Collaboratively driven hydraulic structure system failure path identification and calculation method and system

Abstract

This invention relates to a collaboratively driven method and system for identifying and calculating failure paths in hydraulic engineering systems. The method includes acquiring multi-source measured information and determining the statistical characteristics of random variables; constructing a finite element model based on these characteristics; generating pairs of random variables using a stochastic method, obtaining the structural response of the hydraulic structure under different random variable settings through finite element simulation, and calculating event reliability based on the structural response; identifying failure paths and modes based on the event reliability results using an improved branch-and-bound method, fitting the response surface function for each failure mode, and calculating the system failure probability based on the response surface function; considering the correlation between various failure modes, calculating the overall system failure probability using a matrix method. This invention integrates multi-source measured information, constructs a finite element model, identifies failure paths based on event reliability, and calculates failure probabilities by comprehensively considering various failure modes, making the failure probability closely resemble real-world operating conditions and enhancing the accuracy and comprehensiveness of the calculation.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent diagnosis of hydraulic structures, and in particular to a method and system for identifying and calculating failure paths of a collaboratively driven hydraulic structure system. Background Art

[0002] During long-term operation, hydraulic structures are subject to adverse environmental effects such as freezing and thawing, chemical reactions, and mechanical loads. They may also encounter unexpected situations such as super-standard floods and major earthquakes, which lead to frequent aging diseases and structural performance degradation, thus threatening the service safety of the project. Structural aging is first of all the aging of materials, which further aggravates the attenuation of local structures and overall performance. It is a time-varying irreversible process that leads to a reduction in the safety margin of hydraulic structures. With the increase in the operating time of hydraulic structures in service, dynamic diagnostic research on the safety status of hydraulic structures has received more and more attention.

[0003] Aging diseases have a profound impact on the overall performance of hydraulic structures, which is related to structural diagnosis and is closely related to life prediction. Reasonable analysis of the impact of aging diseases on local and overall performance can help to gain insight into the working state and evolution of hydraulic structures. Systematic research based on multi-source measured data obtained from field tests and monitoring is of great significance for the long-term service and life cycle management of hydraulic structures. At present, there are obvious defects in the existing technology for hydraulic structures with continuous structures with high redundancy: the lack of an adaptive failure path search method makes it difficult to accurately track the occurrence process and diffusion path of system failure, which makes preventive maintenance and risk control lack key support; the traditional system reliability limit method has poor accuracy. When calculating the reliability of hydraulic structure systems, it cannot meet the current high-precision engineering requirements, which can easily lead to reliability assessment deviations and interfere with the accuracy of subsequent engineering decisions and safety assessments. In addition, the advantages of reliability analysis and finite element analysis have not yet been integrated. There is a lack of parameter statistical methods that can reflect the structural evolution of hydraulic structures, as well as failure mode identification and probability calculation methods driven by data-mechanism collaboration. It is difficult to integrate multi-dimensional information to achieve accurate control of the failure of hydraulic structure systems under complex operating conditions.

[0004] Therefore, the identification and calculation of failure paths of collaboratively driven hydraulic structures are still blank, the various analysis links are separated from each other, and the level of safety management throughout the entire life cycle is limited. It is urgent to propose a collaboratively driven failure path identification and calculation method for hydraulic structure systems to overcome the above technical defects. Summary of the invention

[0005] To this end, the technical problem to be solved by the present invention is to overcome the technical defects existing in the prior art, and propose a collaboratively driven failure path identification and calculation method and system for hydraulic structure systems, which integrates multi-source measured information, constructs a finite element model through the statistical characteristics of random variables, identifies failure paths and modes based on event reliability, and calculates the failure probability of the system by considering the correlation of each failure mode. It can intuitively present the failure law, make the system failure probability fit the actual situation, and enhance the accuracy and comprehensiveness of the calculation.

[0006] In order to solve the above technical problems, the present invention provides a collaboratively driven hydraulic structure system failure path identification and calculation method, comprising:

[0007] Acquire multi-source measured information, and determine the statistical characteristics of random variables of the multi-source measured information;

[0008] Construct a finite element model based on the statistical characteristics of random variables of multi-source measured information;

[0009] Based on the finite element model and the statistical characteristics of random variables, a random method is used to generate random variable set pairs, and the structural response of the physical structure of the hydraulic structure under different random variable settings is obtained through finite element simulation according to the random variable set pairs, and the event reliability is calculated according to the structural response;

[0010] The improved branch and bound method is used to identify failure paths and modes based on event reliability results. After the failure path is identified, the response surface function of each failure mode is obtained by fitting the reliability data of the failure path, and the failure probability of the hydraulic structure system is calculated based on the response surface function.

[0011] Using the identified failure modes and their corresponding response surface functions, and considering the correlation of each failure mode, the overall failure probability of the hydraulic structure system is calculated by the matrix method.

[0012] In one embodiment of the present invention, a method for constructing a finite element model based on the statistical characteristics of random variables of multi-source measured information includes:

[0013] Establish an original text file containing nodes, lines, surface units and their collections, generate a finite element mesh model through a mesh divider, divide the collection according to the constitutive relationship, boundary conditions and load conditions, and then determine the boundary conditions, clarify the parameters of each area of ​​the hydraulic structure system and its loading method, and complete the establishment of the finite element execution file;

[0014] Based on the statistical characteristics of random variables of multi-source measured information, the GAWK program is called to modify the finite element execution file and generate the control file.

[0015] In one embodiment of the present invention, a method for calculating event reliability according to the structural response comprises:

[0016] Constructing limit state function based on structural response , where and are damage parameter and damage threshold respectively;

[0017] Calculation of limit state functions The normalized negative gradient vector is:

[0018] ;

[0019] In the formula, Indicates A normalized negative gradient vector, represents the gradient operator, Indicates A limit state function, represents the design points in the space of uncorrelated standard normal random variables, represents the norm of a vector;

[0020] Event correlation is determined based on the normalized negative gradient vector, and event reliability is obtained by combining the reliability analysis method.

[0021] In one embodiment of the present invention, a method for identifying failure paths and modes based on event reliability results using an improved branch and bound method includes:

[0022] Initialize the upper bound and lower bound , perform finite element analysis and stress state analysis on the structural model of the hydraulic structure system and determine the initial nodes;

[0023] The structural model corresponding to the initial node after damage is corrected, and the finite element analysis and stress state analysis are performed on the corrected structural model to determine the first batch of sub-nodes;

[0024] Calculate the event failure probability of each sub-node, and further obtain the failure probability of the hydraulic structure system;

[0025] The limit is determined based on the event failure probability and the system failure probability, and the upper limit is determined by the newly identified fault-free nodes ;

[0026] Select the node with the highest failure probability, perform finite element analysis on the structural model corresponding to the selected node, and determine whether the selected node represents the system failure path. If so, the lower limit of the system failure probability is obtained from the failure probability of the selected node. , if not, then execute the branch operation to find the probability of calculating the child nodes obtained by branching from this node, and repeat the "branch" and "bound" operations until the termination condition is met.

[0027] In one embodiment of the present invention, the method for identifying failure paths and modes based on event reliability results using an improved branch and bound method further includes:

[0028] For continuous structures, let the dual termination condition be:

[0029] ;

[0030] In the formula, and is the set value, represents the upper bound, Indicates the lower limit.

[0031] In one embodiment of the present invention, a method for fitting a response surface function of each failure mode includes:

[0032] Fit the response surface function using a bivariate polynomial without cross terms:

[0033] ;

[0034] In the formula, , and represents the polynomial coefficients, represents the number of random variables, It means that the random errors caused by higher-order terms are ignored.

[0035] In one embodiment of the present invention, a method for calculating the overall failure probability of a hydraulic structure system includes:

[0036] The calculation formula for the overall failure probability of the hydraulic structure system is:

[0037] ;

[0038] In the formula, Indicates system events The probability and represents the probability of basic events, Represents the basic events in the system events. represents the system event vector, Indicated by The probability vector composed of represents transpose, Represents an index variable, where:

[0039] ;

[0040] In the formula, is the conditional probability, , for The joint probability distribution function of Represents a vector of common-source random variables.

[0041] In addition, the present invention also provides a collaboratively driven hydraulic structure system failure path identification and calculation system, comprising:

[0042] A data acquisition and preprocessing module, which is used to obtain multi-source measured information and determine the statistical characteristics of random variables of the multi-source measured information;

[0043] A model building module, which is used to build a finite element model based on the statistical characteristics of random variables of multi-source measured information;

[0044] A reliability calculation module, which is used to generate random variable set pairs using a random method based on the finite element model and the statistical characteristics of random variables, obtain the structural response of the physical structure of the hydraulic structure under different random variable settings through finite element simulation according to the random variable set pairs, and calculate the event reliability according to the structural response;

[0045] A failure mode identification module is used to identify failure paths and modes based on event reliability results using an improved branch and bound method. After the failure path is identified, a response surface function of each failure mode is obtained by fitting the reliability data of the failure path, and the failure probability of the hydraulic structure system is calculated based on the response surface function.

[0046] The system failure probability calculation module uses the identified failure modes and their corresponding response surface functions, considers the correlation of each failure mode, and calculates the overall failure probability of the hydraulic structure system through the matrix method.

[0047] Furthermore, the present invention also provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned method when executing the program.

[0048] Furthermore, the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, the steps of the above-mentioned method are implemented.

[0049] The above technical solution of the present invention has the following advantages compared with the prior art:

[0050] The present invention provides a collaboratively driven method and system for identifying and calculating failure paths of hydraulic structure systems, which integrates multi-source measured information, accurately determines the statistical characteristics of random variables, incorporates real uncertainties into calculations, and realizes accurate risk quantification, so that the failure probability fits the actual working conditions; a finite element model is constructed through the statistical characteristics of random variables to simulate a large number of working conditions efficiently, obtain structural response quantities under different settings, and enhance the accuracy and comprehensiveness of reliability calculations; failure paths and modes are intelligently identified based on event reliability, and failure laws can be intuitively presented; finally, the failure probability is calculated considering the correlation of each failure mode, so as to achieve system-level risk control, provide accurate navigation for operation and maintenance, and reinforcement investment, and effectively ensure the safety and economy of hydraulic structures throughout their life cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to make the contents of the present invention more clearly understood, the present invention is further described in detail below based on specific embodiments of the present invention in conjunction with the accompanying drawings.

[0052] Figure 1 It is a flow chart of a collaboratively driven hydraulic structure system failure path identification and calculation method proposed by the present invention.

[0053] Figure 2 It is a flowchart diagram of step S20 proposed by the present invention.

[0054] Figure 3 It is a flowchart diagram of step S40 proposed by the present invention.

[0055] Figure 4 It is the finite element mesh of a certain concrete dam system proposed in the present invention.

[0056] Figure 5 It is a finite element mesh detail diagram of a concrete dam system proposed in the present invention.

[0057] Figure 6 It is a flow chart of the improved branch and bound method proposed in the present invention.

[0058] Figure 7 These are the upper and lower bounds of the system failure probability obtained by the improved branch and bound method proposed in the present invention.

[0059] Figure 8 It is the main failure path of the bedrock proposed by the present invention.

[0060] Fig. 9 This is the relationship between the average elastic modulus of the dam body and the service time proposed by the present invention.

[0061] Fig.10 This is the relationship between the dam body strength reliability and failure probability and the service time proposed by the present invention.

[0062] Fig.11 This is the dam body strength failure path proposed by the present invention.

[0063] Fig.12 It is the time-varying failure probability change process of the system proposed by the present invention. DETAILED DESCRIPTION

[0064] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention.

[0065] Reference Figure 1 As shown, an embodiment of the present invention provides a collaboratively driven hydraulic structure system failure path identification and calculation method, comprising the following steps:

[0066] Step S10, obtaining multi-source measured information, and determining the statistical characteristics of random variables of the multi-source measured information;

[0067] Step S20, constructing a finite element model according to the statistical characteristics of random variables of multi-source measured information;

[0068] Step S30: Based on the finite element model and the statistical characteristics of random variables, a random method is used to generate random variable set pairs, and the structural response of the physical structure of the hydraulic structure under different random variable settings is obtained through finite element simulation according to the random variable set pairs, and the event reliability is calculated according to the structural response;

[0069] Step S40, using the improved branch and bound method to identify failure paths and modes based on event reliability results, after the failure paths are identified, the response surface function of each failure mode is obtained by fitting according to the reliability data of the failure paths, and the failure probability of the hydraulic structure system is calculated based on the response surface function;

[0070] Step S50: using the identified failure modes and their corresponding response surface functions, taking into account the correlation of each failure mode, the overall failure probability of the hydraulic structure system is calculated by matrix method.

[0071] The present invention provides a collaboratively driven method for identifying and calculating failure paths of hydraulic structure systems, which integrates multi-source measured information, accurately determines the statistical characteristics of random variables, incorporates realistic uncertainties into calculations, and achieves accurate risk quantification, so that the failure probability fits the actual working conditions; a finite element model is constructed through the statistical characteristics of random variables to efficiently simulate massive working conditions, obtain structural response quantities under different settings, and enhance the accuracy and comprehensiveness of reliability calculations; failure paths and modes are intelligently identified based on event reliability, and failure laws can be intuitively presented; finally, the overall probability is calculated considering the correlation of each failure mode, so as to achieve system-level risk control, provide accurate navigation for operation and maintenance, and reinforcement investment, and effectively ensure the safety and economy of hydraulic structures throughout their life cycle.

[0072] Among them, in step S10, it specifically includes obtaining multi-source measured information. In the field of hydraulic structures, it includes field test data, such as data obtained by mechanical property tests of building materials on site; in-situ observation data, such as dynamic information obtained by long-term monitoring of structural deformation and crack development in the actual operating environment of the building; and non-destructive detection detection information, such as the use of non-destructive detection technologies such as ultrasound and radar to grasp the data of internal defects and material conditions without destroying the structure of the building. For hydraulic structures, many influencing factors are random, that is, the values ​​of each measurement are not fixed. Therefore, after collecting multi-source measured information, it is necessary to conduct statistical analysis on these random variables to clarify their statistical characteristics, such as calculating the mean to know the average level of the variable; finding the standard deviation to grasp the degree of discreteness of the data; determining the probability distribution type, such as normal distribution, lognormal distribution, etc., to outline the change law of random variables. By accurately determining these statistical characteristics, a scientific data foundation can be provided when the performance and failure risk of hydraulic structures are evaluated by finite element analysis, reliability calculation and other means.

[0073] Among them, in step S20, the method of constructing a finite element model according to the statistical characteristics of random variables of multi-source measured information specifically includes the following steps:

[0074] Step S21, establishing an original text file containing nodes, lines, surface units and their sets, generating a finite element mesh model through a mesh divider, dividing the set according to the constitutive relationship, boundary conditions and load conditions, and then determining the boundary conditions, clarifying the parameters of each area of ​​the hydraulic structure system and its loading method, and completing the establishment of the finite element execution file;

[0075] Step S22: Based on the statistical characteristics of random variables of multi-source measured information, the GAWK program is called to modify the finite element execution file and generate a control file.

[0076] In step S22, based on the statistical characteristics of random variables of multi-source measured information, the GAWK program is called, and the constructed finite element execution file is modified in a targeted manner with the help of the random variable set pairs transmitted by the interface, so that the finite element model can be integrated into the uncertainty factors in the actual working conditions. Then, the GAWK program is called again to generate a finite element control file, which is very important. It integrates input files such as finite element execution files and finite element meshes, and also covers message files, text files that record the intermediate process and final results of the operation, and the result finite element binary file that is convenient for subsequent data processing, laying the foundation for the orderly advancement of the finite element analysis process.

[0077] Among them, in step S30, the limit state of the failure mode of the hydraulic structure system can be expressed as a function of structural response quantities such as stress, strain, displacement and energy. The limit state of the failure mode of the hydraulic structure-foundation system (hereinafter referred to as the hydraulic structure system) can be expressed as a function of structural response quantities such as stress, strain, displacement and energy. The failure of the hydraulic structure system is related to material properties, relative strength, aging rate, etc. Therefore, the structure and foundation are regarded as an integral structural system, and the coupling effect between them is considered to calculate the structural response quantity in the limit state of the hydraulic structure system.

[0078] Optionally, a random method is first used to generate a random variable set pair, and a structural response of the physical structure of the hydraulic structure under different random variable settings is obtained through finite element simulation according to the random variable set pair; a limit state function is constructed according to the structural response , where and are damage parameter and damage threshold respectively; the limit state function is calculated from this The normalized negative gradient vector is ; In the formula, Indicates A normalized negative gradient vector, represents the gradient operator, Indicates A limit state function, represents the design points in the space of uncorrelated standard normal random variables, Represents the norm of the vector, that is, the above-mentioned standardized negative gradient vector involves the design point of the limit state function and the standard normal random variable space. The limit state function is usually used to describe the conditions under which a structure or system will fail. For example, in structural reliability analysis, when the value of the limit state function is equal to 0, it may indicate that the structure is in a critical failure state.

[0079] Then, the event correlation is determined based on the normalized negative gradient vector, and the event reliability is obtained by combining the reliability analysis method. In the reliability analysis stage, the gradient vector and the normalized gradient vector are represented by row vectors. This vector representation method is convenient for mathematical processing and calculation. As a preferred method, the limit state function is constructed by taking damage as the structural response quantity. , where and They are damage parameter and damage threshold, respectively. In actual engineering, structural damage is an important consideration. By incorporating damage parameter and damage threshold into the limit state function, the failure state of the structure can be described more accurately. The damage threshold is a key indicator. When different damage thresholds are set, they correspond to different damage standards. Different damage standards will lead to different system failure modes and corresponding failure probabilities. For example, the overall instability caused by the attenuation of concrete strength due to aging diseases can be used as the standard for the failure of the concrete dam-bedrock system. According to certain damage standards (such as the distance from the occurrence of damage to the upstream dam heel), the target reliability index value is established to design the gravity dam; it is believed that in the failure mode analysis of the gravity dam, the allowable value of the depth of the dam heel failure path should be specified in advance so that the gravity dam can work stably under a certain load. Here, it is believed that when the number of finite element unit failures reaches 5 and meets the convergence condition (that is, the system failure probability is stable), the concrete dam-bedrock system fails as a whole, that is, when the threshold is exceeded, the event (finite element) is considered to have failed.

[0080] The following is a specific implementation method and steps of event finite element reliability analysis:

[0081] (1) Generation of random variable pairs and file modification: First, design methods such as star, complete factorial, central composite, Box-Behnken, etc. are used to generate random variable pairs. These design methods are used to reasonably select sample points in the random variable space to ensure that the value range of the variables is covered and to consider the interaction between variables to a certain extent. Then the generated random variable pairs are passed to the GAWK program. The GAWK program is a text processing tool, which is used here to modify the CA execution file and control file. By modifying these files, the execution folder of the finite element analysis (that is, specifying where the finite element analysis is performed) and the response quantities to be output (such as key physical quantities such as structural stress, strain, displacement, etc.) can be determined.

[0082] (2) Finite element analysis and result output: After the file modification is completed, the finite element analysis program is called. The finite element analysis program simulates and calculates the structural response of the structure under corresponding conditions based on the set random variable set. The finite element analysis will output the result file containing the structural response in the corresponding folder as required. These result files are the basic data source for subsequent calculations.

[0083] (3) Result file processing: Download the result file from the corresponding folder and read the structural response data therein.

[0084] (4) Limit state function and reliability calculation: First, based on the read structural response, the limit state function and its gradient with respect to the random variable are calculated. The limit state function is used to determine whether the structure has failed. For example, in structural reliability analysis, when the limit state function is less than or equal to 0, the structure is considered to have failed. Then, the finite element reliability calculation module is called to calculate the reliability of the event using the previously calculated limit state function and its gradient information. The reliability calculation module is usually based on certain reliability theories and algorithms, such as the second-order moment method and the Monte Carlo simulation method.

[0085] Among them, in step S40, the main failure path (failure mode) is identified by improving the branch and bound method, and the convergence standard of the branch and bound method is comprehensively considered. With reference to the main failure mode of hydraulic structures, the main failure mode of the actual project is determined. The improved branch and bound method program repeatedly calls the structural finite element analysis program to obtain the stress state distribution under given random variables and damage conditions, and outputs the stress distribution results, and evaluates the unit state through the state evaluation program. The specific operation method includes the following steps:

[0086] Step S41, initialization upper limit and lower bound , perform finite element analysis and stress state analysis on the structural model of the hydraulic structure system and determine the initial nodes;

[0087] Step S42, correcting and obtaining a structural model corresponding to the initial node after damage, performing finite element analysis and stress state analysis on the corrected structural model to determine the first batch of child nodes;

[0088] Step S43, calculating the event failure probability of each child node, and further obtaining the failure probability of the hydraulic structure system;

[0089] Step S44: Determine the limit based on the event failure probability and the system failure probability, and determine the upper limit based on the newly identified fault-free node ;

[0090] Step S45: Select the node with the highest failure probability, perform finite element analysis on the structural model corresponding to the selected node, and determine whether the selected node represents a system failure path. If so, obtain the lower limit of the system failure probability from the failure probability of the selected node. , if not, then execute the branch operation to find the probability of calculating the child nodes obtained by branching from this node, and repeat the "branch" and "bound" operations until the termination condition is met.

[0091] Furthermore, for a continuous structure with high redundancy, it may contain multiple paths with similar failure probabilities. In this case, a lot of time is often spent on identifying those system failure probabilities that could be ignored, while the lower bound of the failure probability has hardly changed. To avoid such unnecessary calculations, in this case, when the change in the lower bound is found to be negligible, the search is stopped. Therefore, for the continuous structure, the double termination condition is set as follows:

[0092] ;

[0093] In the formula, and is the set value, represents the upper bound, Indicates the lower limit.

[0094] Among them, in step S40, the method of fitting the response surface function of each failure mode according to the reliability data of the failure path specifically includes: when the response quantity of the system With input variables When the relationship function between is nonlinear and difficult to express analytically, the finite element method can be used to obtain it. Using the response surface method, the true limit state function Polynomial response surface function To approximate, for the input variable , the bivariate polynomial with cross terms is:

[0095] ;

[0096] In the formula, , , and are the polynomial coefficients, is the number of random variables, Because the random errors caused by high-order terms are ignored. Since the cross terms make the function contain a large number of variables, in practical applications, a bivariate polynomial response surface function without cross terms is generally used, that is, , failure can be expressed as the relationship between the structural capacity R and the structural response G:

[0097] Z = R / G ;

[0098] In the formula, a coefficient is given K ,when Z > K , indicating a safe state; when Z = K, represents the limit state; when Z < K , indicating a failed state.

[0099] Among them, in step S50, after obtaining the failure probability and main failure path (failure mode) of the hydraulic structure system, the response surface function of each failure mode is obtained by fitting the reliability of the failure path, and the overall failure probability of the hydraulic structure system is further calculated by the system failure probability calculation method. Specifically, the failure probability of a complex system under multiple events can be extended to by matrix multiplication:

[0100] ;

[0101] In the formula, , Indicates The event vector of events; , Indicates probability vector under conditions; matrix Indicates Line The column element is Under the conditions The failure probability of a system.

[0102] To solve the system failure probability expressed by the matrix, it is necessary to identify the vector and calculate vector , when the system is relatively simple, the event vector of simple system events can be directly identified; when the system is more complex, it is necessary to obtain it through matrix operations on each event vector. Indicates general events The event vector of When there is only one event of system failure, the probability vector can be expressed as , , and are the probabilities of event 1 and the corresponding complement respectively. Similarly, the probability vectors of systems containing 2 and 3 events can be calculated respectively:

[0103] ;

[0104] ;

[0105] Therefore, through sequential calculation, we can get n The system probability vector of events is .

[0106] When there is a statistical correlation among events, we cannot simply obtain the vector by multiplying the probabilities of the events. However, for most structural systems, if we can obtain random variables that represent environmental correlation and common source effects, we can obtain conditional independence of events. represents a vector of common source random variables, and from the total probability theorem we can get:

[0107] ;

[0108] In the formula, is the conditional probability, , for The joint probability distribution function of .

[0109] The failure probability of the system is:

[0110] ;

[0111] In the formula, is the conditional probability vector of non-repeated and non-missing events.

[0112] when , due to the conditional independence of the event, it can be calculated in order from the system probability vector For common source random variables, they can be represented by the DS (Dunnett-Sobel) correlation coefficient matrix. From the definition of system events, we know that the event vector It is not affected by the common origin of random variables, so:

[0113] .

[0114] Assumptions is a DS type standard normal random variable, and Correlation coefficient ( ), .thereby, Can be obtained by ( ) unrelated standard normal random variables are represented by:

[0115] ; In the formula, and are independent standard normal random variables. For a given result , Assuming that the normal random variable of the safety margin or limit state function of the system event is or has been converted into a DS type random variable, it can be obtained by a single random variable Identify common source effects. A special case is when all events are equally correlated, i.e. ( ).

[0116] For a general correlation coefficient matrix, we can find the true correlation The set with the smallest error , and is fitted by the DS class. If the difference in correlation coefficients leads to significant errors in the estimated system failure probability, the DS class can be generalized by adding common source random variables. random variables with common source, ( ) to improve the fitting accuracy, It can be expressed as:

[0117] ;

[0118] In the formula, Represents the correlation coefficient.

[0119] so, and The correlation coefficient between them is:

[0120] .

[0121] The conditions Substitute into the DS type standard normal random variable , we can calculate the probability of an event that is independent of the condition, that is, or .

[0122] For hydraulic structures in service, understanding the sensitivity of various parameters to the probability of system failure is of great significance for operation and management. For example, the sensitivity of various parameters to the failure probability of old and sick dams has reference value for operation, maintenance and risk prevention and reinforcement. In the risk management process, the importance based on sensitivity can be used to measure the importance of uncertain parameters. Specifically, the relative importance of each parameter is compared by the change in the probability of failure in the quantitative risk caused by the percentage of a certain parameter deviating from a fixed standard deviation. In the process of uncertainty management and quality control, the importance of variables is quantified using the sensitivity-based importance expressed in the following formula. In order to facilitate investment decisions for reinforcement, the sensitivity-based importance needs to be quantitatively expressed as:

[0123] ;

[0124] In the formula, Indicates the importance of sensitivity, represents the failure probability, express Parameters The partial derivative of For the parameters, For parameters The change in the value of (such as parameter changes caused by reinforcement), the system sensitivity is:

[0125] ;

[0126] ; In the formula, To include The formula for the probability of an event The probability vector obtained by the iterative procedure (except 1 and -1, using and replace); is the marginal event probability vector

[0127] The following is a detailed description of a method for identifying and calculating a failure path of a collaboratively driven hydraulic structure system proposed by the present invention through a specific implementation method.

[0128] Taking a concrete dam as an example, a river-blocking dam of a hydropower station in service is a Class 1 hydraulic structure. The normal water storage level of the reservoir is 173m, the dam top elevation is 179.00m, the design and verification flood levels are 174.76m (0.2%) and 177.80m (0.02%) respectively, and the downstream water level is basically stable, which is an incompletely regulated reservoir. The dam top width is 7.0m, the dam bottom elevation is 80.0m, the dam bottom width is 73m, and the dam base is horizontal. The comprehensive elastic modulus of the concrete dam system is 27.71GPa, and the Poisson's ratio is 27.71GPa. , self-respect , Mazars model parameters , , , , and The geological survey results show that the foundation rock is mostly slightly weathered granite, and the rock / rock shear strength is , MPa. Parameters of the linear hardening DP model for slightly weathered granite at the dam foundation: elastic modulus is MPa, Poisson's ratio , yield strength MPa, hardening modulus constant MPa, ultimate plastic strain Recommended values ​​for the friction coefficient and cohesion of the concrete / bedrock bond and the shear strength , cohesion MPa; Based on the tensile strength in Griffith's theory of brittle material fracture and cohesion Relationship Approximate tensile strength of the bonded surface; normal and tangential stiffness MPa, tensile strength MPa, the roughness is at a normal level. Curtain lift pressure reduction factor at the initial stage of operation It can be calculated from the measured data In addition, the geometric shape of the dam is considered to be a time-invariant variable, while the strength of the dam concrete, the shear strength of the concrete / bedrock bonding surface, the upstream and downstream water levels, and the dam foundation uplift pressure are time-varying variables, and all of them are random variables. The KS test was performed on the measured data of the upstream reservoir water level. The annual maximum value of the upstream reservoir water level obeys the right-truncated normal distribution. , ; Downstream water level is negligible.

[0129] The change of dam foundation uplift pressure is one of the most common structural aging signals, which is mainly affected by factors such as upstream and downstream water levels, dam foundation temperature, and curtain body anti-seepage effect. The design uplift pressure reduction coefficient is 0.3. According to the measured data, the uplift pressure change at the curtain lags behind the change of the upstream water level. The uplift pressure reduction coefficient is slightly larger than the design value. Under the condition of similar reservoir water level upstream, the uplift pressure has a certain increasing trend. This is caused by the blockage of the drainage system and the erosion of the grouting curtain during the operation of the project.

[0130] Based on the measured data, the evolution model of the uplift pressure reduction coefficient is obtained by fitting the following formula:

[0131] ;

[0132] In the formula, the failure rate of uplift pressure reduction is 0.00285, and the curtain construction quality parameter is =23; Assuming that no extreme events (such as earthquakes and major floods) that may cause curtain rupture occur during future operation, .

[0133] The elastic modulus, tensile strength and compressive strength of the dam concrete are highly correlated, which is assumed to be 0.9; the upstream water level is related to the uplift pressure reduction coefficient, and the correlation coefficient obtained from the measured data is 0.425.

[0134] Considering random variables such as the elastic modulus and Poisson's ratio of the dam body, the bulk density of the dam concrete, the elastic modulus of the bedrock, the upstream water level, the curtain uplift pressure reduction factor, the cohesion and friction coefficient of the bedrock (reflected as the yield strength in the DP model), and the tensile strength, friction coefficient and cohesion of the dam concrete / bedrock bonding surface, the statistical characteristics of the random variables are shown in Table 1.

[0135] Table 1 Random variables and statistical characteristics

[0136]

[0137] With the help of full factorial method to design sample points, the finite element reliability analysis of each failure event requires calling the finite element program hundreds of times.

[0138] The finite element mesh model of the concrete dam system is as follows: Figure 4 As shown, the detail unit number is as follows Figure 5 As shown, use Figure 6 The improved branch and bound method shown in the figure analyzes the destruction process of the concrete dam system and determines the failure probability of each link in the failure sequence based on the limit state function. Each finite element calculation can identify a branch of the failure sequence or determine a system failure path. In this process, the upper and lower bounds of the system failure probability decrease and increase respectively.

[0139] Taking the gravity dam as an example, the failure mechanisms of different parts of the gravity dam are different, and there may be multiple failure criteria. Different failure evaluation criteria are used for different parts. The improved branch and limit method program repeatedly calls the structural finite element analysis program to obtain the stress state distribution under given random variables and damage conditions, and outputs the stress distribution results. The unit state is evaluated through the state judgment program. The target value of the upper and lower bound difference ratio of the probability is taken as a relatively large value. The relationship between the number of finite element simulation calculations and the upper and lower bounds of the system failure probability is as follows: Figure 7 shown.

[0140] After 121 finite element analysis, the upper and lower bounds of the system failure probability are calculated as follows: and The main failure modes obtained include bedrock failure and tensile failure along the cementation surface. There are two main failure paths of the bedrock, starting from the left side of the dam heel and the unit directly below the dam heel, with unit numbers 7632 and 6186 respectively; the unit compositions of each failure path are 7632-7631-7633-4177-4173 and 6186-4177-4173-4178-4179, as shown in Figure 8 As shown; another failure path identified is the concrete / bedrock bonding surface unit, and the failure path is along the bonding surface from upstream to downstream, that is, 10348-10349-10350-10351-10352.

[0141] Table 2 lists the cumulative failure probabilities corresponding to the three main failure paths.

[0142] Table 2 Main failure paths and failure probabilities

[0143]

[0144] The main failure path of anti-sliding failure was identified, and its unit composition was 10420-10419-10418-10417-10416. The response surface method was used to fit the response surface functions of the first unit failure of each failure path, such as cementing surface fracture failure, dam foundation failure path 1 and dam foundation failure path 2, and anti-sliding instability of the cementing surface, as follows:

[0145] ;

[0146] ;

[0147] ;

[0148] ;

[0149] In the formula, They represent upstream water level, bedrock elastic modulus, tensile strength of cementing surface, bedrock internal friction coefficient, bedrock cohesion, dam body bulk density, uplift pressure reduction factor, cementing surface friction coefficient, cementing surface cohesion and dam body elastic modulus respectively. The units of each variable are the same as those in Table 1.

[0150] Regarding the calculation method of the system failure probability, it is necessary to calculate the sensitivity of the random variables of each failure mode (respectively denoted as NM1~NM4) and the negative normalized gradient vector. The calculation results of the mean value of the variable in each failure mode to the failure probability are listed in Table 3 and Table 4 respectively.

[0151] Table 3 Failure probability and normalized gradient vector of each failure mode

[0152]

[0153] Table 4 Sensitivity of failure probability to the average value of each parameter ( )

[0154]

[0155] The correlation coefficient matrix between the main failure modes is calculated as follows:

[0156] ;

[0157] The failure probability of the concrete dam system under the corresponding load and parameter conditions can be calculated as .

[0158] The influence of aging on the failure probability of concrete dams is systematically analyzed. The changes in the properties of dam materials are as follows: Fig. 9As shown. At the same time, the material variation coefficient increases from 0.1 at the beginning of service to 0.2 when the dam body is completely dissolved. The random variables of the load conditions are listed in Table 1. The average value at the beginning of operation is used as the input parameter, and the most likely failure of the dam heel unit is obtained by calculation and analysis. Based on the platform calculation and analysis, the response surface function of the unit failure is obtained as follows:

[0159] ;

[0160] In the formula, the random variable and are the bulk density of dam concrete and Poisson’s ratio respectively. The units of random variables are the same as those in Table 1.

[0161] The sensitivity of failure probability to the statistical characteristics of each parameter is listed in Fig.10 The time-varying reliability and failure probability of dam strength failure are listed in Table 5. The negative normalized gradient vector is .

[0162] Table 5 Sensitivity of failure probability to the average value of each parameter

[0163]

[0164] On this basis, the influence of dam body corrosion and aging on the system failure probability is analyzed. The main path of dam body strength failure is obtained by platform identification (this failure path is recorded as NM5). Fig.11 The failed unit consists of units 4165-4166-916-2172-2171 (the unit grid numbers are the same as Figure 5 ), the corresponding cumulative failure probability is listed in Table 6; at the same time, the cumulative probability of strength failure in the corrosion aging process is calculated and also listed in Table 6.

[0165] Table 6 Dam strength failure paths and corresponding failure probabilities during service

[0166]

[0167] The correlation coefficient matrix between the main failure paths NM1, NM2, NM3 and NM5 is calculated as follows:

[0168] ;

[0169] The failure of the concrete dam system is regarded as a series combination of four failure modes, namely NM1, NM2, NM3 and NM5. The calculation method of the system failure probability is used to obtain the following: Fig.12 The failure probability time-varying process curve of the concrete dam system is shown in Figure 2. Fig.12 It can be seen that the probability of system failure in the early stage of service is After 10 years of service, affected by the corrosion and attenuation of the dam concrete, the dam strength failure became the main failure mode of the concrete dam system. After 30 years of operation, the failure probability of the system was as high as 0.084.

[0170] Corresponding to the above method embodiment, the present invention also provides a collaboratively driven hydraulic structure system failure path identification and calculation system, comprising:

[0171] A data acquisition and preprocessing module, which is used to obtain multi-source measured information and determine the statistical characteristics of random variables of the multi-source measured information;

[0172] A model building module, which is used to build a finite element model based on the statistical characteristics of random variables of multi-source measured information;

[0173] A reliability calculation module, which is used to generate random variable set pairs using a random method based on the finite element model and the statistical characteristics of random variables, obtain the structural response of the physical structure of the hydraulic structure under different random variable settings through finite element simulation according to the random variable set pairs, and calculate the event reliability according to the structural response;

[0174] A failure mode identification module is used to identify failure paths and modes based on event reliability results using an improved branch and bound method. After the failure path is identified, a response surface function of each failure mode is obtained by fitting the reliability data of the failure path, and the failure probability of the hydraulic structure system is calculated based on the response surface function.

[0175] The system failure probability calculation module is used to utilize the identified failure modes and their corresponding response surface functions, consider the correlation of each failure mode, and accurately calculate the overall failure probability of the hydraulic structure system through the matrix method.

[0176] The collaboratively driven hydraulic construction system failure path identification and calculation system of this embodiment is used to implement the aforementioned collaboratively driven hydraulic construction system failure path identification and calculation method. Therefore, its specific implementation method can refer to the description of the corresponding various partial embodiments and will not be introduced in detail here.

[0177] In addition, since the cooperatively driven hydraulic structure system failure path identification and calculation system of this embodiment is used to implement the aforementioned cooperatively driven hydraulic structure system failure path identification and calculation method, its function corresponds to that of the aforementioned method and will not be repeated here.

[0178] Corresponding to the above method embodiment, an embodiment of the present invention further provides a computer device, including:

[0179] A memory for storing computer programs;

[0180] A processor is used to implement the steps of the above-mentioned collaboratively driven hydraulic structure system failure path identification and calculation method when executing a computer program.

[0181] In the embodiment of the present invention, the processor may be a central processing unit (CPU), an application specific integrated circuit, a digital signal processor, a field programmable gate array or other programmable logic devices.

[0182] The processor may call the program stored in the memory. Specifically, the processor may execute the operations in the embodiment of the above-mentioned collaboratively driven hydraulic structure system failure path identification and calculation method.

[0183] The memory is used to store one or more programs, which may include program codes, and the program codes include computer operating instructions.

[0184] In addition, the memory may include a high-speed random access memory and may also include a non-volatile memory, such as at least one disk storage device or other volatile solid-state storage device.

[0185] Corresponding to the above method embodiment, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned collaboratively driven hydraulic structure system failure path identification and calculation method are implemented.

[0186] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0187] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0188] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0189] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0190] Obviously, the above embodiments are merely examples for clear explanation and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived from these are still within the protection scope of the invention.

Claims

1. A collaboratively driven hydraulic structure system failure path identification and calculation method, characterized by: include: Acquire multi-source measured information, and determine the statistical characteristics of random variables of the multi-source measured information; Construct a finite element model based on the statistical characteristics of random variables of multi-source measured information; Based on the finite element model and the statistical characteristics of random variables, a random method is used to generate random variable set pairs, and the structural response of the physical structure of the hydraulic structure under different random variable settings is obtained through finite element simulation according to the random variable set pairs, and the event reliability is calculated according to the structural response; The improved branch and bound method is used to identify failure paths and modes based on event reliability results. After the failure path is identified, the response surface function of each failure mode is obtained by fitting the reliability data of the failure path, and the failure probability of the hydraulic structure system is calculated based on the response surface function. Using the identified failure modes and their corresponding response surface functions, and considering the correlation of each failure mode, the overall failure probability of the hydraulic structure system is calculated by the matrix method. The method for calculating the event reliability according to the structural response comprises: Construct the limit state function g according to the structural response i (x) = σ i -[σ] i , where σ and [σ] are damage parameter and damage threshold respectively; Calculate the limit state function g i The normalized negative gradient vector of (x) is: In the formula, α i represents the i-th normalized negative gradient vector, represents the gradient operator, g i (x) represents the i-th limit state function, denotes the design point in the space of uncorrelated standard normal random variables, ||·|| denotes the norm of the vector; Event correlation is determined based on the normalized negative gradient vector, and event reliability is obtained by combining the reliability analysis method.

2. The method for identifying and calculating failure paths of a collaboratively driven hydraulic structure system according to claim 1, characterized in that: The method of constructing a finite element model based on the statistical characteristics of random variables of multi-source measured information includes: Establish an original text file containing nodes, lines, surface units and their collections, generate a finite element mesh model through a mesh divider, divide the collection according to the constitutive relationship, boundary conditions and load conditions, and then determine the boundary conditions, clarify the parameters of each area of ​​the hydraulic structure system and its loading method, and complete the establishment of the finite element execution file; Based on the statistical characteristics of random variables of multi-source measured information, the GAWK program is called to modify the finite element execution file and generate the control file.

3. The method for identifying and calculating failure paths of a collaboratively driven hydraulic structure system according to claim 1, characterized in that: The method of identifying failure paths and modes based on event reliability results using the improved branch and bound method includes: Initialize the upper bound P upp and the lower bound P low , perform finite element analysis and stress state analysis on the structural model of the hydraulic structure system and determine the initial nodes; The structural model corresponding to the initial node after damage is corrected, and the finite element analysis and stress state analysis are performed on the corrected structural model to determine the first batch of sub-nodes; Calculate the event failure probability of each sub-node, and further obtain the failure probability of the hydraulic structure system; The limit is determined based on the event failure probability and the system failure probability, and the upper limit P is determined by the newly identified fault-free node upp ; Select the node with the highest failure probability, perform finite element analysis on the structural model corresponding to the selected node, and determine whether the selected node represents the system failure path. If so, the lower limit P of the system failure probability is obtained from the failure probability of the selected node. low , if not, then execute the branch operation to find the probability of calculating the child nodes obtained by branching from this node, and repeat the "branch" and "bound" operations until the termination condition is met.

4. The method for identifying and calculating failure paths of a collaboratively driven hydraulic structure system according to claim 3 is characterized in that: The method for identifying failure paths and modes based on event reliability results using an improved branch and bound method also includes: For continuous structures, let the dual termination condition be: In the formula, ε1 and ε2 are set values, P upp represents the upper bound, P low Indicates the lower limit.

5. The method for identifying and calculating failure paths of a collaboratively driven hydraulic structure system according to claim 1, characterized in that: The method of fitting the response surface function of each failure mode includes: Fit the response surface function using a bivariate polynomial without cross terms: In the formula, a, b i and c i Represents the polynomial coefficient, n represents the number of random variables, and ε represents the random error caused by ignoring high-order terms.

6. The method for identifying and calculating failure paths of a collaboratively driven hydraulic structure system according to claim 1, characterized in that: The method for calculating the overall failure probability of a hydraulic structure system includes: The calculation formula for the overall failure probability of the hydraulic structure system is: In the formula, P(E sys ) indicates system event E sys The probability and p j represents the probability of basic events, e j represents the basic event in the system event, c represents the system event vector, and p represents the j The probability vector composed of, T represents the transpose, j represents the index variable, where: P sys =∫ x P(E sys |x)f x (x)dx; In the formula, P(E sys |x) is the conditional probability, X = x, f x (x) is the joint probability distribution function of X, where X represents a vector of common-source random variables.

7. A collaboratively driven hydraulic structure system failure path identification and calculation system, characterized by: include: A data acquisition and preprocessing module, which is used to obtain multi-source measured information and determine the statistical characteristics of random variables of the multi-source measured information; A model building module, which is used to build a finite element model according to the statistical characteristics of random variables of multi-source measured information; a reliability calculation module, which is used to generate random variable set pairs using a random method based on the finite element model and the statistical characteristics of random variables, obtain the structural response of the physical structure of the hydraulic structure under different random variable settings through finite element simulation according to the random variable set pairs, and calculate the event reliability according to the structural response; A failure mode identification module is used to identify failure paths and modes based on event reliability results using an improved branch and bound method. After the failure path is identified, a response surface function of each failure mode is obtained by fitting the reliability data of the failure path, and the failure probability of the hydraulic structure system is calculated based on the response surface function. The system failure probability calculation module uses the identified failure modes and their corresponding response surface functions, considers the correlation of each failure mode, and calculates the overall failure probability of the hydraulic structure system through the matrix method; The method for calculating the event reliability according to the structural response comprises: Construct the limit state function g according to the structural response i (x) = σ i -[σ] i , where σ and [σ] are damage parameter and damage threshold respectively; Calculate the limit state function g i The normalized negative gradient vector of (x) is: In the formula, α i represents the i-th normalized negative gradient vector, represents the gradient operator, g i (x) represents the i-th limit state function, denotes the design point in the space of uncorrelated standard normal random variables, ||·|| denotes the norm of the vector; Event correlation is determined based on the normalized negative gradient vector, and event reliability is obtained by combining the reliability analysis method.

8. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images