the feature representation of the inter-indicator attention module can be calculated is the set of neighbor nodes between the indices of node i. The final output of the variable graph attention layer It is obtained by concatenating the inputs of the three attention modules and is calculated using equations (39) and (40):

[0305]

[0306] In the formula, is the final representation of node i at layer l+1, represents the weight matrix of the l+1th layer, represents the bias vector of the l+1th layer, || is the concatenation operation, By taking the intermediate features of the l+1th layer and Splicing obtained.

[0307] (2) Temporal Graph Attention Network (T-GAT). Assume that Z l is the feature representation of T-GAT at the layer, and the initial input of T-GAT is in is the learnable transformation matrix of the input data, and U is the matrix composed of node representation vectors. The time graph attention layer takes the above-constructed time graph structure as input, combines position encoding, and uses a multi-head attention module to aggregate the information of neighboring nodes to update the feature representation of each time point, which is calculated by formula (41):

[0308]

[0309] In the formula, is the feature representation of node i at the l+1th layer, || is the concatenation operation, S represents the number of attention heads, is the set of neighbor nodes of node i in the adjacency matrix A′ representing the time graph above, is the attention score of node i and node j at the sth attention head in layer l, is the weight matrix of the sth attention head in the lth layer, is the feature representation of node j at the,th layer, The calculation steps are the same as above The calculation steps are similar, and are specifically calculated by equations (42), (43) and (44):

[0310]

[0311] In the formula, represents the attention score of node i and node j in the time graph at the sth attention head in the sth layer, is the learnable weight matrix of the sth attention head in the lth layer of the temporal graph attention module, a T is a learnable bias vector, || is a concatenation operation, and LeakyReLU is a nonlinear activation function.

[0312] The output of the variable graph attention network is an N×ω-dimensional matrix, where a row of the matrix represents the relationship between a node in the variable feature graph and its neighboring nodes captured by the graph attention network. Similarly, the output of the temporal graph attention network is an ω×N-dimensional matrix. The outputs of the two graph attention layers are concatenated with the original time series data to form an ω×3N-dimensional matrix, where a row of the matrix represents a 3N-dimensional feature vector of a timestamp within the input time window. Finally, the ω×3N-dimensional matrix is ​​used as the input of the gated convolutional network (GRU). As a variant of the recurrent convolutional network, GRU can well capture the sequence pattern information in the data and obtain the target features.

[0313] In order to take advantage of both reconstruction-based and prediction-based models, the loss function of MTS-GAT contains two objectives: capturing the distribution of the entire input data in the reconstruction module and accurately predicting the value at the next timestamp in the prediction module. is the output of XχXGRU. The loss function of joint optimization is defined by equation (45):

[0314]

[0315] in, is the loss function of the reconstruction module, is the loss function of the prediction module, and γ1 is the hyperparameter for balancing the weights of the two modules.

[0316] Prediction module usage To predict the observation value at the next timestamp, this section uses a multi-layer perceptron (MLP) as the prediction module, and the loss function is defined by formula (46):

[0317]

[0318] In the formula, x i，t+1 is the measured value of the i-th time series at t+1, is the predicted value of the i-th sequence at time t+1.

[0319] The reconstruction module is to learn the reconstruction probability of the input data. In order to enhance the robustness of the model, two discriminators D E(·) and DD(·) for the autoencoder G A Adversarial training is performed as a reconstruction-based model. G A The encoder G E (·) and decoder G D (·) can be regarded as two generators. Figure 5 As shown, for a given input z is the potential representation of the autoencoder, p(z) represents the prior distribution of z, and the posterior distribution q(z) generated by the autoencoder in the latent space is calculated by formula (47):

[0320]

[0321] In the formula, represents the encoding distribution, Denotes the input data distribution. Adversarial network D E (·) is used to adjust the posterior distribution q(z) to satisfy the prior distribution p(z), that is, to maximize the loss function Calculated by formula (48):

[0322]

[0323] The corresponding generator G E (·) Mixed D E (·), which minimizes the loss function Calculated by formula (49):

[0324]

[0325] Similarly, adversarial network D D (·) Avoid overfitting of the model by increasing the difference between the input data and the reconstructed data, that is, maximizing the loss function Calculated by formula (50):

[0326]

[0327] The corresponding generator GD(·) needs to minimize the loss function Calculate by formula (51):

[0328]

[0329] Reconstructed representation of input data Finally, use and As adversarial regularization to ensure the robustness of the reconstruction model, the loss function of the reconstruction model is defined by Equation (52) and Equation (53):

[0330]

[0331] In the formula, represents the reconstruction loss.

[0332] For the i-th univariate time series, at any timestamp t, the forecast module generates a forecast value The reconstruction module generates the reconstruction probability p i The final anomaly score for each timestamp is obtained by balancing the weights of these two modules through equation (54):

[0333]

[0334] In the formula, x i is the measured value, and γ2 is the hyperparameter for balancing the two modules. In the anomaly detection stage, when the anomaly score of a timestamp is greater than the given anomaly threshold, the timestamp is marked as an "abnormal timestamp", otherwise it is a "normal timestamp". The selection of the anomaly threshold is performed on the validation set using the peaks-over-threshold (POT) algorithm.

[0335] Figure 6 Schematic diagram of a device for judging the operating performance of a concrete dam shown in the present application. Figure 6 As shown, the concrete dam operation performance evaluation device 600 includes a data acquisition module 601, a first determination module 602, a second determination module 603 and an evaluation and analysis module 604, wherein:

[0336] The data acquisition module 601 is used to arrange a monitoring system on the concrete dam to collect the same type of multi-point monitoring data and multi-type multi-point monitoring data of the concrete dam, wherein the same type of multi-point monitoring data includes deformation monitoring data, seepage monitoring data, stress strain monitoring data, and temperature monitoring data.

[0337] The first determination module 602 is used to determine various index analysis results of the concrete dam based on the same type of multi-point monitoring data, wherein the various index analysis results include deformation analysis results, seepage analysis results, stress-strain analysis results, and temperature analysis results.

[0338] The second determination module 603 is used to determine the operation score results of various indicators of the concrete dam based on the multi-type and multi-point monitoring data, wherein the operation score results of various indicators include deformation score results, seepage score results, and stress-strain score results.

[0339] The evaluation and analysis module 604 is used to perform an overall evaluation and analysis on the concrete dam based on the analysis results of each indicator and the operation scoring results of each indicator.

[0340] In order to implement the above-mentioned embodiments, the embodiments of the present application also propose an electronic device, which includes: a processor and a memory communicatively connected to the processor, the memory storing instructions executable by at least one processor, and the instructions being executed by at least one processor to implement the method for formulating evaluation indicators based on a hierarchical architecture for online evaluation of the operating performance of a concrete dam as shown in the above-mentioned embodiments.

[0341] In order to implement the above-mentioned embodiments, the embodiments of the present application also propose a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable the computer to implement the method for formulating evaluation indicators based on the hierarchical architecture for online evaluation of the operating performance of concrete dams as shown in the above-mentioned embodiments.

[0342] In order to implement the above embodiment, the embodiment of the present application also proposes a computer program product, including a computer program, which, when executed by a processor, implements the method for formulating evaluation indicators based on the hierarchical architecture for online evaluation of the operating performance of concrete dams as shown in the above embodiment.

[0343] In the description of the present application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0344] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0345] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0346] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for formulating evaluation indicators based on a hierarchical framework for online evaluation of concrete dam operating performance, characterized in that: include: Obtaining an online evaluation hierarchical model for the operation performance of a concrete dam, wherein the online evaluation hierarchical model includes an overall engineering layer, a key part layer, a monitoring project layer, an evaluation index layer, a diagnostic method layer, and a monitoring data layer; The overall engineering layer is the target of online evaluation of the concrete dam operation performance and is used to conduct a comprehensive evaluation of the overall operation performance of the concrete dam; The key parts layer is the focus of the online evaluation of the operating performance of the concrete dam. The key parts are determined based on the calculation analysis in the design stage of the concrete dam, or the cluster analysis of the long-term operation monitoring data. The key parts layer analyzes the special evaluation results of the deformation, seepage, stress-strain, and temperature monitoring effect derived from the monitoring project layer, and the key parts are distributed in the upper dam body, middle dam body, riverbed dam section dam heel, riverbed dam section dam toe, dam foundation, and the defect distribution area of ​​the long-term focus of the project operation; The monitoring project layer is the associated medium for online evaluation of the operating performance of concrete dams. According to the currently implemented concrete dam design and safety monitoring specifications, the deformation, seepage, stress-strain and temperature monitoring effect quantities are comprehensively analyzed; The evaluation index layer is the process center of the online evaluation of the operation performance of the concrete dam. Four types of evaluation indicators, namely deformation, seepage, stress-strain and temperature, are set corresponding to the deformation, seepage, stress-strain and temperature monitoring effect quantities of the operation safety of the concrete dam in the monitoring project layer. Among them, under each type of evaluation indicator, there are further divided into two indicator formulation methods: one is to calculate the evaluation indicators of the same type of multi-measurement point combination according to the structural characteristics; the other is a multi-type multi-measurement point partition comprehensive evaluation indicator based on data-driven; The diagnostic method layer is the core of the online evaluation of the operating performance of concrete dams. Based on the periodic effects of reservoir water and temperature loads during the operation of concrete dams and the changes in the internal structure and materials of concrete dams, the monitoring effect quantity analysis of multiple points of the same type or multiple types of points is carried out according to the combined calculation of multiple measuring points of the same type and the comprehensive evaluation indicators of multiple measuring points of multiple types proposed in the evaluation index layer. The monitoring data layer is the information base for online evaluation of the operating status of concrete dams, including engineering safety monitoring, inspection defect data and geophysical detection results data; When the overall evaluation and analysis of the concrete dam is performed based on the online evaluation hierarchical model, the following steps are included: Collecting the same type of multi-point monitoring data and multi-type multi-point monitoring data of the concrete dam, wherein the same type of multi-point monitoring data includes deformation monitoring data, seepage monitoring data, stress strain monitoring data, and temperature monitoring data; Based on the same type of multi-measurement point monitoring data, determining various index analysis results of the concrete dam, wherein the various index analysis results include deformation analysis results, seepage analysis results, stress-strain analysis results, and temperature analysis results; Based on the multi-type and multi-point monitoring data, determining the operation scoring results of various indicators of the concrete dam, wherein the operation scoring results of various indicators include deformation scoring results, seepage scoring results, and stress-strain scoring results; Performing an overall evaluation and analysis on the concrete dam based on the analysis results of each indicator and the operation scoring results of each indicator; Wherein, the operation score results of various indicators of the concrete dam are determined based on the multi-type and multi-point monitoring data, including: Divide the key operating parts of the concrete dam into regions, and obtain the time series measurement data of different types of monitoring instruments in a certain region from the multi-type and multi-measurement point monitoring data; Establishing graph structures for the time series measurement data in time and variable dimensions respectively, to obtain a time characteristic graph and a variable characteristic graph; Inputting the time feature graph and the variable feature graph into a time graph attention network and a variable graph attention network respectively to obtain a time attention matrix and a variable attention matrix; splicing the time series measurement data, the time attention matrix and the variable attention matrix together and inputting them into a gated convolutional network to obtain target features; Calculating an anomaly score according to the target feature, comparing the anomaly score with a preset index threshold, and obtaining an operation score result of the corresponding index of the concrete dam; Calculating the anomaly score according to the target feature includes: Inputting the target feature into a prediction module and a reconstruction module to obtain a prediction value and a reconstruction probability; The anomaly score is calculated based on the predicted value and the reconstruction probability.

2. The method according to claim 1, characterized in that The step of determining the deformation analysis results in the various index analysis results of the concrete dam based on the same type of multi-measurement point monitoring data includes: Based on the deformation monitoring data in the same type of multi-point monitoring data, the deformation analysis parameters of the concrete dam are determined, wherein the deformation analysis parameters include vertical distribution of dam foundation deformation, overall horizontal displacement of the dam body, coordination degree of horizontal displacement of the dam body, overall deflection of the dam section, overall settlement of the dam body, micro-deformation between dam body layers, overall opening degree of the foundation surface, and overall opening degree of the structural joint; Based on the deformation analysis parameters, deformation analysis is performed on the concrete dam to obtain the deformation analysis results.

3. The method according to claim 2, characterized in that The method of determining the seepage analysis results in the various indicator analysis results of the concrete dam based on the same type of multi-measurement point monitoring data includes: Based on the seepage monitoring data in the same type of multi-point monitoring data, the seepage analysis parameters of the concrete dam are determined, wherein the seepage analysis parameters include the overall uplift pressure reduction coefficient of the dam foundation, the anti-sliding stability of the dam foundation, the seepage pressure gradient of the structural joints, and the seepage amplitude of the key areas; Based on the seepage analysis parameters, a seepage analysis is performed on the concrete dam to obtain the seepage analysis result.

4. The method according to claim 3, characterized in that The step of determining the stress-strain analysis results in the analysis results of various indicators of the concrete dam based on the same type of multi-measurement point monitoring data includes: Based on the stress and strain monitoring data in the same type of multi-measurement point monitoring data, the stress and strain analysis parameters of the concrete dam are determined, wherein the stress and strain analysis parameters include the vertical distribution of dam foundation stress, the overall horizontal stress of the dam body, the overall stress in the vertical beam direction, and the amplitude of strain measurement points at special locations; Based on the stress-strain analysis parameters, a stress-strain analysis is performed on the concrete dam to obtain the stress-strain analysis result.

5. The method according to claim 4, characterized in that The step of determining the temperature analysis results in the analysis results of various indicators of the concrete dam based on the same type of multi-measurement point monitoring data includes: Determine the temperature analysis parameters of the concrete dam based on the temperature monitoring data in the same type of multi-measurement point monitoring data, wherein the temperature analysis parameters include the dam foundation temperature distribution and the dam body temperature gradient; Based on the same temperature analysis parameters, a temperature analysis is performed on the concrete dam to obtain the temperature analysis result.

6. The method according to claim 1, characterized in that The prediction module is a multi-layer perceptron.

7. An online evaluation device for the operating performance of a concrete dam, characterized in that: The device implements the method according to claim 1, and the device comprises: A data acquisition module, used to collect the same type of multi-measurement point monitoring data and multi-type multi-measurement point monitoring data of the concrete dam, wherein the same type of multi-measurement point monitoring data includes deformation monitoring data, seepage monitoring data, stress strain monitoring data, and temperature monitoring data; A first determination module is used to determine various index analysis results of the concrete dam based on the same type of multi-measurement point monitoring data, wherein the various index analysis results include deformation analysis results, seepage analysis results, stress-strain analysis results, and temperature analysis results; A second determination module is used to determine the operation score results of various indicators of the concrete dam based on the multi-type and multi-point monitoring data, wherein the operation score results of various indicators include deformation score results, seepage score results, and stress-strain score results; The evaluation and analysis module is used to perform an overall evaluation and analysis on the concrete dam based on the analysis results of each indicator and the operation scoring results of each indicator.

Images