Fuzzy comprehensive intelligent evaluation method for seepage-proofing effect of grouting curtain

Abstract

The invention discloses a grouting curtain anti-seepage effect fuzzy comprehensive intelligent evaluation method which comprises the steps that a layer evaluation system of grouting curtain anti-seepage effect indexes is established, the layer evaluation system comprises a target layer, a criterion layer and a scheme layer from top to bottom in sequence, and each layer comprises a direct index, an effect index and a factor index; establishing a grouting curtain anti-seepage effect evaluation index data set according to the hierarchy evaluation system, and calculating the weight and the weight coefficient vector of each index of the next hierarchy of the hierarchy evaluation system to each index of the previous hierarchy; determining a membership evaluation matrix of each level index of the level evaluation system; multiplying the weight coefficient vector by the membership evaluation matrix, and calculating to obtain a fuzzy comprehensive evaluation result of the seepage-proofing effect of the grouting curtain; and carrying out unification processing on the fuzzy comprehensive evaluation result of the seepage-proofing effect of the grouting curtain. According to the invention, fuzzy comprehensive evaluation of the seepage-proofing effect of the grouting curtain is realized, and the evaluation result of the seepage-proofing effect is more comprehensive and objective.

Description

Technical Field

[0001] The invention belongs to the technical field of grouting curtain anti-seepage for dam bedrock in water conservancy and hydropower engineering, and specifically relates to a fuzzy comprehensive intelligent evaluation method for the anti-seepage effect of a grouting curtain. Background Art

[0002] With the rapid development of water conservancy and hydropower engineering construction in my country, the stability of reservoir dams has become the main factor affecting the safe operation of reservoirs. In the subsequent engineering operation process, the dam grouting curtain has been affected by high-pressure water seepage and groundwater dissolution for a long time, resulting in the attenuation of its anti-seepage effect and even failure. Therefore, it is of great practical significance to study the anti-seepage effect of reservoir dams and their dam foundation grouting curtains during operation.

[0003] At present, there are many methods to evaluate the anti-seepage effect of dam foundation grouting curtain. In engineering, the water pressure test method is generally used, that is, to detect whether the underground structure can effectively prevent groundwater seepage and its water resistance, to ensure the sealing and durability of the structure. In addition, the detection of anti-seepage effect can also be carried out by using the acoustic CT method, borehole imaging method, mathematical statistical analysis method, numerical simulation method, artificial intelligence calculation method, chemical balance method, etc.

[0004] Based on the above research methods, many scholars have conducted more in-depth research on the evaluation of anti-seepage effects in order to achieve accurate evaluation of the effectiveness of anti-seepage projects. Shen Changsong and Wu Zhongru (“Discussion on the Statistical Model of Concrete Dam Foundation Pressure”, Journal of Hohai University, 1989, 17(1):109-116) selected model factors and expressions based on the basic theory of seepage using deterministic functions and physical reasoning, and established statistical models for observation quantities such as concrete dam body seepage and foundation pressure, earth-rock dam foundation pressure gauge water level and leakage, and dam bypass seepage, and conducted quantitative processing and predicted their changing trends, thereby evaluating the anti-seepage effect of grouting curtains at the dam foundation; Peng Peng et al. (“Evaluation of the Anti-seepage Effect of Grouting Curtains Based on Distributed Multi-sensor Fusion Technology”, Surveying Science and Technology, 2009, (6):18-21) conducted multi-point monitoring of grouting curtains at the dam foundation based on distributed multi-sensor fusion technology, and established a dynamic system model to reflect the stage-by-stage changes in the anti-seepage performance of the grouting curtains.Tang et al. (“Meticulous simulation and evaluation of seepage control effects at rightbank dam foundation and underground powerhouse of Jinping I hydropowerstation”, Earth and Environment Science, 2018, 189: 02-2021) used an adaptive penalty function to numerically simulate the three-dimensional seepage field of Jinping Hydropower Station, predicted the distribution of the seepage field during operation, and evaluated its anti-seepage effect; Saleh analyzed the factors affecting seepage by using the SEEPIW model and artificial neural network (“Studying the seepage phenomena under a concrete dam using SEEP / W and Artificial Neural Network models”, Materials Science and Engineering, 2018, 443: 01-2029); Zhang et al. combined the complex geological three-dimensional finite element model with the artificial neural network to establish an inversion model of the dam foundation seepage field and evaluated the anti-seepage effect of the dam foundation grouting curtain (“Research on seepage field of concrete dam foundation based on artificial neural network”, Aleandria Engineering Journal, 2020, 60: 1-14); Ishfaque et al. used recurrent neural network and long short-term memory as time series algorithms to analyze parameters such as dam flow information from 2014 to 2020 and predict dam seepage (“Use of Recurrent Neural Network with Long Short-Term Memory for Seepage Prediction at Tarbela Dam”, KP, Pakistan. Energies, 2022, 15: 3123); Song Zichao conducted a comprehensive evaluation of different anti-seepage schemes from multiple aspects based on fuzzy neural network and gray clustering SPA theory, and selected the optimal anti-seepage scheme (“Optimal comparison analysis of bridge foundation anti-seepage scheme based on fuzzy neural network”, Science and Technology Innovation, 2022, 13: 141-144.).

[0005] In summary, although the existing technology uses different detection methods to make up for the deficiency of traditional anti-seepage effect relying only on water pressure test results as the only evaluation basis to a certain extent, there are the following limitations when only independent analysis is performed based on the results obtained by different detection methods: First, a single detection result only reflects the characteristics of a certain aspect of the rock mass after grouting, and it is difficult to describe the relationship between different detection results and the overall effect of the anti-seepage project, and it cannot be evaluated from multiple angles at the same time; second, the existing anti-seepage effect evaluation method is essentially a mathematical model based on a certain relationship, which ignores the uncertainty problem that is prevalent in the anti-seepage effect evaluation. Therefore, it is very important to establish an anti-seepage effect evaluation system under the influence of multiple factors. Summary of the invention

[0006] 1. Technical issues to be resolved

[0007] The invention aims to solve the problem that the existing evaluation method for the anti-seepage effect of dam bedrock grouting curtain is not comprehensive and objective, and cannot accurately perform anti-seepage analysis and prediction.

[0008] (II) Technical solution

[0009] In order to solve the above technical problems, the present invention proposes a fuzzy comprehensive intelligent evaluation method for the anti-seepage effect of grouting curtain, comprising the following steps:

[0010] S1. Establish a hierarchical evaluation system for the anti-seepage effect index of the grouting curtain. The hierarchical evaluation system includes three layers from top to bottom, namely, the target layer, the criterion layer and the scheme layer. Each layer includes direct indicators, effect indicators and factor indicators respectively. In this hierarchical evaluation system, the data of each indicator of the upper layer is determined by the data of each indicator of the lower layer, but the importance of each indicator of the same layer to a certain indicator of the upper layer may be the same or different.

[0011] S2. Establishing a grouting curtain anti-seepage effect evaluation index data set according to the hierarchical evaluation system, wherein the grouting curtain anti-seepage effect evaluation index data set includes a plurality of factor index data and a plurality of effect index data corresponding to the factor index data;

[0012] S3. Calculate the weight and weight coefficient vector of each indicator of the next level of the hierarchical evaluation system for each indicator of the previous level.

[0013] S4. Determine the membership evaluation matrix of each level indicator of the hierarchical evaluation system.

[0014] S5. Multiply the weight coefficient vector by the membership evaluation matrix to calculate the fuzzy comprehensive evaluation result of the anti-seepage effect of the grouting curtain.

[0015] S6. The fuzzy comprehensive evaluation results of the anti-seepage effect of the grouting curtain are processed into single values, and a three-dimensional model for anti-seepage effect evaluation is established.

[0016] S7. Establish a cloud model for evaluating the level of anti-seepage effects.

[0017] According to a preferred embodiment of the present invention, step S2 comprises:

[0018] S2.1. Obtain preliminary factor index data of the anti-seepage effect of grouting curtain;

[0019] S2.2, using a numerical simulation method and based on the preliminary factor indicator data, calculating preliminary effect indicator data corresponding to the preliminary factor indicator data;

[0020] S2.3. Build an artificial neural network model for calculating effect indicators, and use the preliminary factor indicator data and preliminary effect indicator data of the grouting curtain anti-seepage effect to train the artificial neural network model. Use the trained artificial neural network model to perform expansion calculations to obtain effect indicator data under more different factor indicator data, thereby establishing a grouting curtain anti-seepage effect evaluation indicator data set.

[0021] According to a preferred embodiment of the present invention, step S4 includes: according to each grouting curtain anti-seepage effect evaluation index data set, determining the value of each element in the evaluation matrix through the frequency of the evaluation factor value in each interval and the membership function.

[0022] According to a preferred embodiment of the present invention, step S6 includes: establishing an evaluation level V and a decision set E, wherein the decision set E is a fuzzy subset of the evaluation set V, and is a description of the degree of grading of the comprehensive condition of each evaluated object; processing the decision set E using a single-value method, that is, assigning a score to each level in the evaluation set V to form a score vector:

[0023] F=(f1,f2,···,f m ) T

[0024] Perform matrix product calculation on the decision set E and the score vector F, finally obtain the calculation result p, and establish a three-dimensional model for anti-seepage effect evaluation.

[0025] According to a preferred embodiment of the present invention, step S7 comprises:

[0026] The normal cloud model is adopted, and three digital characteristics, namely expectation (Ex), entropy (En) and excess entropy (He), are used to represent the evaluation level of the curtain anti-seepage effect, and a cloud model for the comprehensive evaluation results of the anti-seepage effect is established.

[0027] (III) Beneficial effects

[0028] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0029] (1) The present invention comprehensively considers the permeability of the grouting curtain, the shear seepage of the grouting-filled fractures, and the changes in the upstream and downstream water heads of the dam at different times, combines the influencing factors in the anti-seepage process with numerical simulation, indoor experiments and field data, and constructs a complete intelligent anti-seepage effect evaluation system, which can perform anti-seepage analysis and prediction more accurately.

[0030] (2) The present invention takes into account the fuzziness of multi-index decision-making problems through fuzzy comprehensive evaluation method and cloud model theory, and also handles the randomness in the evaluation system in a good way, organically combining the fuzziness and randomness of the evaluation object, and realizing the fuzzy comprehensive evaluation of the anti-seepage effect of the grouting curtain, making the evaluation result of the anti-seepage effect more comprehensive and objective. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a structural diagram of the hierarchical evaluation system of the grouting curtain anti-seepage effect index of the present invention.

[0032] Figure 2 It is a structural diagram of a grouting curtain anti-seepage effect index hierarchy evaluation system according to an embodiment of the present invention.

[0033] Figure 3 It is a plan and three-dimensional geological model map of the study area established by an embodiment of the present invention.

[0034] Figure 4 It is an artificial neural network model established by an embodiment of the present invention.

[0035] Figure 5 An embodiment of the present invention is to establish a cloud model for evaluating the level of anti-seepage effect.

[0036] Figure 6 It is an embodiment of the present invention to establish a comparison diagram of a cloud model of comprehensive evaluation results of anti-seepage effect and a cloud model of evaluation grade.

[0037] Figure 7 It is a three-dimensional model diagram for evaluating anti-seepage effect established in one embodiment of the present invention. DETAILED DESCRIPTION

[0038] The anti-seepage effect of the dam bedrock grouting curtain has no direct quantitative relationship with physical parameters such as flow rate and pressure, so the present invention introduces a fuzzy evaluation method to further analyze it. The fuzzy evaluation method converts qualitative evaluation into quantitative evaluation based on the membership theory of fuzzy mathematics, that is, fuzzy mathematics is used to make an overall evaluation of affairs or objects constrained by multiple factors, making the evaluation results of the anti-seepage effect of the dam bedrock grouting curtain more objective.

[0039] In general, the present invention proposes a new fuzzy comprehensive evaluation method for the anti-seepage effect of grouting curtain based on numerical simulation, neural network model and cloud model.

[0040] The fuzzy evaluation method of the anti-seepage effect of the grouting curtain of the present invention comprises the following steps:

[0041] Firstly, a hierarchical evaluation system of grouting curtain anti-seepage effect indicators is established.

[0042] The grouting curtain anti-seepage effect index evaluation system proposed by the present invention is a hierarchical evaluation system based on a fuzzy comprehensive evaluation method. Figure 1 The structure diagram of the grouting curtain anti-seepage effect index level evaluation system of the present invention is shown. Figure 1 As shown in the figure, the hierarchical evaluation system is divided into three layers from top to bottom, namely the target layer, the criterion layer and the program layer. Each layer includes different indicators. The indicators included in the target layer are called direct indicators, the indicators included in the criterion layer are called effect indicators, and the indicators included in the program layer are called factor indicators.

[0043] In this hierarchical evaluation system, the data of each indicator in the upper layer is determined by the data of each indicator in the lower layer, but the importance of each indicator in the same layer to a certain indicator in the upper layer may be the same or different.

[0044] Direct index refers to the index that directly describes the anti-seepage effect of the grouting curtain, that is, the final evaluation index that can directly reflect the anti-seepage effect of the grouting curtain to be obtained by the present invention, which is represented by T here; effect index refers to the index that can characterize the penetration effect or penetration degree of the grouting curtain, which is represented by C1, C2, ... here. Factor index refers to the index that can affect the anti-seepage performance of the grouting curtain, which is represented by F1, F2, ... here.

[0045] Furthermore, in order to calculate the final evaluation index T, the present invention needs to establish a grouting curtain anti-seepage effect evaluation index data set, which includes factor index data and effect index data corresponding to the factor index data. Furthermore, in order to fully reflect the anti-seepage performance of the grouting curtain, the numerical distribution range of the factor index data should be wide and uniform.

[0046] Specifically, the present invention constructs factor indicators based on the influencing factors of the specific target problem of grouting curtain anti-seepage. As a preferred embodiment, the present invention constructs factor indicators of grouting curtain anti-seepage effect based on the influencing factors of dam grouting curtain anti-seepage effect in actual water conservancy and hydropower projects, including: grouting curtain permeability coefficient, grouting filling fracture shear seepage permeability coefficient, and upstream and downstream water head of the dam.

[0047] The effect index is a more direct evaluation index for evaluating the penetration effect of the grouting curtain, for example, including the pressure and flow rate of each section of the grouting curtain.

[0048] Secondly, based on this hierarchical evaluation system, a data set of evaluation indicators for the anti-seepage effect of grouting curtains was established.

[0049] The grouting curtain anti-seepage effect evaluation index data set of the present invention includes a plurality of factor index data and a plurality of effect index data corresponding to the factor index data.

[0050] As a preferred embodiment, the present invention establishes a three-dimensional geological model according to the hydrogeological parameters of the dam, and obtains the pressure and flow information of each section of the grouting curtain through numerical simulation methods and factor index calculations as effect index data.

[0051] As a more preferred embodiment, after obtaining a part of the data set consisting of factor indicators and effect indicators through the data simulation method, the present invention expands the above data through an artificial neural network model. For example, an artificial neural network model is established, and the pressure and flow data sets of each section of the grouting curtain obtained through numerical simulation under different factor indicators are used as training sets to optimize the artificial neural network model, and then a large amount of effect indicator data under more different factor indicators, namely pressure and flow data, is calculated through the artificial neural network model. Through this step, repeated numerical simulation processes can be avoided.

[0052] Then, the fuzzy comprehensive evaluation results of anti-seepage effect are calculated based on the evaluation index data set of grouting curtain anti-seepage effect.

[0053] This step determines the weight coefficient vector and weight of each grouting curtain anti-seepage effect evaluation index according to the grouting curtain anti-seepage effect evaluation index data set, and determines the evaluation matrix of each grouting curtain anti-seepage effect evaluation index, and based on this, establishes a three-dimensional geological model map for anti-seepage effect evaluation and / or an anti-seepage effect evaluation grade cloud model.

[0054] Determine the weight matrix and weight coefficient vector D of the grouting curtain anti-seepage effect evaluation index. According to the established anti-seepage effect evaluation index system, the weight of the anti-seepage effect evaluation index is calculated using the hierarchical analysis method. The importance of each evaluation index of the anti-seepage effect is compared in pairs, and the weight coefficient vector of the anti-seepage effect evaluation is calculated.

[0055] Furthermore, the evaluation matrix R of the anti-seepage effect evaluation index of each grouting curtain is determined specifically including: according to the measured data of each anti-seepage unit, combined with the anti-seepage effect evaluation grade and numerical simulation results, the value of each element in the evaluation matrix R is determined by the frequency and membership function of the evaluation factor value in each interval.

[0056] Furthermore, the calculation of the fuzzy comprehensive evaluation result of the anti-seepage effect includes: based on the weight coefficient vector D of the determined anti-seepage effect evaluation index and the calculation result of the evaluation matrix R of each determined anti-seepage effect evaluation index, according to the basic operation rules, the fuzzy comprehensive evaluation result vector B of the anti-seepage effect of the grouting curtain is obtained:

[0057] B=D×R

[0058] Among them, D represents the weight coefficient vector of the anti-seepage effect evaluation index, R represents the evaluation matrix of the anti-seepage effect evaluation index, and B represents the fuzzy evaluation result vector of the anti-seepage effect evaluation index.

[0059] Furthermore, the decision set E is a fuzzy subset of the evaluation set V, which is a description of the degree of grading of the comprehensive status of each evaluated object. The decision set E is processed using the single-value method, that is, a score is assigned to each level in the evaluation set to form a score vector:

[0060] F=(f1,f2,···,f m ) T

[0061] Furthermore, for the establishment of a three-dimensional geological model map for anti-seepage effect evaluation, the decision set E and the score vector F are matrix multiplied to finally obtain the calculation result p, which reflects the comprehensive information brought by the hierarchical fuzzy subset E and the equivalent score vector F, and is a comprehensive evaluation of the anti-seepage effect. Repeat this process for different sections of the corridor, and establish a three-dimensional geological model of anti-seepage effect evaluation indicators based on the obtained data.

[0062] Furthermore, the fuzzy comprehensive evaluation of anti-seepage effect based on numerical simulation and cloud model includes: using the normal cloud model, using the three digital features of expectation (Ex), entropy (En) and hyperentropy (He) to represent the evaluation level of the anti-seepage effect of the grouting curtain, and establishing a cloud model of the comprehensive evaluation result of the anti-seepage effect. According to the fuzzy comprehensive evaluation method, the weight cloud matrix of the anti-seepage evaluation index is established by the hierarchical analysis method, and the membership cloud model is established by the forward cloud generator, and then the cloud model of the comprehensive evaluation result of the anti-seepage effect is obtained. The evaluation level cloud model is compared with the determined anti-seepage effect comprehensive evaluation result cloud model, and the evaluation level is judged by the intuitive judgment criterion.

[0063] In order to make the objectives, technical solutions and advantages of the present invention more clear, specific implementation modes of the present invention are further described in detail below with reference to the accompanying drawings.

[0064] S1. Establish a hierarchical evaluation system for the anti-seepage effect of grouting curtains.

[0065] Figure 2 The structural diagram of the grouting curtain anti-seepage effect index level evaluation system of this embodiment of the present invention is shown. Figure 2As shown, it is the present invention Figure 2 The specific implementation method of the hierarchical evaluation system shown is divided into three layers from top to bottom, namely the target layer, the criterion layer and the program layer. Each layer includes different indicators, namely direct indicators, effect indicators and factor indicators.

[0066] In this embodiment, the direct index is the evaluation index T of the grouting curtain anti-seepage effect. For example, the evaluation set V of the evaluation level of the grouting curtain anti-seepage effect can be determined as five levels: "excellent", "good", "average", "poor", and "bad", that is:

[0067] V={V1,V2,V3,V4,V5}={excellent, good, fair, poor, bad}.

[0068] The effect indicators include pressure index C1 and flow index C2, and the factor indicators include water head F1, grouting curtain permeability coefficient F2 and grouting filling fracture permeability coefficient F3.

[0069] S2. Establishing a grouting curtain anti-seepage effect evaluation index data set according to the hierarchical evaluation system. The grouting curtain anti-seepage effect evaluation index data set includes a plurality of factor index data and a plurality of effect index data corresponding to the factor index data.

[0070] S2.1. Obtain preliminary factor index data of the anti-seepage effect of grouting curtain.

[0071] In this embodiment, the factor indexes include water head index, grouting curtain permeability index, and grouting filling fracture shear seepage permeability index. Part of the data of these indexes can be obtained by grouting curtain anti-seepage hydrogeological survey and indoor test, which is called preliminary factor index data.

[0072] (1) Head index. Head represents the mechanical energy of river fluid and is composed of pressure head and position head. By surveying the hydrogeological information upstream and downstream of the study area, the range of river water level changes and head changes are obtained:

[0073]

[0074] Where H is the total water head of the river, in m; p is the pressure of the river fluid, in MPa; ρ is the density of the river fluid, in kg / m 3 ; g is the acceleration due to gravity, unit is m / s 2 ; is the pressure head of the river, in meters; z is the position head based on the reference surface height, in meters.

[0075] (2) Grouting curtain permeability index. The permeability of the grouting curtain is obtained through on-site water pressure test and grouting numerical simulation method. The permeability coefficient is used to represent the permeability characteristics of the grouting curtain. The permeability coefficient refers to the unit flow rate under unit hydraulic gradient in an isotropic medium, which indicates the difficulty of the fluid passing through the pore skeleton:

[0076]

[0077] Where K is the permeability coefficient of the grouting curtain, in m / s; k is the permeability of the grouting curtain, in m 2 ; μ is the fluid viscosity, unit: Pa·s.

[0078] (3) Permeability coefficient index of shear seepage of grouting-filled fissures. Through the shear seepage test of indoor grouting curtain grouting-filled fissures, the long-term permeability performance change of grouting-filled fissures under shear disturbance is obtained and expressed by the permeability coefficient:

[0079]

[0080] in is the porosity of the fracture filling medium; b is the fracture opening, in m; d is the pore size of the filling particles, in m.

[0081] S2.2. Calculate preliminary effect index data corresponding to the preliminary factor index data using numerical simulation methods and based on the preliminary factor index data.

[0082] In this embodiment, a three-dimensional geological model is established through a numerical simulation method and based on the hydrogeological information of the study area to calculate the pressure and flow rate of the study area. This method uses Feflow groundwater numerical simulation software for calculation, and its control equations and boundary conditions are as follows:

[0083]

[0084] H(x,y,z,0)=H0(x,y,z)∈Ω

[0085]

[0086] H(x,y,z,t)=H1(x,y,z)∈S1

[0087] Where K xx , K yy , K zz are the permeability coefficients in the x, y, and z directions, respectively, in m / s; w is the source and sink term, including evaporation, rainfall infiltration recharge, well pumping, etc.; μ sis the water storage rate; H0 is the initial groundwater level; Ω is the groundwater seepage area; H1 is the river level; S1 is the first type of boundary (given water level boundary conditions); S2 is the second type of boundary (given flow boundary conditions); q(x, y, z, t) represents the flow at different times at different positions on the boundary. The pressure and flow data are the preliminary effect indicator data.

[0088] Figure 3 The plan view and three-dimensional geological model map of the study area established in this embodiment are shown.

[0089] S2.3. Build an artificial neural network model for calculating effect indicators, and use the preliminary factor indicator data and preliminary effect indicator data of the grouting curtain anti-seepage effect to train the artificial neural network model. Use the trained artificial neural network model to perform expansion calculations to obtain effect indicator data under more different factor indicator data, thereby establishing a grouting curtain anti-seepage effect evaluation indicator data set.

[0090] In this embodiment, an artificial neural network model for calculating effect indicators is first established. Different preliminary factor indicators and pressure and flow data of each section of the grouting curtain obtained through seepage simulation are used as training data sets, and network parameters are continuously adjusted through optimization algorithms such as gradient descent to minimize prediction errors, thereby optimizing the established artificial neural network model.

[0091] Furthermore, the upstream and downstream water heads of the dam, the permeability coefficient of the grouting curtain, and the shear seepage permeability coefficient of the grouting-filled fractures are taken as input parameters. By optimizing the trained artificial neural network model, the pressure and flow data of each section of the grouting curtain are output, thereby establishing a data set of evaluation indicators for the anti-seepage effect of the grouting curtain.

[0092] Figure 4 A schematic diagram of an artificial neural network model constructed according to an embodiment of the present invention is shown.

[0093] S3. Calculate the weight coefficient vector and weight of the anti-seepage effect evaluation index.

[0094] Determining the weights and weight coefficient vectors of each grouting curtain anti-seepage effect evaluation index specifically includes: using the hierarchical analysis method to calculate the weights of the anti-seepage effect evaluation index.

[0095] The anti-seepage effect evaluation index judgment matrix is ​​used to compare the importance of each anti-seepage effect evaluation index in pairs and obtain the weight of the anti-seepage effect evaluation index.

[0096] One of the important steps is to select a suitable scaling method to establish a pairwise judgment matrix. In this embodiment, an anti-seepage effect evaluation index judgment matrix is ​​established as shown in Table 1, and the importance of each anti-seepage effect evaluation index is compared pairwise to obtain the weight of the anti-seepage effect evaluation index.

[0097] Table 1 Scaling criteria for evaluation indexes of anti-seepage effect

[0098]

[0099] In the hierarchical evaluation system of grouting curtain anti-seepage effect indicators, for an element in the previous layer, the weight of the importance order of the elements in this layer that are related to an element in the previous layer is calculated according to the judgment matrix. The maximum eigenvalue of the judgment matrix can be calculated to obtain its eigenvector and weight coefficient vector. The calculation steps are as follows:

[0100] Compute the product M of each row of the matrix:

[0101]

[0102] Calculate M i The nth root of:

[0103]

[0104] Vector Normalization:

[0105]

[0106] Then we get the eigenvector W of the judgment matrix T, C1, and C2 T =(W1,W2) T ,

[0107]

[0108] Specifically, this embodiment conducts a comprehensive analysis on the anti-seepage effect of the grouting curtain of the Dongzhuang Water Conservancy Project.

[0109] The weight matrix and eigenvalues ​​of the evaluation index of the anti-seepage effect of the grouting curtain are calculated, and the importance of each element at the same level with respect to a criterion in the previous level is compared pairwise. The judgment matrix T of the criterion layer to the target layer and the judgment matrix C1 and C2 of the index layer to the two options of the criterion layer are established, as shown in Table 2, Table 3 and Table 4.

[0110] Table 2T judgment matrix

[0111]

[0112] Table 3C1 Judgment Matrix

[0113]

[0114] Table 4C2 Judgment Matrix

[0115]

[0116] In the table, x ij,y ij 、z ij They represent the importance of the factors in the i-th row relative to the elements in the j-th column to the evaluation level. If they are greater than 1, it means that the factors in the i-th row are more important; if they are less than 1, it means that the elements in the j-th column are more important; if they are equal to 1, it means that the two are equally important. λ1, λ2, and λ3 are the maximum eigenvalues ​​of the judgment matrix.

[0117] According to the hierarchical analysis method, as shown in Table 1, the weights of the matrices are assigned, that is, the matrices T, C1, and C2 are assigned natural numbers of 1 to 9 according to the importance of each element, and the characteristic vector W and weight coefficient vector W of each anti-seepage effect evaluation index are obtained as follows:

[0118] W T =(W1,W2) T

[0119]

[0120] D=[d1,d2,d3]

[0121] S4. Determine the evaluation matrix of the anti-seepage effect evaluation index of each grouting curtain, which specifically includes: according to the measured data and numerical simulation results of each anti-seepage unit, combined with the anti-seepage effect evaluation level, determine the value of each element in the evaluation matrix R through the frequency and membership function of the evaluation factor value in each interval.

[0122] By using the method for establishing the membership evaluation matrix of the grouting curtain anti-seepage effect evaluation index of the present invention, taking the Dongzhuang Water Conservancy Project as an example, the upstream and downstream water heads of the study area are, the grouting curtain permeability coefficient is, and the grouting filling fracture permeability coefficient is. According to the measured data and numerical simulation results of the study area, the anti-seepage effect evaluation grade classification standard is obtained through the expert scoring method and the membership function, and the results are shown in Table 5.

[0123] Table 5 Classification standard of anti-seepage effect evaluation

[0124] The membership evaluation matrix of each evaluation index in the study area is obtained as follows:

[0125]

[0126] S5. Calculation of the fuzzy comprehensive evaluation results of anti-seepage effect includes: based on the calculation results of the weight coefficient vector cloud model of the anti-seepage effect evaluation index determined above and the membership cloud model of each anti-seepage effect evaluation index determined, according to the basic operation rules of the cloud model, a fuzzy comprehensive evaluation method of the anti-seepage effect of the grouting curtain based on the cloud model is proposed:

[0127] B=D×R

[0128] Among them, D represents the weight coefficient vector of the anti-seepage effect evaluation index, R represents the evaluation matrix of the anti-seepage effect evaluation index, and B represents the fuzzy evaluation result vector of the anti-seepage effect evaluation index.

[0129] By using the fuzzy comprehensive evaluation method for the anti-seepage effect of grouting curtains described in the present invention, the weight matrix of the grouting anti-seepage effect evaluation index is multiplied by the membership evaluation matrix of the evaluation index of the study area, and the fuzzy comprehensive evaluation result of the anti-seepage effect of the study area can be calculated:

[0130]

[0131] where b 11 ,b 12 ,b 13 ,b 14 ,b 15 , respectively represent the probability of excellent, good, fair, poor and bad anti-seepage effects.

[0132] S6. The fuzzy comprehensive evaluation results of the grouting curtain anti-seepage effect are processed into single values ​​to obtain a three-dimensional geological model of the anti-seepage effect. The decision set E is a fuzzy subset of the evaluation set V. It is a description of the degree of grading of the comprehensive status of each evaluated object. It cannot be directly used for ranking and evaluation among the evaluated objects. It must be further analyzed and processed before it can be used. The decision set E is processed by the single value method, that is, a score is assigned to each level in the evaluation set to form a score vector:

[0133] F=(f1,f2,···,f m ) T

[0134] The decision set E and the score vector F are matrix-producted, and the calculation result p is finally obtained. This value reflects the comprehensive information brought by the hierarchical fuzzy subset E and the equivalent score vector F, which is a comprehensive evaluation of the anti-seepage effect. Based on the obtained data, a three-dimensional geological model of the anti-seepage effect evaluation index is established, such as Figure 7 shown.

[0135] The decision set F is processed by the singularization method, that is, a score is assigned to each level in the evaluation set to form a score vector:

[0136] F=(f1,f2,···,f m ) T

[0137] p=F×B

[0138] Perform matrix product calculation on the decision set E and the score vector F, and finally obtain the calculation result p. This value reflects the comprehensive information brought by the graded fuzzy subset E and the equivalent score vector F, and is a comprehensive evaluation of the anti-seepage effect, such as Figure 7 As shown, a three-dimensional geological model of anti-seepage effect evaluation indicators is established based on the obtained data.

[0139] S7. Establish a cloud model for evaluating the level of anti-seepage effect. By using the cloud model establishment method for evaluating the level of anti-seepage effect described in the present invention, the three cloud digital eigenvalues ​​of each evaluation level of the anti-seepage effect of the grouting curtain and their ranges on the x-axis are calculated, as shown in Table 6. A program is compiled in Matlab software to draw a normal cloud model for each evaluation level of the anti-seepage effect of the grouting curtain, as shown in Table 6. Figure 5 shown.

[0140] According to the fuzzy comprehensive evaluation method, the elements of each row in the judgment matrix are calculated through the hierarchical analysis method to obtain the three digital eigenvalues ​​of the weight cloud, namely, expectation (Ex), entropy (En) and hyperentropy (He), and the weight cloud matrix of the anti-seepage evaluation index is established.

[0141] A forward cloud generator is used to generate N cloud droplets. According to the generated cloud droplets and the principle of a reverse cloud generator, the three digital eigenvalues ​​of the membership cloud model R for comprehensive evaluation of each index of anti-seepage effect are obtained, and the membership cloud model is established.

[0142] The multiplication of the two results in a cloud model of comprehensive evaluation results of anti-seepage effect with three digital eigenvalues: specific expectation (Ex), entropy (En) and super entropy (He), such as Figure 6 As shown, the evaluation grade cloud model is compared with the determined anti-seepage effect comprehensive evaluation result cloud model, and the evaluation grade is judged by using intuitive judgment criteria.

[0143] The normal cloud model is adopted, and the three digital features of expectation (Ex), entropy (En) and super entropy (He) are used to represent the evaluation grade cloud model of the grouting curtain anti-seepage effect. The specific division of the evaluation grade adopts the cloud generation method of the golden ratio. In the interval [0, 1], the central point 0.5 is the expectation of the middle evaluation grade "general" of the five evaluation grades. The closer to the center of the interval, the smaller the entropy and super entropy of the evaluation grade cloud, and the smaller the entropy and super entropy of the adjacent evaluation grades is 0.618 times the larger one. According to this principle, the three cloud digital eigenvalues ​​of each evaluation grade of the grouting curtain anti-seepage effect and their range on the x-axis are determined, as shown in Table 6. A program is compiled in the Matlab software to draw the normal cloud model of each evaluation index of the grouting curtain anti-seepage effect, as shown in Table 6. Figure 5 shown.

[0144] Table 6 Cloud digital characteristic values ​​of evaluation levels

[0145]

[0146] According to the fuzzy comprehensive evaluation method, the elements of each row in the judgment matrix are calculated through the hierarchical analysis method to obtain the three digital eigenvalues ​​of the weight cloud, namely, the expectation (Ex), entropy (En) and super entropy (He), and the weight cloud matrix of the anti-seepage evaluation index is established:

[0147] W 0 =(Ex 0 ,En 0 ,He 0 ) T

[0148] A forward cloud generator is used to generate N cloud droplets. Based on the generated cloud droplets and the principle of the reverse cloud generator, the three digital eigenvalues ​​of the membership cloud model R for comprehensive evaluation of each index of the anti-seepage effect are obtained, and the membership cloud model is established:

[0149]

[0150] The multiplication of the two results in a cloud model of comprehensive evaluation results of anti-seepage effect with three digital eigenvalues: specific expectation (Ex), entropy (En) and super entropy (He), such as Figure 6 As shown, the evaluation grade cloud model is compared with the determined anti-seepage effect comprehensive evaluation result cloud model, and the evaluation grade is judged by using intuitive judgment criteria.

[0151] B 0 =W 0 ×R 0

[0152] The present invention is not limited to the embodiments described above. The above description of the specific embodiments is intended to describe and illustrate the technical solution of the present invention. The above specific embodiments are merely illustrative and not restrictive. Without departing from the scope of the present invention and the scope of protection of the claims, a person of ordinary skill in the art can also make many forms of specific changes under the guidance of the present invention, which all fall within the scope of protection of the present invention.

Claims

1. A fuzzy comprehensive intelligent evaluation method for grouting curtain anti-seepage effect, characterized in that: The steps include: S1. Establish a hierarchical evaluation system for the anti-seepage effect index of the grouting curtain. The hierarchical evaluation system includes three layers from top to bottom, namely, the target layer, the criterion layer and the scheme layer. Each layer includes direct indicators, effect indicators and factor indicators respectively. In this hierarchical evaluation system, the data of each indicator of the upper layer is determined by the data of each indicator of the lower layer, but the importance of each indicator of the same layer to a certain indicator of the upper layer may be the same or different. S2. Establishing a grouting curtain anti-seepage effect evaluation index data set according to the hierarchical evaluation system, wherein the grouting curtain anti-seepage effect evaluation index data set includes a plurality of factor index data and a plurality of effect index data corresponding to the factor index data; S3, calculating the weight and weight coefficient vector of each indicator of the next level of the hierarchical evaluation system for each indicator of the previous level; S4, determining the membership evaluation matrix of each level indicator of the hierarchical evaluation system; S5, multiplying the weight coefficient vector by the membership evaluation matrix to calculate the fuzzy comprehensive evaluation result of the anti-seepage effect of the grouting curtain; S6. Performing single-value processing on the fuzzy comprehensive evaluation results of the anti-seepage effect of the grouting curtain to obtain a three-dimensional geological model of the anti-seepage effect; S7. Establish a cloud model for evaluating the level of anti-seepage effects.

2. The fuzzy comprehensive intelligent evaluation method for the anti-seepage effect of grouting curtain according to claim 1 is characterized in that: The step S2 comprises: S2.

1. Obtain preliminary factor index data of the anti-seepage effect of grouting curtain; S2.2, using a numerical simulation method and based on the preliminary factor indicator data, calculating preliminary effect indicator data corresponding to the preliminary factor indicator data; S2.

3. Build an artificial neural network model for calculating effect indicators, and use the preliminary factor indicator data and preliminary effect indicator data of the grouting curtain anti-seepage effect to train the artificial neural network model. Use the trained artificial neural network model to perform expansion calculations to obtain effect indicator data under more different factor indicator data, thereby establishing a grouting curtain anti-seepage effect evaluation indicator data set.

3. The fuzzy comprehensive intelligent evaluation method for the anti-seepage effect of grouting curtain according to claim 1 is characterized in that: Step S4 includes: According to the evaluation index data sets of each grouting curtain anti-seepage effect, the value of each element in the evaluation matrix is ​​determined by the frequency of the evaluation factor value in each interval and the membership function.

4. The fuzzy comprehensive intelligent evaluation method for the anti-seepage effect of grouting curtain according to claim 1 is characterized in that: Step S6 includes: Establish the evaluation level V and decision set E. The decision set E is a fuzzy subset of the evaluation set V, which is a description of the degree of grading of the comprehensive status of each evaluated object; The decision set E is processed by the singularization method, that is, a score is assigned to each level in the evaluation set V to form a score vector: F=(f1,f2,···,f m ) T The decision set E and the score vector F are matrix multiplied to obtain the calculation result p, and a three-dimensional model for anti-seepage effect evaluation is established.

5. The fuzzy comprehensive intelligent evaluation method for the anti-seepage effect of grouting curtain according to claim 1 is characterized in that: The step S7 comprises: The normal cloud model is adopted, and the three digital characteristics of expectation (Ex), entropy (En) and excess entropy (He) are used to represent the evaluation level of the curtain anti-seepage effect, and a cloud model for the comprehensive evaluation results of the anti-seepage effect is established.

Images