Concrete dam deformation safety comprehensive evaluation method

Abstract

The invention discloses a concrete dam deformation safety comprehensive evaluation method, which belongs to the technical field of dam safety monitoring, and comprises the following steps: step 1, determining adverse working conditions; 2, constructing a concrete dam deformation safety comprehensive evaluation index system; 3, determining an identification framework of the deformation safety state of the concrete dam; step 4, calculating a subjective weight by a G1 method; step 5, objective weight is calculated through a CRITIC method; step 6, establishing a G1-CRITIC-GT model to obtain a subjective and objective combination weight; 7, drawing up a deformation monitoring threshold value, and dividing a safety state interval; step 8, obtaining BPA of the bottom layer evaluation indexes; step 9, carrying out weighting processing on the BPA of the bottom layer evaluation index, and introducing uncertainty; step 10, performing fusion diagnosis on the evaluation indexes; and 11, the deformation safety state of the concrete dam is determined. The method achieves the collaborative optimization of subjective and objective weights, improves the reliability of an evaluation result, quantifies the fuzziness and uncertainty of monitoring data, and achieves the reasonable distribution of conflict evidences and the deep fusion of multi-source information.

Description

Technical Field

[0001] This invention relates to a comprehensive evaluation method for the deformation safety of concrete dams, belonging to the field of dam safety monitoring technology. Background Technology

[0002] Concrete dam deformation monitoring systems typically integrate multiple sensors, including tension wires, static levels, plumb lines, and lines of sight. These systems are characterized by wide acquisition ranges, diverse acquisition methods, and high acquisition frequencies, resulting in increasingly numerous and complex deformation monitoring data. These multi-source, heterogeneous data may exhibit complementary and redundant relationships, as well as contradictions, posing a significant challenge to the comprehensive evaluation of concrete dam deformation safety. Furthermore, the concrete dam deformation monitoring environment is extremely complex, and sensors are susceptible to interference from various factors during data acquisition, leading to significant ambiguity and uncertainty in the data. In conducting comprehensive evaluations of concrete dam deformation safety, this uncertainty manifests as significant differences or even conflicts in the safety status information provided by different sensors within the same time period. Accurately handling this uncertainty becomes a key technical challenge in the comprehensive evaluation of concrete dam deformation safety.

[0003] For the comprehensive evaluation of the deformation safety of concrete dams, relevant scholars have proposed a variety of analytical methods, mainly including finite element calculation, multi-attribute decision-making, artificial neural networks and Bayesian framework. These methods have achieved certain results in practical applications, but they all have inherent defects: (1) Inadequate handling of uncertainty: Most traditional methods are difficult to handle both randomness and fuzziness at the same time. For example, finite element calculation relies on precise physical models and parameters, which are difficult to fully reflect the uncertainty in actual monitoring data; Bayesian methods require prior probability distributions, which often have strong subjectivity in practical applications. (2) Difficulty in resolving information conflicts: When there are high conflicts between multi-source monitoring information, traditional methods lack effective conflict resolution mechanisms. For example, artificial neural network methods often operate as "black boxes" when processing conflicting information, lacking interpretability. (3) Over-reliance on sample quantity: Many machine learning methods require a large number of labeled samples for training, while dam safety events, especially dangerous cases, are relatively rare, resulting in insufficient model generalization ability.

[0004] In summary, to improve the reliability and accuracy of the comprehensive evaluation of the deformation safety of concrete dams, there is an urgent need for a multi-source information fusion method that can effectively characterize uncertain information without relying on prior knowledge and can integrate fuzzy and uncertain monitoring data. Summary of the Invention

[0005] The purpose of this invention is to provide a comprehensive evaluation method for the deformation safety of concrete dams. This method constructs a conflict evidence identification and weighting mechanism based on G1-CRITIC-GT and designs a multi-level evidence fusion process based on improved DS evidence theory, which can effectively handle highly conflicting evidence. On this basis, it integrates multi-source monitoring information to achieve accurate and reliable diagnosis of the deformation safety status of concrete dams, thus providing technical support for the safe operation of concrete dams.

[0006] This invention is achieved through the following technical solution: A comprehensive evaluation method for the deformation safety of concrete dams includes the following steps: Step 1: Based on the layout of the concrete dam deformation monitoring system, select sensor measurement points for deformation safety evaluation, and determine one of the unfavorable working conditions of the concrete dam during its operation as the basis for subsequent analysis. Step 2: Select deformation safety evaluation indicators based on the actual engineering characteristics of concrete dams and relevant data, and construct a comprehensive evaluation index system for deformation safety of concrete dams; Step 3: Determine the identification framework for the deformation safety state of the concrete dam. ; Step 4: Calculate the subjective weights of various bottom-level evaluation indicators in the comprehensive evaluation index system for deformation safety of concrete dams using the G1 method; Step 5: Calculate the objective weights of various bottom-level evaluation indicators in the comprehensive evaluation index system for deformation safety of concrete dams using the CRITIC method; Step 6: Establish the G1-CRITIC-GT combined weighting model to obtain the subjective and objective combined weights of various bottom-level evaluation indicators in the comprehensive evaluation index system for deformation safety of concrete dams, and quantify the importance of different bottom-level evaluation indicators in multi-source information fusion. Step 7: Based on the environmental data and deformation monitoring data of the concrete dam, establish a statistical model between the deformation effect of each underlying evaluation index and different loads using the stepwise regression method, and formulate the identification framework for each underlying evaluation index using the confidence interval method. The deformation monitoring threshold is used to divide its safe state range; Step 8: Obtain the basic probability distribution of various underlying evaluation indicators, i.e., BPA, through the interval number method, as the original evidence; Step 9: Use the subjective and objective combined weights calculated in Step 6 to weight the BPA of the underlying evaluation index, and then apply the weights within the identification framework. The introduction of uncertainty allocation may lead to information conflicts; Step 10: Use the improved DS evidence theory to perform fusion diagnosis on the deformation evaluation index of concrete dam, and realize the layer-by-layer fusion of multi-source deformation monitoring information; Step 11: Determine the deformation safety status of the concrete dam based on the principle of maximum confidence.

[0007] The identification framework in step 3 It consists of five safety states and one uncertainty state, namely: {normal, basically normal, slightly abnormal, severely abnormal, malignant abnormal, and uncertainty}.

[0008] Step 4 includes the following steps: Step 4.1: Invite several experts to rank the various underlying evaluation indicators from highest to lowest importance based on their own experience and subjective judgment. Assume that one type of underlying evaluation indicator includes n indicators, namely... The order of the evaluation indicators obtained after expert ranking is as follows: ; Step 4.2: Several experts score the relative importance of adjacent indicators. For any two adjacent indicators... and The relative importance ratio assigned by experts Represented as: (1), In the formula: , The larger the value, the better the indicator. Compared to indicators The greater the importance; Step 4.3: Due to the differences in the professional background, engineering experience and authority of different experts, we will comprehensively consider the experts' education, professional title, years of service and familiarity with the field, establish expert information evaluation rules, and score each expert. Step 4.4: Normalize the total scores of each expert and assign weights to each expert; Step 4.5: Calculate the subjective weights of each expert for the underlying evaluation index using the G1 method according to equations (2) and (3). ; (2), (3), in, ,and ; Step 4.6: Use the expert weights from Step 4.4 to weight the subjective weights of different experts in Step 4.5 using the G1 method. The sum of the subjective weights of the same underlying evaluation index after being weighted by several experts is the final subjective weight of that underlying evaluation index.

[0009] Step 5 includes the following steps: Step 5.1: Given n samples to be evaluated and p evaluation indicators, form the original indicator data matrix. : (4), In the formula, This represents the value of the i-th sample on the j-th evaluation index; Step 5.2: Process the original indicator data matrix. Perform dimensionless processing and construct a standardized matrix. ; Step 5.3: Calculate the normalized matrix using the following formula. Standard deviation To determine the volatility among the various evaluation indicators. (5), (6), In the formula, Let j be the average value of the data for the j-th evaluation indicator; Step 5.4: Use the correlation coefficient between the i-th evaluation index and the j-th evaluation index. To measure the conflict between various evaluation indicators. ; (7), (8), Step 5.5, based on volatility and conflict Calculate information carrying capacity ; (9), Step 5.6: Take the proportion of the information content of the j-th evaluation indicator to the total information content as the objective weight of that evaluation indicator. , (10) Step 5.7: From this, the objective weights of various underlying evaluation indicators can be obtained sequentially using the CRITIC method.

[0010] Step 6 includes the following steps: Step 6.1: Suppose that L different weighting methods are used to calculate L weight vectors, and each weight vector is represented as: (11), In the formula, n represents the number of evaluation indicators.

[0011] Step 6.2, Let any linear combination of these L vectors be: (12) In the formula, The linear combination coefficients of the weights obtained by the k-th weighting method are... For a given possible weight vector; Step 6.3: Using game theory, optimize the linear combination coefficients of each indicator to make... and Minimize the deviation between: (13) Step 6.4: Solve the equivalent linear equation system to obtain the optimal linear combination coefficient vector: (14) Step 6.5: For the optimal linear combination coefficient vector Perform a normalization operation to obtain the normalized combination coefficients. and the corresponding combined weighted vectors : (15) (16).

[0012] Step 8 includes the following steps: Step 8.1: Determine the measured values ​​of each underlying evaluation index under unfavorable working conditions as determined in Step 1, and construct an interval number model corresponding to the measured value, denoted as . ; Step 8.2: Based on the underlying evaluation indicators obtained in Step 7, in the identification framework... The safe state range below and the interval number model in step 8.1 Calculate the distance between two intervals. ; Step 8.3: Based on the distance between the two interval numbers in Step 8.2 Calculate the similarity between interval numbers ; Step 8.4: Normalize the similarity results between the interval numbers in Step 8.3 to generate the BPA of each underlying evaluation index.

[0013] The distance between the two intervals in step 8.2 The following formula is used for calculation: (17) The similarity between interval numbers in step 8.3 The following formula is used for calculation: (18) In the formula, the support coefficient The value is usually 5.

[0014] Step 9 includes the following steps: Step 9.1, let the first... Individual evidence within the identification framework The BPA function inside is The corresponding focal elements are respectively The BPA function, weighted by the subjective and objective combination weights calculated in step 6, is expressed as follows: (19) Step 9.2: After weight allocation as in Step 9.1, the sum of the BPAs of each piece of evidence is no longer 1. Therefore, within the identification framework... Adding an additional uncertainty, the BPA assigned to the uncertainty of this body of evidence is: (20) Step 9.3, from which we can obtain the first... Each underlying evaluation metric in the identification framework The following BPAs are respectively , , , , , .

[0015] Step 10 includes the following steps: Step 10.1: Merge the weighted BPA of the underlying evaluation indicators obtained in Step 9 to obtain the BPA distribution of their respective superior evaluation indicators. Let there be two weighted BPA functions of the underlying evaluation indicators, m1 and m2, with their corresponding focal elements being m1 and m2 respectively. and The Dempster composition rule can then be expressed as follows: (twenty one) In the formula, K is used to measure the degree of conflict between pieces of evidence, and the conflict coefficient K is defined as: (twenty two) Step 10.2: Based on the BPA distribution of the superior evaluation indicators to which the lower-level evaluation indicators belong in Step 10.1, the Dempster synthesis rule is also used to fuse them. Step 10.3: Repeat step 10.2 until the fusion yields the recognition framework. The following is a BPA distribution regarding the deformation safety state of concrete dams.

[0016] The underlying evaluation indicator is the monitoring point, and its superior evaluation indicator is the monitoring instrument.

[0017] The beneficial effects of this invention are as follows: 1. The comprehensive evaluation method for deformation safety of concrete dams provided by this invention calculates the subjective and objective weights of the underlying evaluation indicators using the G1 method and the CRITIC method, respectively, and obtains the combined subjective and objective weights that comprehensively reflect expert opinions and the attributes of the data itself using the GT method. This achieves synergistic optimization of subjective and objective weights, significantly reduces the risks caused by deviations in weight setting, and thus ensures the scientific nature of the multi-source monitoring information fusion process, while improving the reliability of the comprehensive evaluation results of dam deformation safety.

[0018] 2. The comprehensive evaluation method for deformation safety of concrete dams provided by this invention addresses the shortcomings of traditional DS evidence theory in handling conflicting evidence and synthesis rules. It innovatively introduces a conflict coefficient to weight and modify the traditional DS evidence theory, significantly improving the robustness of the evidence fusion mechanism. In addition, by extending an uncertainty proposition after the identification framework, the fuzziness and uncertainty of concrete dam deformation monitoring data are effectively quantified, realizing the reasonable allocation of conflicting evidence and the deep fusion of multi-source information. Attached Figure Description

[0019] Figure 1 This is a flowchart of the method of the present invention.

[0020] Figure 2 This is a typical framework for a comprehensive evaluation index system for the deformation safety of concrete dams. Detailed Implementation

[0021] The technical solution of the present invention is further described below, but the scope of protection is not limited to what is described.

[0022] like Figure 1 As shown, the comprehensive evaluation method for deformation safety of concrete dams according to the present invention includes the following steps: Step 1: Based on the layout of the concrete dam deformation monitoring system, select sensor measurement points for deformation safety evaluation, and determine one of the unfavorable working conditions of the concrete dam during its operation as the basis for subsequent analysis.

[0023] Specifically, unfavorable operating conditions include high water levels and temperature drops.

[0024] Step 2: Select deformation safety evaluation indicators based on the actual engineering characteristics of concrete dams and relevant data, and construct a comprehensive evaluation index system for deformation safety of concrete dams.

[0025] Figure 2The figure shows a typical framework for a comprehensive evaluation index system for the deformation safety of concrete dams, which includes four categories of bottom-level evaluation indexes.

[0026] Step 3: Determine the identification framework for the deformation safety state of the concrete dam. .

[0027] Step 4: Use the G1 method to calculate the subjective weights of various bottom-level evaluation indicators in the comprehensive evaluation index system for deformation safety of concrete dams.

[0028] Step 5: Use the CRITIC method to calculate the objective weights of various bottom-level evaluation indicators in the comprehensive evaluation index system for deformation safety of concrete dams.

[0029] Step 6: Establish the G1-CRITIC-GT combined weighting model to obtain the subjective and objective combined weights of various bottom-level evaluation indicators in the comprehensive evaluation index system for deformation safety of concrete dams, and quantify the importance of different bottom-level evaluation indicators in multi-source information fusion.

[0030] Step 7: Based on the environmental data and deformation monitoring data of the concrete dam, establish a statistical model between the deformation effect of each underlying evaluation index and different loads using the stepwise regression method, and formulate the identification framework for each underlying evaluation index using the confidence interval method. The deformation monitoring threshold is used to divide the safe state range.

[0031] Step 8: Obtain the basic probability assignment (BPA) of various underlying evaluation indicators using the interval number method, as the original evidence.

[0032] Step 9: Use the subjective and objective combined weights calculated in Step 6 to weight the BPA of the underlying evaluation index, and then apply the weights within the identification framework. The introduction of uncertainty allocation may lead to information conflicts.

[0033] Step 10: Use the improved DS evidence theory to perform fusion diagnosis on the deformation evaluation index of concrete dam, and realize the layer-by-layer fusion of multi-source deformation monitoring information.

[0034] Step 11: Determine the deformation safety status of the concrete dam based on the principle of maximum confidence.

[0035] The identification framework in step 3 It consists of five safety states and one uncertainty state, namely: {normal, basically normal, slightly abnormal, severely abnormal, malignant abnormal, and uncertainty}.

[0036] Step 4 includes the following steps: Step 4.1: Invite several experts to rank the various underlying evaluation indicators from highest to lowest importance based on their own experience and subjective judgment. Assume that one type of underlying evaluation indicator includes n indicators, namely... The order of the evaluation indicators obtained after expert ranking is as follows: .

[0037] Step 4.2: Several experts score the relative importance of adjacent indicators. For any two adjacent indicators... and The relative importance ratio assigned by experts Represented as: (1), In the formula: , The larger the value, the better the indicator. Compared to indicators The greater its importance. Refer to Table 1 for the values.

[0038] Table 1 Importance of Adjacent Indicators

[0039] Step 4.3: Due to differences in the professional background, engineering experience, and authority of different experts, an expert information evaluation rule is established, taking into account each expert's education, professional title, years of service, and familiarity with the field, and each expert is scored. The expert information evaluation rule is shown in Table 2.

[0040] Table 2 Expert Information Evaluation Rules

[0041] Step 4.4: Normalize the total scores of each expert and assign weights to each expert; Step 4.5: Calculate the subjective weights of each expert for the underlying evaluation index using the G1 method according to equations (2) and (3). ; (2), (3), in, ,and ; Step 4.6: Use the expert weights from Step 4.4 to weight the subjective weights of different experts in Step 4.5 using the G1 method. The sum of the subjective weights of the same underlying evaluation index after being weighted by several experts is the final subjective weight of that underlying evaluation index.

[0042] Step 5 includes the following steps: Step 5.1: Given n samples to be evaluated and p evaluation indicators, form the original indicator data matrix. : (4), In the formula, This represents the value of the i-th sample on the j-th evaluation index; Step 5.2: Process the original indicator data matrix. Perform dimensionless processing and construct a standardized matrix. ; Step 5.3: Calculate the normalized matrix using the following formula. Standard deviation To determine the volatility among the various evaluation indicators. (5), (6), In the formula, Let j be the average value of the data for the j-th evaluation indicator; Step 5.4: Use the correlation coefficient between the i-th evaluation index and the j-th evaluation index. To measure the conflict between various evaluation indicators. ; (7), (8), Step 5.5, based on volatility and conflict Calculate information carrying capacity ; (9), Step 5.6: Take the proportion of the information content of the j-th evaluation indicator to the total information content as the objective weight of that evaluation indicator. , (10) Step 5.7: From this, the objective weights of various underlying evaluation indicators can be obtained sequentially using the CRITIC method.

[0043] Step 6 includes the following steps: Step 6.1: Suppose that L different weighting methods are used to calculate L weight vectors, and each weight vector is represented as: (11), In the formula, n represents the number of evaluation indicators.

[0044] Step 6.2, Let any linear combination of these L vectors be: (12) In the formula, The linear combination coefficients of the weights obtained by the k-th weighting method are... For a given possible weight vector; Step 6.3: Using game theory, optimize the linear combination coefficients of each indicator to make... and Minimize the deviation between: (13) Step 6.4: Solve the equivalent linear equation system to obtain the optimal linear combination coefficient vector: (14) Step 6.5: For the optimal linear combination coefficient vector Perform a normalization operation to obtain the normalized combination coefficients. and the corresponding combined weighted vectors : (15) (16).

[0045] Step 8 includes the following steps: Step 8.1: Determine the measured values ​​of each underlying evaluation index under unfavorable working conditions as determined in Step 1, and construct an interval number model corresponding to the measured value, denoted as . ; Step 8.2: Based on the underlying evaluation indicators obtained in Step 7, in the identification framework... The safe state range below and the interval number model in step 8.1 Calculate the distance between two intervals. ; Step 8.3: Based on the distance between the two interval numbers in Step 8.2 Calculate the similarity between interval numbers ; Step 8.4: Normalize the similarity results between the interval numbers in Step 8.3 to generate the BPA of each underlying evaluation index.

[0046] The distance between the two intervals in step 8.2 The following formula is used for calculation: (17) The similarity between interval numbers in step 8.3 The following formula is used for calculation: (18) In the formula, the support coefficient The value is usually 5.

[0047] Step 9 includes the following steps: Step 9.1, let the first... Individual evidence within the identification framework The BPA function inside is The corresponding focal elements are respectively The BPA function, weighted by the subjective and objective combination weights calculated in step 6, is expressed as follows: (19) Step 9.2: After weight allocation as in Step 9.1, the sum of the BPAs of each piece of evidence is no longer 1. Therefore, within the identification framework... Adding an additional uncertainty, the BPA assigned to the uncertainty of this body of evidence is: (20) Step 9.3, from which we can obtain the first... Each underlying evaluation metric in the identification framework The following BPAs are respectively , , , , , .

[0048] Step 10 includes the following steps: Step 10.1: Merge the weighted BPA of the underlying evaluation indicators obtained in Step 9 to obtain the BPA distribution of their respective superior evaluation indicators. Let there be two weighted BPA functions of the underlying evaluation indicators, m1 and m2, with their corresponding focal elements being m1 and m2 respectively. and The Dempster composition rule can then be expressed as follows: (twenty one) In the formula, K is used to measure the degree of conflict between pieces of evidence, and the conflict coefficient K is defined as: (twenty two) Step 10.2: Based on the BPA distribution of the superior evaluation indicators to which the lower-level evaluation indicators belong in Step 10.1, the Dempster synthesis rule is also used to fuse them. Step 10.3: Repeat step 10.2 until the fusion yields the recognition framework. The following is a BPA distribution regarding the deformation safety state of concrete dams.

[0049] The underlying evaluation indicator is the monitoring point, and its superior evaluation indicator is the monitoring instrument.

Claims

1. A comprehensive evaluation method for the deformation safety of concrete dams, characterized in that: Includes the following steps: Step 1: Based on the layout of the concrete dam deformation monitoring system, select sensor measurement points for deformation safety evaluation, and determine one of the unfavorable working conditions of the concrete dam during its operation as the basis for subsequent analysis. Step 2: Select deformation safety evaluation indicators based on the actual engineering characteristics of concrete dams and relevant data, and construct a comprehensive evaluation index system for deformation safety of concrete dams; Step 3: Determine the identification framework for the deformation safety state of the concrete dam. ; Step 4: Calculate the subjective weights of various bottom-level evaluation indicators in the comprehensive evaluation index system for deformation safety of concrete dams using the G1 method; Step 5: Calculate the objective weights of various bottom-level evaluation indicators in the comprehensive evaluation index system for deformation safety of concrete dams using the CRITIC method; Step 6: Establish the G1-CRITIC-GT combined weighting model to obtain the subjective and objective combined weights of various bottom-level evaluation indicators in the comprehensive evaluation index system for deformation safety of concrete dams, and quantify the importance of different bottom-level evaluation indicators in multi-source information fusion. Step 7: Based on the environmental data and deformation monitoring data of the concrete dam, establish a statistical model between the deformation effect of each underlying evaluation index and different loads using the stepwise regression method, and formulate the identification framework for each underlying evaluation index using the confidence interval method. The deformation monitoring threshold is used to divide its safe state range; Step 8: Obtain the basic probability distribution of various underlying evaluation indicators, i.e., BPA, through the interval number method, as the original evidence; Step 9: Use the subjective and objective combined weights calculated in Step 6 to weight the BPA of the underlying evaluation index, and then apply this weighting to the identification framework. The introduction of uncertainty allocation may lead to information conflicts; Step 10: Use the improved DS evidence theory to perform fusion diagnosis on the deformation evaluation index of concrete dam, and realize the layer-by-layer fusion of multi-source deformation monitoring information; Step 11: Determine the deformation safety status of the concrete dam based on the principle of maximum confidence.

2. The comprehensive evaluation method for deformation safety of concrete dams as described in claim 1, characterized in that: The identification framework in step 3 It consists of five safety states and one uncertainty state, namely: {normal, basically normal, slightly abnormal, severely abnormal, malignant abnormal, and uncertainty}.

3. The comprehensive evaluation method for deformation safety of concrete dams as described in claim 1, characterized in that: Step 4 includes the following steps: Step 4.1: Invite several experts to rank the various underlying evaluation indicators from highest to lowest importance based on their own experience and subjective judgment. Assume that one type of underlying evaluation indicator includes n indicators, namely... The order of the evaluation indicators obtained after expert ranking is as follows: ; Step 4.2: Several experts score the relative importance of adjacent indicators. For any two adjacent indicators... and The relative importance ratio assigned by experts Represented as: （1）， In the formula: , The larger the value, the better the indicator. Compared to indicators The greater the importance; Step 4.3: Due to the differences in the professional background, engineering experience and authority of different experts, we will comprehensively consider the experts' education, professional title, years of service and familiarity with the field, establish expert information evaluation rules, and score each expert. Step 4.4: Normalize the total scores of each expert and assign weights to each expert; Step 4.5: Calculate the subjective weights of each expert for the underlying evaluation index using the G1 method according to equations (2) and (3). ; （2）， （3）， in, ,and ; Step 4.6: Use the expert weights from Step 4.4 to weight the subjective weights of different experts in Step 4.5 using the G1 method. The sum of the subjective weights of the same underlying evaluation index after being weighted by several experts is the final subjective weight of that underlying evaluation index.

4. The comprehensive evaluation method for deformation safety of concrete dams as described in claim 1, characterized in that: Step 5 includes the following steps: Step 5.1: Given n samples to be evaluated and p evaluation indicators, form the original indicator data matrix. : （4）， In the formula, This represents the value of the i-th sample on the j-th evaluation index; Step 5.2: Process the original indicator data matrix. Perform dimensionless processing and construct a standardized matrix. ; Step 5.3: Calculate the normalized matrix using the following formula. Standard deviation To determine the volatility among the various evaluation indicators. （5）， （6）， In the formula, Let j be the average value of the data for the j-th evaluation indicator; Step 5.4: Use the correlation coefficient between the i-th evaluation index and the j-th evaluation index. To measure the conflict between various evaluation indicators. ; （7）， （8）， Step 5.5, based on volatility and conflict Calculate information carrying capacity ; （9）， Step 5.6: Take the proportion of the information content of the j-th evaluation indicator to the total information content as the objective weight of that evaluation indicator. , （10）， Step 5.7: From this, the objective weights of various underlying evaluation indicators can be obtained sequentially using the CRITIC method.

5. The comprehensive evaluation method for deformation safety of concrete dams as described in claim 1, characterized in that: Step 6 includes the following steps: Step 6.1: Suppose that L different weighting methods are used to calculate L weight vectors, and each weight vector is represented as: （11）， In the formula, n represents the number of evaluation indicators. Step 6.2, Let any linear combination of these L vectors be: （12）， In the formula, The linear combination coefficients of the weights obtained by the k-th weighting method are... For a given possible weight vector; Step 6.3: Using game theory, optimize the linear combination coefficients of each indicator to make... and Minimize the deviation between: （13）， Step 6.4: Solve the equivalent linear equation system to obtain the optimal linear combination coefficient vector: （14）， Step 6.5: For the optimal linear combination coefficient vector Perform a normalization operation to obtain the normalized combination coefficients. and the corresponding combined weighted vectors : （15）， （16）。 6. The comprehensive evaluation method for deformation safety of concrete dams as described in claim 1, characterized in that: Step 8 includes the following steps: Step 8.1: Determine the measured values ​​of each underlying evaluation index under unfavorable working conditions as determined in Step 1, and construct an interval number model corresponding to the measured value, denoted as . ; Step 8.2: Based on the underlying evaluation indicators obtained in Step 7, in the identification framework... The safe state range below and the interval number model in step 8.1 Calculate the distance between two intervals. ; Step 8.3: Based on the distance between the two interval numbers in Step 8.2 Calculate the similarity between interval numbers ; Step 8.4: Normalize the similarity results between the interval numbers in Step 8.3 to generate the BPA of each underlying evaluation index.

7. The comprehensive evaluation method for deformation safety of concrete dams as described in claim 6, characterized in that: The distance between the two intervals in step 8.2 The following formula is used for calculation: （17）， The similarity between interval numbers in step 8.3 The following formula is used for calculation: （18） In the formula, the support coefficient The value is usually 5.

8. The comprehensive evaluation method for deformation safety of concrete dams as described in claim 1, characterized in that: Step 9 includes the following steps: Step 9.1, let the first... Individual evidence within the identification framework The BPA function inside is The corresponding focal elements are respectively The BPA function, weighted by the subjective and objective combination weights calculated in step 6, is expressed as follows: （19） Step 9.2: After weight allocation as in Step 9.1, the sum of the BPAs of each piece of evidence is no longer 1. Therefore, within the identification framework... Adding an additional uncertainty, the BPA assigned to the uncertainty of this body of evidence is: （20） Step 9.3, from which we can obtain the first... Each underlying evaluation metric in the identification framework The following BPAs are respectively , , , , , .

9. The comprehensive evaluation method for deformation safety of concrete dams as described in claim 1, characterized in that: Step 10 includes the following steps: Step 10.1: Merge the weighted BPA of the underlying evaluation indicators obtained in Step 9 to obtain the BPA distribution of their respective superior evaluation indicators. Let there be two weighted BPA functions of the underlying evaluation indicators, m1 and m2, with their corresponding focal elements being m1 and m2 respectively. and The Dempster composition rule can then be expressed as follows: （21） In the formula, K is used to measure the degree of conflict between pieces of evidence, and the conflict coefficient K is defined as: （22） Step 10.2: Based on the BPA distribution of the superior evaluation indicators to which the lower-level evaluation indicators belong in Step 10.1, the Dempster synthesis rule is also used to fuse them. Step 10.3: Repeat step 10.2 until the fusion yields the recognition framework. The following is a BPA distribution regarding the deformation safety state of concrete dams.

10. The comprehensive evaluation method for deformation safety of concrete dams as described in claim 1, characterized in that: The underlying evaluation indicator is the monitoring point, and its superior evaluation indicator is the monitoring instrument.

Images