Method and system for evaluating anti-seismic performance of dam project

Abstract

The invention discloses a dam project anti-seismic performance evaluation method and system, and belongs to the technical field of geological data, structural analysis and computer simulation. The problems of inaccuracy of model establishment, one-sidedness of anti-seismic performance evaluation and complexity of seismic action caused by simplified hypothesis of a traditional anti-seismic analysis model in calculation results are solved. The method comprises the following steps: classifying anti-seismic performance key variables of dam engineering, analyzing the influence of different types of key variables on dam performance, and performing basic analysis on the key variables; on this basis, an anti-seismic performance evaluation model based on a deep learning algorithm is established, and a plurality of groups of experimental models of different key variable combinations are generated; through multiple combination arrangements, construction of an evaluation model and analysis of performance evaluation precision, the influence weight of a single key variable on the anti-seismic performance of the dam project can be accurately obtained, the deviation between the model and the actual situation is reduced, and the anti-seismic performance of the dam project can be more accurately predicted.

Description

Technical Field

[0001] The invention belongs to the technical field of geological data, structural analysis and computer simulation, and specifically relates to a method and system for evaluating the seismic performance of a dam project. Background Art

[0002] As an important part of water conservancy projects, dam projects play a vital role in water storage, flood control, power generation, etc. However, natural disasters such as earthquakes pose severe challenges to the safety performance of dam projects. The seismic performance of dam projects refers to their ability to maintain stability and integrity under earthquakes, which is directly related to the safety of life and property of people in downstream areas and the prevention of secondary disasters. Therefore, comprehensively improving the seismic performance of dam projects and ensuring their safe operation under earthquake conditions is an important task in the current field of water conservancy construction.

[0003] The defects in the calculation results of traditional seismic analysis models mainly include the inaccuracy of model establishment caused by simplified assumptions, the one-sidedness of seismic performance evaluation and the complexity of earthquake effects; these defects require necessary corrections and improvements to the traditional models in practical applications, while actively developing and applying modern seismic analysis techniques to improve the scientificity and accuracy of structural seismic design; through continuous research and practice, we can better understand and respond to the impact of earthquakes on structures. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a method for evaluating the seismic performance of dam projects, which is used to solve the defects of traditional seismic analysis models in calculation results, mainly including the limitations of simplified assumptions, inadequacies in calculation methods, inaccuracy in model establishment, one-sidedness of seismic performance evaluation, lag in standards and methods, disconnection between numerical simulation and experimental verification, and complexity of earthquake effects.

[0005] To solve the above problems, the first aspect of the present invention provides a method for evaluating the seismic performance of a dam project, comprising the following steps:

[0006] S1. Investigate the technical methods for evaluating the seismic performance of relevant dam projects, and establish a key variable set for extracting the characteristics of dam projects; based on the key variable set, collect relevant earthquake case data and generate a project data set; perform data cleaning on the project data set;

[0007] S2. Classify the key variables of the seismic performance of dam projects, analyze the impact of different categories of key variables on dam performance, and conduct basic analysis of key variables; establish a dam project seismic performance evaluation model based on the fusion of multiple key variables based on deep learning algorithms, generate several groups of experimental models with different key variable combinations, and analyze the impact weights of different key variables on seismic performance;

[0008] S3. Define the seismic performance gradients of dam projects and calibrate the evaluation criteria for each gradient.

[0009] As a further solution of the present invention: the step S1, generating an engineering data set and cleaning the data, includes the following steps:

[0010] The steps of generating the engineering data set include: retrieving the key variables in the historical cases and the earthquake cases involving the key variables, counting the earthquake attribute information of each earthquake case, the key variable data of the corresponding key variable set of the building and the degree of damage to the building

[0011] The data cleaning includes removing redundant and repeated information in the data; filling or correcting missing values ​​and outliers in the data; standardizing data from different sources into a unified format to ensure compliance with predetermined specifications; performing necessary data conversion and processing based on business needs; extracting keywords, vocabulary and semantic information from text data and converting it into structured data form.

[0012] As a further solution of the present invention: Step S2, classifying key variables, analyzing the impact of key variables of different categories on dam performance, and performing basic analysis on key variables, includes the following steps:

[0013] The key variables include: strain, allowable stress, dynamic strain, allowable dynamic stress, initial strain, allowable creep strain, elastic modulus and elastic modulus after environmental factors, as well as time variation, temperature and humidity, analysis of stress-strain relationship, dynamic effects, environmental effects and long-term performance;

[0014] Analyze key variables and generate key variable coefficients:

[0015] Based on strain, elastic modulus and allowable stress, stress-strain analysis includes:

[0016] σ=E×ε;

[0017] Where σ is stress, E is elastic modulus, and ε is strain;

[0018] Stress coefficient analysis includes:

[0019]

[0020] Set the stress qualification coefficient. The setting method is as follows: set multiple groups of data, each group of data includes actual stress and allowable stress, and the actual stress increases with the corresponding data of the adjacent groups in a gradient manner. Calculate the stress coefficient of each group of data, and take the stress coefficient corresponding to the actual stress greater than the minimum stress standard of the dam seismic resistance as the stress qualification coefficient sc0;

[0021] Where sc is the stress coefficient, δ is the actual stress, δ allow is the allowable stress;

[0022] Dynamic response: σ_dynamic=E×ε_dynamic, where σ_dynamic represents dynamic stress and ε_dynamic represents dynamic strain; dynamic response analysis includes:

[0023]

[0024] Among them, σ allow is the allowable dynamic stress, ac is the dynamic stress coefficient;

[0025] Set the dynamic response qualification coefficient, the setting method is: set multiple groups of data, each group of data includes dynamic stress and allowable dynamic stress, and the dynamic stress is increased with the corresponding data of the adjacent group, calculate the dynamic stress coefficient of each group of data, and take the dynamic stress coefficient corresponding to the dynamic stress greater than the minimum dynamic stress standard of the dam seismic resistance as the dynamic response qualification coefficient ac0;

[0026] Environmental and humidity influence: E_env=E×g(T,H) where E_env represents the elastic modulus affected by environmental factors, and g(T,H) is a function of temperature T and humidity H.

[0027]

[0028] Where v is the original volume of the material, Δv is the volume change, ΔT is the temperature change, and ΔW is the temperature change; the environmental and humidity influence coefficients are:

[0029]

[0030] Where E allow is the allowable initial elastic modulus; ecc is the environmental and humidity influence coefficient;

[0031] Set the qualified coefficient of environmental and humidity influence. The setting method is: set multiple groups of data, each group of data includes initial elastic modulus and allowable initial elastic modulus, and the initial elastic modulus increases with the corresponding data of the adjacent group in a gradient manner, calculate the environmental and humidity influence coefficient of each group of data, and take the environmental and humidity influence coefficient corresponding to the initial elastic modulus greater than the minimum stress standard of dam seismic resistance in the data as the qualified coefficient of environmental and humidity influence ecc0;

[0032] Long-term performance: -creep =ε_0×ΔTwhere, ξ allow represents creep strain, ε_0 represents initial strain, ΔT is the time change; the long-term performance coefficient is:

[0033]

[0034] where ξ allow is the permissible creep strain, etc is the long-term performance coefficient;

[0035] Set the long-term performance qualification coefficient. The setting method is: set multiple groups of data, each group of data includes creep strain and allowable creep strain, and the creep strain increases gradiently with the corresponding data of the adjacent groups, calculate the long-term performance coefficient of each group of data, and use the long-term performance coefficient corresponding to the creep strain greater than the minimum seismic stress standard of the dam in the data as the long-term performance qualification coefficient etc0.

[0036] As a further solution of the present invention: the step S2, establishing a dam engineering seismic performance evaluation model based on a deep learning algorithm and integrating multiple key variables, generating several groups of experimental models with different key variable combinations, and analyzing the influence weights of different key variables on seismic performance, comprises the following steps:

[0037] Establishing a dam engineering seismic performance evaluation model based on deep learning algorithm and fusion of multiple key variables includes the following steps:

[0038] According to the engineering data set, key variables are classified, a separate folder is set for each category of data and sorted, and the training data set for the dam engineering seismic performance evaluation model is constructed, and then cleaned and preprocessed;

[0039] Select several key variables, and input the data in the folder corresponding to the selected key variables in the training data set into the dam project seismic performance assessment model, execute the multivariate regression analysis algorithm, input the selected key variables of the dam project, and generate key variable coefficients, where the key variables include: strain, allowable stress, dynamic strain, allowable dynamic stress, initial strain, allowable creep strain, elastic modulus and elastic modulus after being affected by environmental factors, as well as time variation, temperature and humidity;

[0040] Comprehensively analyze the key variable coefficients to generate the comprehensive performance index P:

[0041] P=(sc-sc0)+(ac-ac0)+(ecc-ecc0)+(etc-etc0);

[0042] On this basis, several groups of experimental models with different key variable combinations are established to analyze the impact of a single key variable on the seismic performance of the dam project;

[0043] As a further solution of the present invention: the analysis of the influence of a single key variable on the seismic performance of a dam project comprises the following steps:

[0044] Two experimental groups, A and B, are set up. In the experimental group A, the steps of establishing a dam project seismic performance evaluation model based on deep learning algorithm with multiple key variables fused are repeated by arranging and combining key variables, and key variables are selected to build a dam project seismic performance evaluation model with multiple key variables fused. The number of key variables included in a single dam project seismic performance evaluation model is greater than or equal to two and less than or equal to three.

[0045] Experimental group B established a dam engineering seismic performance assessment model integrating multiple key variables, including all key variables;

[0046] Comprehensively analyze the two experimental groups A and B to generate the impact weight of each key variable.

[0047] As a further solution of the present invention: comprehensively analyzing the two experimental groups A and B to generate the influence weight of each key variable, including the following steps:

[0048] Compare all the seismic performance assessment models of dam projects in Group A and Group B. For a single key variable, count all the seismic performance assessment models of dam projects that contain the key variable, record them as one group; count all the seismic performance assessment models of dam projects that do not contain the variable, record them as Group B; according to the principle of controlling variables, find the only corresponding seismic performance assessment model of dam projects in Group B for all the seismic performance assessment models of dam projects in Group A;

[0049] Among them, the key variables included in the dam engineering seismic performance assessment model corresponding to the principle group one are the only ones that are different from the dam engineering seismic performance assessment model corresponding to the second group;

[0050] Calculate the absolute value of the difference in the comprehensive performance index P between the corresponding two dam project seismic performance evaluation models, and calculate the mean of all absolute values, which is recorded as the percentage value; repeat the above steps to calculate the percentage values ​​of all key variables, analyze the relative proportional relationship between the corresponding percentage values ​​of the key variables, and then derive the influence weight of each key variable.

[0051] As a further solution of the present invention: the step S3, defining the seismic performance gradient of the dam project and correcting the evaluation criteria of each gradient, comprises the following steps:

[0052] The seismic performance of dam projects is divided into three levels: qualified, unqualified and extremely unqualified;

[0053] Among the key variables, the coefficients of the key variables are all greater than or equal to the corresponding qualified coefficients and are recorded as qualified;

[0054] For key variables with an impact weight less than c, if the key variable coefficient involved in the calculation is less than the corresponding qualified coefficient, it will be recorded as unqualified;

[0055] For key variables with an impact weight greater than or equal to c, if the key variable coefficient involved in the calculation is less than the corresponding qualified coefficient, it will be recorded as extremely unqualified;

[0056] Qualified standard calibration, each individual key variable is divided into qualified coefficients, the division steps are as follows:

[0057] According to the weight of each key variable's influence on the seismic performance of the dam project, when calculating a single key variable, all historical cases are counted in which the building is not affected in actual use except for the appearance after the earthquake, and the coefficient of the key variable is calculated. The mean is calculated and the variance is defined. If the variance is greater than the limit variance, the maximum and minimum values ​​of the key variable coefficient in the case are removed, and the above operation is repeated until the calculated variance is less than or equal to the limit variance. At this time, the key variable coefficient is the qualified coefficient corresponding to the key variable.

[0058] As another aspect of the present invention: a system for evaluating the seismic performance of a dam project, the system specifically adopts a method for evaluating the seismic performance of a dam project to implement the evaluation of the seismic performance of the dam project.

[0059] Compared with the prior art, the present invention has the following beneficial effects:

[0060] The present invention classifies the key variables of the seismic performance of dam projects, analyzes the influence of key variables of different categories on dam performance, and performs basic analysis on key variables; on this basis, establishes a seismic performance evaluation model of dam projects based on the fusion of multiple key variables of deep learning algorithm, and generates several groups of experimental models with different key variable combinations; through multiple combinations and arrangements, the evaluation model is constructed and the performance evaluation accuracy is analyzed, so that the influence weight of a single key variable on the seismic performance of the dam project can be accurately obtained, and a seismic performance evaluation of the dam project that is closer to reality can be obtained, which reduces the deviation between the model and the actual situation, reduces the reliance on simplified assumptions, enhances the generalization ability of the model, and can more accurately predict the seismic performance of the dam project;

[0061] At the same time, a control variable reference model of the dam project seismic performance assessment model based on the fusion of key variables was established, and the influence of different influencing characteristics on the seismic performance of the dam project was compared, effectively responding to the challenges brought by the complexity of earthquakes;

[0062] Investigate relevant technical methods for evaluating the seismic performance of dam projects, establish a key variable set for extracting dam project characteristics, collect relevant earthquake case data, generate an engineering data set, and then perform data cleaning; by summarizing the existing data, quickly locate the key variables that have an impact on the performance of dam projects, and find and fill in the gaps, so as to make a more comprehensive assessment of the seismic performance of dam projects and avoid missing key factors that may lead to large errors in the performance assessment of dam projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0064] Figure 1 It is a schematic diagram of the process of the present invention;

[0065] Figure 2 It is a schematic diagram of the system framework of the present invention. DETAILED DESCRIPTION

[0066] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0067] See also Figure 1-Figure 2 The first embodiment of the present invention provides a method for evaluating the seismic performance of a dam project, comprising the following steps:

[0068] S1. Investigate the technical methods for evaluating the seismic performance of relevant dam projects, and establish a key variable set for extracting the characteristics of dam projects; based on the key variable set, collect relevant earthquake case data and generate a project data set; perform data cleaning on the project data set;

[0069] S2. Classify the key variables of the seismic performance of dam projects, analyze the impact of different categories of key variables on dam performance, and conduct basic analysis of key variables; establish a dam project seismic performance evaluation model based on the fusion of multiple key variables based on deep learning algorithms, generate several groups of experimental models with different key variable combinations, and analyze the impact weights of different key variables on seismic performance;

[0070] S3. Define the seismic performance gradients of dam projects and calibrate the evaluation criteria for each gradient.

[0071] By classifying the key variables of the seismic performance of dam projects, analyzing the impact of different categories of key variables on dam performance, and conducting basic analysis of key variables; on this basis, establishing a dam project seismic performance evaluation model based on the fusion of multiple key variables based on deep learning algorithms, generating several groups of experimental models with different key variable combinations; through a variety of combinations and arrangements, constructing evaluation models and analyzing performance evaluation accuracy, the influence weight of a single key variable on the seismic performance of dam projects can be accurately obtained, and a seismic performance evaluation of dam projects that is closer to reality can be obtained, which reduces the deviation between the model and the actual situation, reduces reliance on simplified assumptions, enhances the generalization ability of the model, and can more accurately predict the seismic performance of dam projects;

[0072] As a further solution of the present invention: the step S1, generating an engineering data set and cleaning the data, includes the following steps:

[0073] The step of generating the engineering data set includes: retrieving key variables in historical cases and earthquake cases involving key variables, counting earthquake attribute information of each earthquake case, key variable data of the corresponding key variable set of the building, and the degree of damage to the building;

[0074] Investigate the technical methods for evaluating the seismic performance of relevant dam projects, establish a key variable set for extracting dam project characteristics, collect relevant earthquake case data, generate engineering data sets, and then perform data cleaning; by summarizing the existing data, quickly locate the key variables that affect the performance of dam projects, and check for deficiencies and fill in the gaps, so as to make the evaluation of the seismic performance of dam projects more comprehensive and avoid missing key factors that may cause large errors in the performance evaluation of dam projects;

[0075] The data cleaning includes removing redundant and repeated information in the data; filling or correcting missing values ​​and outliers in the data; standardizing data from different sources into a unified format to ensure compliance with predetermined specifications; performing necessary data conversion and processing based on business needs; extracting keywords, vocabulary and semantic information from text data and converting it into structured data form.

[0076] As a further solution of the present invention: Step S2, classifying key variables, analyzing the impact of key variables of different categories on dam performance, and performing basic analysis on key variables, includes the following steps:

[0077] The key variables include: strain, allowable stress, dynamic strain, allowable dynamic stress, initial strain, allowable creep strain, elastic modulus and elastic modulus after environmental factors, as well as time variation, temperature and humidity, analysis of stress-strain relationship, dynamic effects, environmental effects and long-term performance;

[0078] Analyze key variables and generate key variable coefficients:

[0079] Based on strain, elastic modulus and allowable stress, stress-strain analysis includes:

[0080] σ=E×ε;

[0081] Where σ is stress, E is elastic modulus, and ε is strain;

[0082] Stress coefficient analysis includes:

[0083]

[0084] Set the stress qualification coefficient. The setting method is: set multiple groups of data, each group of data includes actual stress and allowable stress, and the actual stress increases gradually with the corresponding data of the adjacent groups. Calculate the stress coefficient of each group of data, and use the stress coefficient corresponding to the actual stress in the data that is greater than the minimum stress standard for dam seismic resistance as the stress qualification coefficient sc0; here the stress qualification coefficient is 0.8.

[0085] Where sc is the stress coefficient, δ is the actual stress, δ allow is the allowable stress;

[0086] Example: δ is 85, δ allow is 100, then sc is 0.85.

[0087] Dynamic response: σ_dynamic = E × ε_dynamic, where σ_dynamic represents dynamic stress and ε_dynamic represents dynamic strain, both of which can be obtained by real-time monitoring of relevant testing equipment; dynamic response analysis includes:

[0088]

[0089] Among them, σ allow is the allowable dynamic stress, ac is the dynamic stress coefficient;

[0090] Example: σ allow is 82, σ_dynamic is 108, then ac is 0.76.

[0091] The dynamic response qualification coefficient is set as follows: multiple groups of data are set, each group of data includes dynamic stress and allowable dynamic stress, and the dynamic stress is increased gradually with the corresponding data of the adjacent groups, and the dynamic stress coefficient of each group of data is calculated, and the dynamic stress coefficient corresponding to the dynamic stress in the data that is greater than the minimum dynamic stress standard for dam seismic resistance is used as the dynamic response qualification coefficient ac0; here the dynamic response qualification coefficient is 0.75.

[0092] Environmental and humidity influence: E_env=E×g(T,H) where E_env represents the elastic modulus affected by environmental factors, and g(T,H) is a function of temperature T and humidity H.

[0093]

[0094] Where v is the original volume of the material, Δv is the volume change, ΔT is the temperature change, and ΔW is the temperature change; the environmental and humidity influence coefficients are:

[0095]

[0096] Where E allow is the allowable initial elastic modulus; ecc is the environmental and humidity influence coefficient;

[0097] Example: E allow is 145, E_env is 200, then ecc is 0.725.

[0098] Set the qualified coefficient of environmental and humidity influence. The setting method is: set multiple groups of data, each group of data includes an initial elastic modulus and an allowable initial elastic modulus, and the initial elastic modulus increases gradually with the corresponding data of the adjacent groups, calculate the environmental and humidity influence coefficient of each group of data, and use the environmental and humidity influence coefficient corresponding to the initial elastic modulus greater than the minimum seismic stress standard of the dam in the data as the qualified coefficient of environmental and humidity influence ecc0; here the environmental and humidity influence coefficient is 0.6.

[0099] Long-term performance: -creep =ε_0×ΔTwhere, ξ allow represents creep strain, ε_0 represents initial strain, ΔT is the time change; the long-term performance coefficient is:

[0100]

[0101] where ξ allow is the permissible creep strain, etc is the long-term performance coefficient;

[0102] Example: ξ allow is 0.75, ξ -creep is 1.1, then etc is 0.68.

[0103] Set the long-term performance qualification coefficient. The setting method is: set multiple groups of data, each group of data includes creep strain and allowable creep strain, and the creep strain increases with the corresponding data of the adjacent group in a gradient manner. Calculate the long-term performance coefficient of each group of data, and take the long-term performance coefficient corresponding to the creep strain greater than the minimum stress standard of the dam seismic resistance in the data as the long-term performance qualification coefficient etc0; here the long-term performance coefficient is 0.9;

[0104] The present invention analyzes the existing dam engineering seismic performance assessment method technology, identifies and classifies the influencing features, and reduces the deviation between the model and the actual situation, reduces the reliance on simplified assumptions, and enhances the generalization ability of the model by learning a large amount of actual data through a model based on a deep learning algorithm.

[0105] As a further solution of the present invention: the step S2, establishing a dam engineering seismic performance evaluation model based on a deep learning algorithm and integrating multiple key variables, generating several groups of experimental models with different key variable combinations, and analyzing the influence weights of different key variables on seismic performance, comprises the following steps:

[0106] Establishing a dam engineering seismic performance evaluation model based on deep learning algorithm and fusion of multiple key variables includes the following steps:

[0107] According to the engineering data set, key variables are classified, a separate folder is set for each category of data and sorted, and the training data set for the dam engineering seismic performance evaluation model is constructed, and then cleaned and preprocessed;

[0108] Select several key variables, and input the data in the folder corresponding to the selected key variables in the training data set into the dam project seismic performance assessment model, execute the multivariate regression analysis algorithm, input the selected key variables of the dam project, and generate key variable coefficients, where the key variables include: strain, allowable stress, dynamic strain, allowable dynamic stress, initial strain, allowable creep strain, elastic modulus and elastic modulus after being affected by environmental factors, as well as time variation, temperature and humidity;

[0109] Comprehensively analyze the key variable coefficients to generate the comprehensive performance index P:

[0110] P=(sc-sc0)+(ac-ac0)+(ecc-ecc0)+(etc-etc0);

[0111] On this basis, several groups of experimental models with different key variable combinations are established to analyze the impact of a single key variable on the seismic performance of the dam project;

[0112] As a further solution of the present invention: the analysis of the influence of a single key variable on the seismic performance of a dam project comprises the following steps:

[0113] Two experimental groups, A and B, are set up. In the experimental group A, the steps of establishing a dam project seismic performance evaluation model based on deep learning algorithm with multiple key variables fused are repeated by arranging and combining key variables, and key variables are selected to build a dam project seismic performance evaluation model with multiple key variables fused. The number of key variables included in a single dam project seismic performance evaluation model is greater than or equal to two and less than or equal to three.

[0114] Experimental group B established a dam engineering seismic performance assessment model integrating multiple key variables, including all key variables;

[0115] Comprehensively analyze the two experimental groups A and B to generate the impact weight of each key variable.

[0116] As a further solution of the present invention: the comprehensive analysis of the two experimental groups A and B to generate the influence weight of each key variable includes the following steps:

[0117] Compare all the seismic performance assessment models of dam projects in Group A and Group B. For a single key variable, count all the seismic performance assessment models of dam projects that contain the key variable, record them as one group; count all the seismic performance assessment models of dam projects that do not contain the variable, record them as Group B; according to the principle of controlling variables, find the only corresponding seismic performance assessment model of dam projects in Group B for all the seismic performance assessment models of dam projects in Group A;

[0118] Among them, the key variables included in the dam engineering seismic performance assessment model corresponding to the principle group one are the only ones that are different from the dam engineering seismic performance assessment model corresponding to the second group;

[0119] Calculate the absolute value of the difference in the comprehensive performance index P between the corresponding two dam engineering seismic performance evaluation models, and calculate the mean of all absolute values, which is recorded as the percentage value; repeat the above steps to calculate the percentage values ​​of all key variables, analyze the relative proportional relationship between the percentage values ​​corresponding to the key variables, and then obtain the impact weight of each key variable;

[0120] A control variable reference model of the dam project seismic performance assessment model based on the integration of key variables was established, and the influence of different influencing characteristics on the seismic performance of the dam project was compared, effectively responding to the challenges brought by the complexity of earthquakes;

[0121] As a further solution of the present invention: the step S3, defining the seismic performance gradient of the dam project and correcting the evaluation criteria of each gradient, comprises the following steps:

[0122] The seismic performance of dam projects is divided into three levels: qualified, unqualified and extremely unqualified;

[0123] Among the key variables, the coefficients of the key variables are all greater than or equal to the corresponding qualified coefficients and are recorded as qualified;

[0124] For key variables with an impact weight less than c, if the key variable coefficient involved in the calculation is less than the corresponding qualified coefficient, it will be recorded as unqualified;

[0125] For key variables with an impact weight greater than or equal to c, if the key variable coefficient involved in the calculation is less than the corresponding qualified coefficient, it will be recorded as extremely unqualified; the value of c is 5%;

[0126] Qualified standard calibration, each individual key variable is divided into qualified coefficients, the division steps are as follows:

[0127] According to the weight of each key variable's influence on the seismic performance of the dam project, when calculating a single key variable, all historical cases are counted in which the building is not affected in actual use except for the appearance after the earthquake, and the coefficient of the key variable is calculated. The mean is calculated and the variance is defined. If the variance is greater than the limit variance, the maximum and minimum values ​​of the key variable coefficient in the case are removed, and the above operation is repeated until the calculated variance is less than or equal to the limit variance. At this time, the key variable coefficient is the qualified coefficient corresponding to the key variable.

[0128] In another embodiment of the present invention, a dam engineering seismic performance assessment system includes:

[0129] Data Integration Unit: Investigate the technical methods for evaluating the seismic performance of relevant dam projects, and establish a key variable set for extracting the characteristics of dam projects; based on the key variable set, collect relevant earthquake case data and generate engineering data sets; perform data cleaning on engineering data sets;

[0130] Basic analysis unit: classify the key variables of the seismic performance of dam projects, analyze the impact of different categories of key variables on dam performance, and conduct basic analysis of key variables;

[0131] Weight analysis unit: Establish a dam project seismic performance evaluation model based on deep learning algorithm and fusion of multiple key variables, generate several groups of experimental models with different key variable combinations, and analyze the influence weights of different key variables on seismic performance;

[0132] Standard correction unit: defines the seismic performance gradient of the dam project and corrects the evaluation standards of each gradient.

[0133] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A method for evaluating the seismic performance of a dam project, characterized in that: The following steps are involved: S1. Investigate the technical methods for evaluating the seismic performance of relevant dam projects, and establish a key variable set for extracting the characteristics of dam projects; based on the key variable set, collect relevant earthquake case data and generate a project data set; perform data cleaning on the project data set; S2. Classify the key variables of the seismic performance of dam projects, analyze the impact of different categories of key variables on dam performance, and conduct basic analysis of key variables; establish a dam project seismic performance evaluation model based on the fusion of multiple key variables based on deep learning algorithms, generate several groups of experimental models with different key variable combinations, and analyze the impact weights of different key variables on seismic performance; S3. Define the seismic performance gradients of dam projects and calibrate the evaluation criteria for each gradient.

2. A method for evaluating the seismic performance of a dam project according to claim 1, characterized in that: The step S1, generating an engineering data set and cleaning the data, includes the following steps: The step of generating the engineering data set includes: retrieving key variables in historical cases and earthquake cases involving key variables, counting earthquake attribute information of each earthquake case, key variable data of the corresponding key variable set of the building, and the degree of damage to the building; The data cleaning includes removing redundant and repeated information in the data; filling or correcting missing values ​​and outliers in the data; standardizing data from different sources into a unified format to ensure compliance with predetermined specifications; performing necessary data conversion and processing based on business needs; extracting keywords, vocabulary and semantic information from text data and converting it into structured data form.

3. A method for evaluating the seismic performance of a dam project according to claim 1, characterized in that: The step S2, classifying the key variables, analyzing the impact of different categories of key variables on dam performance, and performing basic analysis on the key variables, includes the following steps: The key variables include: strain, allowable stress, dynamic strain, allowable dynamic stress, initial strain, allowable creep strain, elastic modulus and elastic modulus after environmental factors, as well as time variation, temperature and humidity, analysis of stress-strain relationship, dynamic effects, environmental effects and long-term performance; Analyze key variables and generate key variable coefficients: Based on strain, elastic modulus and allowable stress, stress-strain analysis includes: σ=E×ε; Where σ is stress, E is elastic modulus, and ε is strain; Stress coefficient analysis includes: Set the stress qualification coefficient. The setting method is as follows: set multiple groups of data, each group of data includes actual stress and allowable stress, and the actual stress increases with the corresponding data of the adjacent groups in a gradient manner. Calculate the stress coefficient of each group of data, and take the stress coefficient corresponding to the actual stress greater than the minimum stress standard of the dam seismic resistance as the stress qualification coefficient sc0; Where sc is the stress coefficient, δ is the actual stress, δ allow is the allowable stress; Dynamic response: σ_dynamic=E×ε_dynamic, where σ_dynamic represents dynamic stress and ε_dynamic represents dynamic strain; dynamic response analysis includes: Among them, σ allow is the allowable dynamic stress, ac is the dynamic stress coefficient; Set the dynamic response qualification coefficient, the setting method is: set multiple groups of data, each group of data includes dynamic stress and allowable dynamic stress, and the dynamic stress is increased with the corresponding data of the adjacent group, calculate the dynamic stress coefficient of each group of data, and take the dynamic stress coefficient corresponding to the dynamic stress greater than the minimum dynamic stress standard of the dam seismic resistance as the dynamic response qualification coefficient ac0; Environmental and humidity influence: E_env=E×g(T,H) where E_env represents the elastic modulus affected by environmental factors, and g(T,H) is a function of temperature T and humidity H. Where v is the original volume of the material, Δv is the volume change, ΔT is the temperature change, and ΔW is the temperature change; the environmental and humidity influence coefficients are: Where E allow is the allowable initial elastic modulus; ecc is the environmental and humidity influence coefficient; Set the qualified coefficient of environmental and humidity influence. The setting method is: set multiple groups of data, each group of data includes initial elastic modulus and allowable initial elastic modulus, and the initial elastic modulus increases with the corresponding data of the adjacent group in a gradient manner, calculate the environmental and humidity influence coefficient of each group of data, and take the environmental and humidity influence coefficient corresponding to the initial elastic modulus greater than the minimum stress standard of dam seismic resistance in the data as the qualified coefficient of environmental and humidity influence ecc0; Long-term performance: -creep =ε_0×ΔTwhere, ξ allow represents creep strain, ε_0 represents initial strain, ΔT is the time change; the long-term performance coefficient is: where ξ allow is the permissible creep strain, etc is the long-term performance coefficient; Set the long-term performance qualification coefficient. The setting method is: set multiple groups of data, each group of data includes creep strain and allowable creep strain, and the creep strain increases gradiently with the corresponding data of the adjacent groups, calculate the long-term performance coefficient of each group of data, and use the long-term performance coefficient corresponding to the creep strain greater than the minimum seismic stress standard of the dam in the data as the long-term performance qualification coefficient etc0.

4. A method for evaluating the seismic performance of a dam project according to claim 1, characterized in that: The step S2 is to establish a dam project seismic performance evaluation model based on a deep learning algorithm and multiple key variable fusion, generate several groups of experimental models with different key variable combinations, and analyze the influence weights of different key variables on seismic performance, including the following steps: Establishing a dam engineering seismic performance evaluation model based on deep learning algorithm and fusion of multiple key variables includes the following steps: According to the engineering data set, key variables are classified, a separate folder is set for each category of data and sorted, and the training data set for the dam engineering seismic performance evaluation model is constructed, and then cleaned and preprocessed; Select several key variables, and input the data in the folder corresponding to the selected key variables in the training data set into the dam project seismic performance assessment model, execute the multivariate regression analysis algorithm, input the selected key variables of the dam project, and generate key variable coefficients, where the key variables include: strain, allowable stress, dynamic strain, allowable dynamic stress, initial strain, allowable creep strain, elastic modulus and elastic modulus after being affected by environmental factors, as well as time variation, temperature and humidity; Comprehensively analyze the key variable coefficients to generate the comprehensive performance index P: P=(sc-sc0)+(ac-ac0)+(ecc-ecc0)+(etc-etc0); On this basis, several groups of experimental models with different combinations of key variables are established to analyze the impact of a single key variable on the seismic performance of the dam project.

5. A method for evaluating the seismic performance of a dam project according to claim 4, characterized in that: The analysis of the influence of a single key variable on the seismic performance of a dam project includes the following steps: Two experimental groups, A and B, are set up. In the experimental group A, the steps of establishing a dam project seismic performance evaluation model based on deep learning algorithm with multiple key variables fused are repeated by arranging and combining key variables, and key variables are selected to build a dam project seismic performance evaluation model with multiple key variables fused. The number of key variables included in a single dam project seismic performance evaluation model is greater than or equal to two and less than or equal to three. Experimental group B established a dam engineering seismic performance assessment model integrating multiple key variables, including all key variables; Comprehensively analyze the two experimental groups A and B to generate the impact weight of each key variable.

6. A method for evaluating the seismic performance of a dam project according to claim 5, characterized in that: The comprehensive analysis of the two experimental groups A and B to generate the impact weight of each key variable includes the following steps: Compare all the seismic performance assessment models of dam projects in Group A and Group B. For a single key variable, count all the seismic performance assessment models of dam projects that contain the key variable, record them as one group; count all the seismic performance assessment models of dam projects that do not contain the variable, record them as Group B; according to the principle of controlling variables, find the only corresponding seismic performance assessment model of dam projects in Group B for all the seismic performance assessment models of dam projects in Group A; Among them, the key variables included in the dam engineering seismic performance assessment model corresponding to the principle group one are the only ones that are different from the dam engineering seismic performance assessment model corresponding to the second group; Calculate the absolute value of the difference in the comprehensive performance index P between the corresponding two dam project seismic performance evaluation models, and calculate the mean of all absolute values, which is recorded as the percentage value; repeat the above steps to calculate the percentage values ​​of all key variables, analyze the relative proportional relationship between the corresponding percentage values ​​of the key variables, and then derive the influence weight of each key variable.

7. A method for evaluating the seismic performance of a dam project according to claim 1, characterized in that: The step S3, defining the seismic performance gradient of the dam project and correcting the evaluation criteria of each gradient, includes the following steps: The seismic performance of dam projects is divided into three levels: qualified, unqualified and extremely unqualified; Among the key variables, the coefficients of the key variables are all greater than or equal to the corresponding qualified coefficients and are recorded as qualified; For key variables with an impact weight less than c, if the key variable coefficient involved in the calculation is less than the corresponding qualified coefficient, it will be recorded as unqualified; For key variables with an impact weight greater than or equal to c, if the key variable coefficient involved in the calculation is less than the corresponding qualified coefficient, it will be recorded as extremely unqualified; Qualified standard calibration, each individual key variable is divided into qualified coefficients, the division steps are as follows: According to the weight of each key variable's influence on the seismic performance of the dam project, when calculating a single key variable, all historical cases are counted in which the building is not affected in actual use except for the appearance after the earthquake, and the coefficient of the key variable is calculated. The mean is calculated and the variance is defined. If the variance is greater than the limit variance, the maximum and minimum values ​​of the key variable coefficient in the case are removed, and the above operation is repeated until the calculated variance is less than or equal to the limit variance. At this time, the key variable coefficient is the qualified coefficient corresponding to the key variable.

8. A dam engineering seismic performance evaluation system, the system specifically adopts the dam engineering seismic performance evaluation method as described in any one of claims 1 to 5 to realize the evaluation of the seismic performance of the dam engineering.

Images