A method for diagnosing the service behavior of a structure based on the failure mode and failure path of a high dam

Abstract

The present invention provides a method for diagnosing the service behavior of a structure based on the failure modes and failure paths of high dams, including: constructing a high dam failure case database and extracting failure characteristic parameters; applying multi-scale hierarchical adaptive correspondence analysis to solve the problem of unbalanced characteristic weights through characteristic weight calculation, hierarchical weighting of failure modes, establishment of physical similarity indicators, and structural difference compensation, and generating a weighted similar project ranking list; constructing a time-series adaptive network model coupled with multiple failure modes to solve the problem of imperfect time-series evolution mechanism of failure paths and outputting a time-series failure path prediction model; combining the similar project ranking list and the time-series failure path prediction model to conduct engineering analogy analysis and generate a high dam safety status assessment report. It solves the technical problems such as the short operation time of super-high dams, insufficient monitoring information, and the difficulty of numerical analysis results to fully reflect the actual project, and provides a new technical path and solution for high dam safety assessment and risk management.

Description

Technical Field

[0001] The present invention relates to building diagnosis technologies, and in particular to a method for diagnosing the service behavior of a structure based on the failure modes and failure paths of high dams. Background Art

[0002] As major water conservancy infrastructure, high dams and large reservoirs play important roles in flood control, water supply, irrigation and safety. With the increase of the service life and the influence of extreme actions such as earthquakes and over-standard floods, the probability of diseases and risks such as cracks and serious leakage in high dams increases, causing some losses. Therefore, systematic analysis of the failure modes and failure paths of high dams and establishment of a scientific method for diagnosing the service behavior of structures have important theoretical value and engineering significance for ensuring the safe operation of water conservancy projects, and are hot research directions at home and abroad.

[0003] At present, the safety assessment of high dams at home and abroad mainly relies on in-situ monitoring and on-site inspection to obtain disease information, and reveals the physical cause mechanism through establishing monitoring models, carrying out simulation tests and structural numerical calculations. For example, the safety assessment method of high dams based on neural networks uses historical monitoring data for pattern recognition and early warning; the assessment method based on risk probability analysis calculates the failure probability through Monte Carlo simulation; the assessment method based on finite element analysis analyzes the stress, deformation and seepage characteristics of high dams under different working conditions through numerical simulation. In addition, the multiple correspondence analysis method is applied to engineering analogy analysis, and provides a reference basis for safety assessment by comparing the similarity between the high dam to be evaluated and historical cases.

[0004] However, there are still key technical problems in the diagnosis of the failure modes and failure paths of high dams in the existing technologies. On the one hand, equal weight processing is generally adopted in traditional multiple correspondence analysis, ignoring the difference in the influence degree of different characteristic parameters on the failure modes and failure paths, resulting in excessive attention to secondary characteristics and neglect of key factors during engineering analogy, and reducing the accuracy of diagnosis. On the other hand, the existing failure path analysis mainly presents a static topological structure, lacking an accurate description of the time-series evolution relationship between each link during the failure process, unable to reflect the development rate of diseases and the time dimension characteristics of the failure process, making the risk early warning and life assessment of high dam operation lack accuracy in the time scale. These problems are particularly prominent in the 300m-class ultra-high dam projects. Due to the short construction and operation time and insufficient accumulation of monitoring information, they exceed the existing technical level and specification requirements, and there is an urgent need to develop new diagnosis methods. Summary of the Invention

[0005] The object of the invention is to provide a method for diagnosing the service behavior of a structure based on the failure modes and failure paths of high dams, so as to solve at least one technical problem existing in the existing technologies.

[0006] Technical solution: A method for diagnosing the service behavior of a structure based on the failure mode and failure path of a high dam includes the following steps:

[0007] Collect information on high dam failure cases, construct a high dam failure case database, and extract characteristic parameters, failure modes, and failure paths to form a high dam failure case dataset;

[0008] Based on the high dam failure case dataset, apply multi-scale hierarchical adaptive correspondence analysis to generate a weighted similar project ranking list;

[0009] Based on the high dam failure case dataset, construct a time-series adaptive network model with coupled multi-failure modes, and output a time-series failure path prediction model;

[0010] Combine the weighted similar project ranking list and the time-series failure path prediction model to conduct engineering analogy analysis on the high dam to be evaluated, and generate a service safety diagnosis report.

[0011] Beneficial effects: This application provides a new technical path and solution for high dam safety assessment and risk management. The relevant technical effects will be described in detail in combination with cases. Description of the drawings

[0012] Figure 1 is the flowchart of the present invention.

[0013] Figure 2 is the flowchart of the present invention for generating a weighted similar project ranking list.

[0014] Figure 3 is the flowchart of the present invention for calculating the characteristic weight vector.

[0015] Figure 4 is the flowchart of the present invention for constructing the mode characteristic weight matrix.

[0016] Figure 5 is the flowchart of the calculation process for correcting the physical similarity index matrix of the present invention. Detailed implementation manners

[0017] As Figures 1 to 5 shown, the technical solution of this application includes the following steps:

[0018] S1. Collect information on high dam failure cases at home and abroad, construct a structured database, extract failure characteristic parameters, and form a high dam failure case dataset.

[0019] S11. Obtain the original case information from engineering literature, accident reports, and monitoring records, including the basic parameters of the high dam (dam type, dam height, reservoir capacity, construction year, country where it is located) and failure information (failure time, failure type, failure location, treatment measures).

[0020] S12. Clean and standardize the original case information, remove outliers and cases with serious missing data, unify the measurement units and classification criteria, and generate standardized case data.

[0021] S13. Based on the standardized case data, extract the dam structure characteristics (such as dam type coefficient, dam height ratio, thickness coefficient), hydrogeological characteristics (such as reservoir capacity coefficient, geological conditions), and operating environment characteristics (such as meteorological conditions, operating years) to form a characteristic parameter matrix X.

[0022] S14. Analyze the failure modes of the standardized case data, identify the three main failure modes of cracks, abnormal deformation, and seepage and their subtypes, and construct a failure mode classification system Y.

[0023] S15. Analyze the failure evolution process in the standardized case data, identify the logical relationship between the failure causes, development processes, and failure results, and construct an initial failure path network Z.

[0024] S16. Integrate the characteristic parameter matrix X, the failure mode classification system Y, and the initial failure path network Z into a structured database to form a complete high dam failure case dataset.

[0025] S2. Based on the high dam failure case dataset, apply multi-scale hierarchical adaptive correspondence analysis to solve the problem of unbalanced feature weights and generate a weighted list of similar engineering arrangements.

[0026] S21. Extract the characteristic parameter matrix X from the high dam failure case dataset, calculate the point biserial correlation coefficient ρ between each characteristic parameter and the failure mode ij , and construct a characteristic correlation matrix R.

[0027] S22. Based on the characteristic correlation matrix R, calculate the average correlation coefficient ρj of each characteristic and perform normalization processing to obtain an initial characteristic weight vector W0. This solves the limitation of equal weight processing and highlights the importance of key features.

[0028] S23. According to the failure mode classification system Y, divide the high dam failure case dataset into m subsets C1, C2,..., Cm, and calculate the characteristic weights separately for each subset to obtain a mode characteristic weight matrix WM. This solves the problem of differences in feature importance under different failure modes.

[0029] S24. Construct a failure mode correlation matrix R, where the matrix element r ij represents the association degree between failure modes i and j, calculated based on the co-occurrence frequency and transformation probability. This step considers the evolutionary relationship between failure modes.

[0030] S25. Construct a physical similarity index system, including n physical indexes such as stress distribution similarity (σ similarity), deformation mode similarity (ε similarity), seepage field similarity (Φ similarity), etc., and establish a mapping relationship F between physical indexes and basic characteristic parameters: characteristic parameter matrix X → physical similarity index matrix P. Introducing physical mechanisms enhances the scientific nature of similarity analysis.

[0031] S26. Apply the mapping relationship F to the characteristic parameter matrix X to obtain the physical similarity index matrix P of all historical cases. For the special structural characteristics of modern super-high dams, design a structural difference compensation function H to correct the physical similarity indexes, and obtain the corrected physical similarity index matrix P'. This solves the problem of structural representativeness differences between new high dams and historical cases.

[0032] S27. Extract the characteristic parameters of the high dam to be evaluated, construct the characteristic vector X0 of the high dam to be evaluated, and through the mapping relationship F and the compensation function H, obtain the physical similarity index vector P0 of the high dam to be evaluated.

[0033] S28. Calculate the weighted Euclidean distance DX between the characteristic vector X0 of the high dam to be evaluated and each case in the characteristic parameter matrix X in the original characteristic space, and determine the weight comprehensively based on the characteristic correlation matrix R, the mode characteristic weight matrix WM, and the failure mode correlation matrix R. At the same time, calculate the weighted Euclidean distance DP between the physical similarity index vector P0 of the high dam to be evaluated and the corrected physical similarity index matrix P' in the physical similarity space.

[0034] S29. Adopt a multi-scale fusion strategy, comprehensively calculate DX and DP to obtain the final similarity S = α·f(DX) + (1-α)·g(DP), where α is an adaptive weight factor, and f and g are distance-similarity conversion functions. Sort the historical cases based on the final similarity S to generate a weighted similar project ranking list. Making full use of the advantages of the original characteristic space and the physical similarity space improves the reliability of similar project screening.

[0035] S3. Utilize the failure information in the high dam failure case dataset to construct a time-series adaptive network model with multi-failure mode coupling, solve the problem of imperfect time-series evolution mechanism of failure paths, and output a time-series failure path prediction model.

[0036] S31. Extract the initial failure path network Z from the high dam failure case dataset, define each node in the network as a state Si of a Markov chain, and add a normal state S0 and a final failure state Sf to form a complete state space S.

[0037] S32. Based on the historical cases in the high dam failure case dataset, count the transfer frequencies between states, and calculate the state transition probability matrix P, where P ijRepresents the transition probability from state Si to state Sj. It solves the limitation of only representing the logical order without quantifying the probability.

[0038] S33. For each state transition process Si→Sj, extract the time records from the high dam failure case dataset and fit its time distribution function T ij (t) and its parameters (mean μ ij , variance σ ij 2 ), and construct the state duration distribution matrix D. Introducing the time dimension into the failure path analysis solves the problem that the static path cannot reflect the time series characteristics.

[0039] S34. Analyze the main external factors (water level, temperature, load, etc.) affecting the high dam failure process, extract the relevant monitoring data from the high dam failure case dataset, and construct the external factor data matrix F.

[0040] S35. Establish the conditional state transition probability function P ij (F1,F2,...,Fk), which represents the relationship between the state transition probability and the external factors. At the same time, establish the mapping relationship between the time distribution parameters and the external factors: μ ij =g(F1,F2,...,Fk), σ ij 2 =h(F1,F2,...,Fk). It can adapt to the non-stationary failure process under complex working conditions.

[0041] S36. Model different types of failure modes (cracks, deformations, seepage, etc.) as independent network layers respectively, and construct the multi-layer failure mode network M. Define the coupling strength matrix C of the failure modes, where C ij represents the influence strength of failure mode i on mode j. It can depict the interaction between failure modes.

[0042] S37. Adopt a rule-based path generation engine to automatically generate new possible paths in the multi-layer failure mode network M according to the failure mode coupling relationship, and dynamically update the initial failure path network Z to form the extended failure path network Z'. It solves the problem that the preset path cannot cover the path mutation.

[0043] S38. Based on the state transition probability matrix P, the state duration distribution matrix D, F, the multi-layer failure mode network M and the extended failure path network Z', construct a comprehensive time series evolution model to realize the multi-scale (macroscopic state sequence, mesoscopic network spread, microscopic physical evolution) simulation of the failure process. It realizes the full-scale time series modeling of the high dam failure process.

[0044] S39. Generate a large number of possible failure timing paths through the Monte Carlo simulation method, statistically analyze the time characteristics of typical failure paths, identify the key nodes and critical states that may cause path mutations, and form a complete timing failure path prediction model.

[0045] S4. Combine the weighted similar project sorted list and the timing failure path prediction model, conduct engineering analogy analysis for the high dam to be evaluated, and generate a high dam safety status assessment report.

[0046] S41. Select the k projects with the highest similarity from the weighted similar project sorted list as the core analogy project set, analyze their accident occurrence modes, locations and causes, and predict the possible potential accident occurrence modes of the high dam to be evaluated.

[0047] S42. Input the monitoring data of the high dam to be evaluated as the initial state information into the timing failure path prediction model, simulate and predict its possible failure evolution path and time window, and generate a failure risk timing diagram.

[0048] S43. For the typical projects in the core analogy project set, combine the characteristics of the high dam to be evaluated, construct a finite element numerical model, conduct failure mechanism analysis, verify the rationality of the potential accident occurrence modes and the failure risk timing diagram, and form a verification analysis result.

[0049] S44. Based on the potential accident occurrence modes, the failure risk timing diagram and the verification analysis result, comprehensively evaluate the safety status of the high dam to be evaluated, determine the risk level, propose targeted monitoring and intervention strategy suggestions, and generate a high dam safety status assessment report.

[0050] According to one aspect of the present application, step S15 is specifically as follows:

[0051] S151. Extract the accident event sequence of each case from the standardized case data, including accident phenomena, occurrence time, development status and influence scope, and form the accident event timing data E.

[0052] S152. Conduct timing decomposition on the accident event timing data E, identify the starting event, intermediate development event and final result event in each case, and construct a case-level event chain set L. Through time window segmentation and event correlation analysis, the accurate deconstruction of the accident process is realized.

[0053] S153. Conduct semantic normalization processing on all events in the event chain set L, merge and unify the events with different expressions but the same essence, and obtain the standardized event type set T. This step adopts an ontology mapping method based on expert knowledge to solve the problem of inconsistent event descriptions in different projects and different literatures.

[0054] S154. Analyze the forward and backward dependencies among events in the event chain set L based on the event type set T, count the transfer frequencies between each pair of events, and form an event transfer frequency matrix F. The element f ij in this matrix represents the historical occurrence times from event i to event j.

[0055] S155. Normalize the event transfer frequency matrix F, calculate the conditional probability P(j|i) = f ij / ∑fik, and construct an event transfer probability matrix P event . The probability statistical method is introduced to convert the qualitative failure path into a quantifiable probability model.

[0056] S156. Based on the event transfer probability matrix P event , use the minimum spanning tree algorithm to extract the main event transfer paths, eliminate the transfer relationships with probabilities lower than the threshold θ, and obtain the core event transfer network N core .

[0057] S157. Combine the high dam engineering mechanics theory and expert knowledge to verify the physical rationality of the event transfer relationships in the core event transfer network N core , delete the physically unreasonable transfer edges, and supplement the transfer edges that physically necessarily exist but are missing in the data to form a physically corrected event network N phy .

[0058] S158. Classify the events in the event type set T two-dimensionally according to the failure types (such as cracks, deformations, seepage, etc.) and development stages (initiation, development, failure), and construct an event classification matrix C. This matrix provides a structured framework for the multi-dimensional display of the failure path.

[0059] S159. Based on the physically corrected event network N phy and the event classification matrix C, construct an initial failure path network Z in graph structure, where the nodes are event types, the edges are event transfer relationships, and the edge weights are transfer probabilities. This network not only retains the topological structure of the failure process but also integrates probability characteristics, laying a foundation for subsequent time-series evolution analysis.

[0060] According to one aspect of the present application, step S22 is specifically as follows:

[0061] S221. Read the feature correlation matrix R, and extract the correlation coefficient ρ ij between each feature parameter j and all failure modes, where i represents the failure mode index and j represents the feature parameter index, and construct a feature-failure correlation coefficient set ρ_set.

[0062] S222. Conduct statistical analysis on the feature-failure correlation coefficient set ρ_set, and calculate the average correlation coefficient ρ_avg of each feature jj = (1 / M)∑ρ ij , where M is the total number of risk occurrence modes, and the average correlation coefficient vector ρ_avg is obtained. Through the averaging process, the influence ability of features on different risk occurrence modes is comprehensively considered.

[0063] S223. Conduct a significance test on the values in the average correlation coefficient vector ρ_avg, screen out the feature parameters with insignificant correlation (p-value > 0.05), retain the significantly correlated feature subset, and construct the significant feature index set J_sig. This step introduces a statistical test method, improving the scientificity of feature weight calculation.

[0064] S224. For the features in the significant feature index set J_sig, introduce an expert scoring mechanism. Based on the historical case analysis experience and theoretical analysis results, determine the expert weight coefficient e of each feature j , forming the expert weight vector E. Combining expert experience with data statistics enhances the rationality of the weight system.

[0065] S225. Design a fusion function f to perform weighted fusion on the data-driven average correlation coefficient vector ρ avg and the expert weight vector E: f j = α·ρ avgj + (1-α)·e j , where α is the adaptive fusion coefficient, which is dynamically adjusted according to data quality. The fusion weight value vector F is obtained. The fusion of data-driven and expert knowledge is realized, overcoming the limitations of a single weight determination method.

[0066] S226. Conduct a sensitivity analysis on the fusion weight value vector F. By perturbing each feature weight and calculating the influence degree on the result, evaluate the stability influence of each feature, and obtain the stability coefficient vector S.

[0067] S227. Based on the fusion weight value vector F and the stability coefficient vector S, design a stability enhancement function g to calculate the weight value after stability adjustment: g j = F j ·(1+β·S j ), where β is the stability adjustment coefficient. The stable adjustment weight vector G is obtained. Considering the stability of the weights enhances the robustness of the model.

[0068] S228. Conduct normalization processing on the stable adjustment weight vector G: w j = g j / ∑g j, ensure that the sum of all weights is 1 to obtain the final initial feature weight vector W0. This weight vector breaks through the limitation of equal weight processing in traditional multiple correspondence analysis and can more accurately reflect the actual influence ability of different features on the risk occurrence mode.

[0069] According to one aspect of the present application, step S23 is specifically as follows:

[0070] S231. Read out the risk occurrence mode classification system Y and the high dam risk occurrence case data set, group the cases according to the main risk occurrence modes (such as cracks, abnormal deformations, seepage, etc.) to form m mode subsets C1, C2,..., Cm, and each subset contains all cases with the same main risk occurrence mode.

[0071] S232. For each mode subset Ci, extract the case ID list and corresponding feature parameters it contains, and construct a mode-specific feature parameter sub-matrix X i .

[0072] S233. For each feature parameter sub-matrix X i , calculate the point-biserial correlation coefficient between each feature and the sub-mode therein, and construct a mode-specific feature correlation sub-matrix R i . The correlation analysis of the risk occurrence mode stratification is realized, and the difference in feature importance under a specific risk occurrence mode can be captured.

[0073] S234. For each feature correlation sub-matrix R i , apply the feature weight calculation method of the foregoing S22 steps (S221 - S228) to obtain the feature weight vector W for the specific risk occurrence mode i i . The weight calculation specific to the risk occurrence mode is realized.

[0074] S235. Analyze the sample distribution characteristics of each risk occurrence mode subset Ci, and calculate the sample size n i , sample coverage rate c i and sample diversity index d i , and construct a mode sample characteristic matrix D. This matrix is used to evaluate the reliability of the weight calculation for each mode subset.

[0075] S236. Based on the mode sample characteristic matrix D, design a reliability evaluation function h, and calculate the reliability coefficient r of each mode weight vector i = h(n i , c i , d i ), and construct a mode weight reliability vector R mode . A weight reliability evaluation mechanism is introduced to solve the uncertainty problem of weight calculation for small sample modes.

[0076] S237. For the pattern weight reliability vector R mode For the pattern j with reliability lower than the threshold λ, adopt the transfer learning strategy to transfer weight information from similar patterns for reinforcement: W j ' = W j + γ·∑(sim(j,k)·(W_k - W j ))), where sim(j,k) is the similarity between pattern j and k, and γ is the transfer intensity coefficient. Obtain the reinforced weight vector set W'. Solve the problem of weight calculation for rare risk patterns and enhance the applicable scope of the method.

[0077] S238. Combine the feature weight vectors (W i or the reinforced W j ') of all patterns into a pattern-feature two-dimensional matrix to form a complete pattern feature weight matrix WM. Each row of this matrix represents a risk pattern, and each column represents a feature parameter. The matrix element WM ij represents the weight value of feature j in risk pattern i.

[0078] S239. Based on the pattern feature weight matrix WM, calculate the coefficient of variation CV of each feature weight among different patterns, identify pattern-sensitive features (with large CV values) and pattern-stable features (with small CV values), and construct the feature pattern sensitivity vector S feature . This vector provides important reference information for subsequent weight optimization for specific high dams.

[0079] According to one aspect of the present application, step S25 is specifically as follows:

[0080] S251. Read the feature parameter matrix X and the risk pattern classification system Y, analyze the physical mechanism of high dam failure, determine the key physical quantities affecting high dam safety, including stress field, deformation field, seepage field, temperature field, etc., and form the key physical quantity set K.

[0081] S252. For each physical quantity in the key physical quantity set K, define its similarity measurement standard, design a similarity calculation function, and form the physical similarity measurement function set Φ. This function set includes the stress distribution similarity function φ σ , the deformation pattern similarity function φ ε , the seepage field similarity function φ f , etc. Each function can convert the characteristics of the corresponding physical field into a similarity value between 0 and 1.

[0082] S253. Based on the high dam mechanics theory and historical case analysis, identify the key feature parameter combinations affecting the distribution of each physical quantity, and establish the mapping relationship set {M σ , M ε , Mf ,...}. For example, the similarity of stress distribution is mainly affected by parameters such as dam type, dam height, and dam thickness ratio; the similarity of seepage field is mainly affected by parameters such as permeability coefficient, anti-seepage facilities, and geological conditions. A mapping mechanism from the feature space to the physical space is established.

[0083] S254. For the similarity of stress distribution (σ similarity), extract the feature subset X that affects the stress distribution from the feature parameter matrix X σ , including dam type coefficient, dam height, thickness-to-height ratio, bending radius, etc., and apply the non-linear mapping function M σ to convert it into a stress field feature vector, and then calculate the stress similarity index P through the similarity function φ σ . This index can quantify the similarity degree of stress distribution patterns of different dams, providing a mechanical perspective for the safety evaluation of high dams. σ . This index can quantify the similarity degree of stress distribution patterns of different dams, providing a mechanical perspective for the safety evaluation of high dams.

[0084] S255. For the similarity of deformation mode (ε similarity), extract the feature subset X that affects the deformation characteristics from the feature parameter matrix X ε , including material elastic modulus, Poisson's ratio, temperature deformation coefficient, etc., and apply the deformation mapping function M ε to convert it into a deformation field feature vector, and then calculate the deformation similarity index P through the similarity function φ ε . This index can compare the similarity of deformation modes of different high dams, especially suitable for evaluating the risk modes of abnormal deformation. ε . This index can compare the similarity of deformation modes of different high dams, especially suitable for evaluating the risk modes of abnormal deformation.

[0085] S256. For the similarity of seepage field (Φ similarity), extract the feature subset X that affects the seepage characteristics from the feature parameter matrix X f , including permeability coefficient, anti-seepage measures, geological conditions, etc., and apply the seepage mapping function M f to convert it into a seepage field feature vector, and then calculate the seepage similarity index P through the similarity function φ f . This index can evaluate the similarity degree of seepage characteristics of different high dams, which is of great significance for analyzing the risk of leakage. f . This index can evaluate the similarity degree of seepage characteristics of different high dams, which is of great significance for analyzing the risk of leakage.

[0086] S257. Combining the theories of materials science and damage mechanics, design a damage evolution similarity index (D similarity), extract the feature subset X_d that affects the development of material damage from the feature parameter matrix X, including material type, service time, load history, etc., and calculate the damage similarity index P_d by applying the damage mapping function M_d. It can evaluate the similarity of service states of high dams from the perspective of material damage.

[0087] S258. For each pair of high dams (i, j), calculate the values of each physical similarity index to form a physical similarity index vector P ij = [Pσij , P εij , P fij , P_d ij ,…]. Combine the physical similarity index vectors of all high - dam pairs into a three - dimensional tensor, and convert it into a two - dimensional matrix through dimensionality compression to construct a complete physical similarity index matrix P. Each row in the matrix corresponds to a high - dam case, and each column corresponds to a physical similarity index.

[0088] S259. Analyze the correlation between the indicators in the physical similarity index matrix P, identify redundant indicators and complementary indicators, and optimize the physical similarity index system. Finally, determine the mapping relationship F between physical indicators and basic characteristic parameters: characteristic parameter matrix X → physical similarity index matrix P. This mapping relationship F includes the integration of mapping functions for each physical quantity, and can convert the engineering characteristics of high dams into comparable physical similarity indicators, providing a scientific basis for subsequent screening of similar projects based on physical mechanisms.

[0089] According to one aspect of the present application, step S26 is specifically as follows:

[0090] S261. Read the characteristic parameter matrix X and the mapping relationship F, apply physical mapping transformation to the characteristic parameters of each historical case, calculate its physical similarity index value, and construct the original physical similarity index matrix P.

[0091] S262. Analyze the differences between modern extra - high dams and high dams in the historical case library in terms of structural design, material properties, construction technology, etc., identify key difference features, and construct a set D of structural difference features. This set includes unique features of modern extra - high dams, such as new dam - type structures, high - performance concrete materials, special construction technologies, etc., which may lack direct counterparts in historical cases.

[0092] S263. For each difference feature d in the set D of structural difference features i , analyze its influence mechanism on each physical similarity index, and construct a difference - influence mapping table M_diff. This table describes the influence direction and intensity of each structural difference feature on each physical similarity index. Quantifies the structural representative differences between modern extra - high dams and historical cases.

[0093] S264. Based on the difference - influence mapping table M_diff, design a parameterized structural difference compensation function H. This function takes the original physical similarity index P and the structural difference feature d i as inputs, and outputs the corrected physical similarity index P': P' = H(P, d_1, d_2,..., d_k). The function H adopts a piece - wise continuous non - linear mapping form, and can provide adaptive compensation for different degrees of structural differences.

[0094] S265. Use a high-precision finite element model to simulate and analyze the effects of typical structural differences, and determine the parameter values in the structural difference compensation function H. By comparing the physical field distributions under different structural designs, quantify the physical similarity deviation caused by structural differences, optimize the compensation function parameters, and obtain the calibrated structural difference compensation function H cal .

[0095] S266. For the new dam type structures in modern super-high dams (such as double-curvature arch dams, RCC gravity dams, etc.), design the dam type structure compensation sub-function H dam , and focus on adjusting the similarity indexes of stress distribution and deformation mode related to the dam type. By comparing the mechanical behavior differences between the new dam type and the traditional dam type, determine the compensation coefficients, and construct the dam type compensation coefficient matrix C dam .

[0096] S267. For the characteristics of high-performance materials (such as low-heat cement, high-strength concrete, etc.), design the material property compensation sub-function H mat , which mainly adjusts the similarity indexes of damage evolution and temperature field related to material properties. Based on material science theories and experimental data, establish the corresponding relationship between material property parameters and physical behaviors, and form the material compensation coefficient matrix C mat .

[0097] S268. Considering the improvement of modern construction technologies and quality control levels, design the construction quality compensation sub-function H con , which adjusts the similarity indexes of seepage field and interface performance related to construction quality. By comparing the quality differences between modern construction and traditional construction, determine the compensation coefficients, and establish the construction compensation coefficient matrix C con .

[0098] S269. Integrate each special compensation sub-function into the main compensation function H cal , and perform compensation and correction on each index value in the physical similarity index matrix P to obtain the corrected physical similarity index matrix P'. This matrix reflects the physical similarity between historical cases and modern super-high dams considering structural differences, overcomes the limitations of relying only on the comparison of original characteristic parameters, and provides physical mechanism support for accurately screening similar projects.

[0099] According to one aspect of the present application, step S29 is specifically as follows:

[0100] S291. Read the weighted Euclidean distance DX in the original feature space and the weighted Euclidean distance DP in the physical similarity space, and construct the double-space distance vector V dist = [DX, DP]. This vector contains the similarity distances of high dams measured from two different perspectives.

[0101] S292. Design distance-similarity conversion functions f and g to convert distance metrics into similarity metrics: SX = f(DX), SP = g(DP). The conversion functions adopt an exponential decay form: f(d) = exp(-d 2 / σ 2 ), where σ is the scale parameter, and the optimal value is determined through cross-validation. The standardization of different spatial distance metrics is achieved, enabling their effective comparison and fusion.

[0102] S293. Analyze the data distribution characteristics of different types of high dams in the high dam failure case dataset, evaluate the data quality and representativeness of the original feature space and the physical similarity space, and construct a spatial credibility evaluation index set Q = [QX, QP]. The index set includes evaluation indexes in multiple dimensions such as data coverage, sample consistency, and feature integrity.

[0103] S294. Based on the spatial credibility evaluation index set Q, design an adaptive weight factor calculation function h to determine the fusion weight α = h(QX, QP) of the original feature space and the physical similarity space. This function can dynamically adjust the fusion weight according to the data quality of the two spaces, improving the reliability of similarity calculation. It solves the problem of uneven data space quality.

[0104] S295. For different failure modes, design a dedicated spatial weight adjustment coefficient β i , and adjust it according to the degree of dependence of the failure mode on the physical mechanism: α i = α·β i . For example, for the crack failure mode related to stress, increase the weight of the physical similarity space; for the erosion failure mode related to the environment, increase the weight of the original feature space. Construct a mode space weight adjustment matrix B to achieve spatial fusion specialized for failure modes.

[0105] S296. For the high dam to be evaluated, according to the preliminary risk assessment results, determine the risk probability vector p = [p1, p2, ..., pm] of each failure mode. The risk probability can be obtained through preliminary monitoring data analysis, expert scoring, or historical case statistics. Combine the risk probability vector p with the mode space weight adjustment matrix B to calculate the comprehensively adjusted adaptive weight factor: α adj = ∑(pi·α i ), and this weight comprehensively considers the data quality and the characteristics of the failure mode.

[0106] S297. Design a multi-scale similarity fusion function, comprehensively consider the similarity SX of the original feature space and the similarity SP of the physical similarity space, and calculate the final similarity S = α adj ·SX + (1 - α adj)·SP. For special cases, a non-linear fusion strategy is introduced, and methods such as geometric weighted average or harmonic weighted average are used to enhance the fusion effect, and an enhanced multi-scale fusion similarity S is constructed enh .

[0107] S298. For the multi-objective optimization requirement, a hierarchical similarity calculation method is designed, and the similarity S is decomposed into multiple sub-similarity indexes: S struct (structural similarity), S mat (material similarity), S env (environmental similarity), etc., and a multi-dimensional similarity vector S multi is constructed. It can depict the similarity relationship between different high dams in detail from multiple dimensions and support more refined screening of similar projects.

[0108] S299. Based on the final similarity S or the enhanced multi-scale fusion similarity S enh , sort the historical cases, select the k projects with the highest similarity as the core analogy objects, and generate a weighted list of similar projects in sorted order. At the same time, based on the multi-dimensional similarity vector S multi , provide similar project recommendations by category, form a multi-dimensional similar project analysis report, and provide a comprehensive and reliable reference basis for subsequent project analogy analysis. It makes full use of the advantages of the original feature space and the physical similarity space, and improves the reliability and adaptability of similar project screening.

[0109] According to one aspect of the present application, step S33 is specifically as follows:

[0110] S331. Screen a case subset containing complete time records from the high dam failure case dataset to form a time-series complete case set T_cases. It contains cases with clear records of the start time and end time of each failure state, providing a data basis for time-series analysis.

[0111] S332. For each case in the time-series complete case set T_cases, based on the state space S and the state transition probability matrix P, identify the state sequence and state transition process it has experienced, extract the transition time data of each pair of adjacent states Si→Sj, and construct a state transition time dataset T data . This dataset records the actual occurrence time of each state transition in historical cases and is the basic data for fitting the time distribution function.

[0112] S333. Perform data cleaning and outlier detection on the state transition time dataset T data , use the box plot method or the Z-score method to identify abnormal time records, replace the outliers with reasonable estimated values or mark them as missing, and obtain the cleaned time dataset T clean . It improves the data quality of time distribution fitting.

[0113] S334. For the time dataset T after cleaning clean For each pair of state transitions Si→Sj in it, extract its transition time samples T ij = {t1, t2, ..., tn}, analyze the data distribution characteristics, use methods such as the Shapiro-Wilk test to test its distribution type, determine the appropriate theoretical distribution model, and construct a distribution type mapping table M dist . Optional distribution types include exponential distribution, Weibull distribution, gamma distribution, lognormal distribution, etc., which are suitable for describing different types of failure processes.

[0114] S335. Based on the distribution type determined by the distribution type mapping table M dist For the time samples T of each pair of state transitions ij Perform parameter estimation, adopt maximum likelihood estimation (MLE) or method of moments to estimate the distribution parameters, obtain the distribution parameter set {θ ij}, and construct a distribution parameter table P dist . For example, for the exponential distribution, estimate the rate parameter λ ij ; for the Weibull distribution, estimate the shape parameter k ij and the scale parameter λ ij . It realizes the conversion from discrete time data to a continuous distribution model.

[0115] S336. For state transition processes with insufficient sample size (such as rare risk occurrence modes or unique states of new high dams), design a parameter transfer learning method, draw parameter information from similar state transitions, adopt Bayesian inference method to estimate the prior distribution, and update it in combination with limited sample data to obtain the posterior distribution, forming a sparse state transition parameter estimation set P_sparse. It solves the problem of time distribution modeling for data sparse states.

[0116] S337. Based on physical failure mechanisms and reliability theory, design a physical constraint correction function to perform physical rationality verification and correction on the statistically estimated distribution parameters, ensure that the time distribution conforms to physical laws, and obtain a physically corrected parameter set P phy . For example, for the crack propagation process caused by material fatigue, correct the distribution parameters to conform to Paris' law; for the seepage erosion process, correct the parameters to conform to the seepage evolution law.

[0117] S338. For each state transition process Si→Sj, based on its distribution type and parameters, calculate key time characteristic quantities, including the mean μ ij (expected transition time), variance σ ij 2 (time volatility), median m ij (typical transition time) and quantile q ij, α (reliability interval), to form the time feature matrix T feat . This matrix comprehensively describes the time characteristics of each state transition process, providing a quantitative basis for risk assessment and intervention decision-making.

[0118] S339. Integrate the time distribution functions and their parameters of all state transition processes to construct a complete state duration distribution matrix D. Each element D of this matrix ij contains the time distribution function T ij (t) and its parameters. It transforms the static failure path network into a dynamic model with a time dimension, realizing for the first time the temporal expression of the high dam failure process and providing a mathematical basis for subsequent temporal evolution analysis and risk warning.

[0119] According to one aspect of the present application, step S35 is specifically as follows:

[0120] S351. Read the external factor data matrix F, the state transition probability matrix P, and the state duration distribution matrix D, determine the key state transition processes that need to establish conditional relationships, and form a key transition process set K trans . This set contains state transition processes that have a significant impact on the safety of high dams, such as crack initialization → crack propagation, abnormal seepage → piping formation, etc.

[0121] S352. Conduct an importance analysis of the factors in the external factor data matrix F, use statistical measures such as mutual information (MI) and correlation ratio (CR) to evaluate the influence degree of each factor on the state transition probability and time distribution, select k factors with significant influence, and construct a key external factor set F key = {F1, F2, ..., Fk}. Key factors usually include water level changes, temperature cycles, seismic activities, load history, etc.

[0122] S353. Design a data stratification strategy. According to the value ranges of the factors in the key external factor set F key , divide the factor space into multiple intervals to form a multi-dimensional grid, and construct a factor stratification framework G. This framework discretizes the complex multi-factor space into a finite number of factor combination regions, facilitating the establishment of conditional relationships.

[0123] S354. For each key state transition process Si→Sj and each factor combination region g in the factor stratification framework G, screen the state transition cases under the corresponding conditions from the high dam accident case dataset, count the state transition frequencies, calculate the conditional transition probability P(Sj|Si,g), and construct a discrete conditional probability table CP discrete . Realize the preliminary association between the state transition probability and external factors.

[0124] S355. For areas with insufficient discrete sampling points, design a conditional probability interpolation function. Based on physical laws and mathematical models, achieve continuous estimation of probability in the factor space, and construct a continuous conditional probability model CP. continuous . The interpolation method uses non-parametric models such as radial basis function (RBF) networks or Gaussian process regression (GPR), which can effectively handle high-dimensional sparse data. It solves the problem of incomplete data coverage.

[0125] S356. Analyze the interaction between external factors, design an interaction term modeling strategy, capture the non-linear effects under the combined action of multiple factors, and form an interaction effect model I. model . For example, the combined action of water level change and temperature fluctuation may have a greater impact than the individual actions, and this interaction effect is modeled through cross terms or kernel methods.

[0126] S357. Integrate the discrete conditional probability table CP. discrete 、the continuous conditional probability model CP. continuous and the interaction effect model I. model to construct a complete conditional state transition probability function P. ij (F1,F2,...,Fk). This function adopts a piecewise continuous mathematical form, which can calculate the corresponding state transition probability for any given combination of external factors. It realizes the dynamic modulation of the external factors on the transition probability.

[0127] S358. Similarly, based on the time parameters (such as the mean μ ij 、variance σ ij 2 ) in the state duration distribution matrix D and the factor stratification framework G, analyze the relationship between the time distribution parameters and external factors, and establish parameter mapping functions μ ij =g(F1,F2,...,Fk) and σ ij 2 =h(F1,F2,...,Fk). The mapping functions are realized by technologies such as multiple regression, response surface method or neural network, which can predict the state transition time characteristics under different external conditions.

[0128] S359. Verify the performance of the conditional transition probability and time parameter mapping models, use the cross-validation method to evaluate the model prediction accuracy, and determine the response characteristics of the model to external factor changes through sensitivity analysis to form the final non-stationary Markov model parameter set Θ. This parameter set completely describes the variation law of the state transition characteristics with external factors, provides a mathematical tool for simulating the high dam failure process under complex working conditions, and is the core to realize non-stationary failure evolution prediction.

[0129] According to one aspect of the present application, step S36 is specifically as follows:

[0130] S361. Extract the main failure mode types from the high dam failure case dataset and the failure mode classification system Y, including the crack mode (CM), deformation mode (DM), seepage mode (FM), etc., to form the failure mode type set MODE = {CM, DM, FM, ...}. Each failure mode represents a typical physical manifestation of high dam failure.

[0131] S362. For each failure mode m in the failure mode type set MODE, extract the state nodes and transition relationships related to this mode from the initial failure path network Z, construct the mode-specific sub-network Zm, and form the mode sub-network set Z sub = {ZCM, ZDM, ZFM, ...}. Each sub-network focuses on describing the evolution process of a specific failure mode.

[0132] S363. For each mode sub-network Zm, based on the high dam engineering mechanics theory and historical case analysis, supplement the state nodes and transition relationships that may exist but are not fully reflected in the data, and optimize the network topology structure to form the enhanced version mode enhanced sub-network Z'm. This makes up for the deficiencies of historical data and enables the sub-network to more completely describe the failure evolution process.

[0133] S364. Define network parameters for each mode enhanced sub-network Z'm, including node attributes (such as state duration, hazard level) and edge attributes (such as transition probability, transition time), to construct the mode network parameter set Pm. These parameters transform the mode sub-network from a qualitative description to a quantitative model, capable of supporting numerical simulation and quantitative analysis.

[0134] S365. Take all mode enhanced sub-networks Z'm as independent layers, construct a multi-layer network structure, and form the initial multi-layer failure mode network M0. This network separates different types of failure modes into relatively independent layers, facilitating the analysis of the internal evolution laws of each mode, and at the same time providing an infrastructure for subsequent coupling analysis between modes.

[0135] S366. Based on high dam engineering practice and mechanical principles, analyze the physical interaction mechanisms between different failure modes, identify the key inter-mode influence paths, and construct the mode coupling path set L. For example, crack development may affect the seepage path (CM→FM), and abnormal seepage may accelerate material deterioration leading to new cracks (FM→CM). These coupling relationships reflect the mutual promotion or inhibition effects between failure modes.

[0136] S367. For each coupling path l(i,j) (from mode i to mode j) in the mode coupling path set L, design a coupling strength evaluation method, comprehensively consider physical mechanism analysis, historical case statistics, and expert knowledge, and calculate the inter-mode influence strength C ij, construct the failure mode coupling strength matrix C. The coupling strength uses the standardized value in the range of [-1, 1]. A positive value indicates a promoting effect, a negative value indicates an inhibitory effect, and the absolute value indicates the degree of influence.

[0137] S368. Design the coupling dynamics equation to describe the dynamic adjustment mechanism of each state transition probability under the action of mode coupling: P' ij = P ij + ∑(Ckm·f(Sk,Sm)), where f(Sk,Sm) is the interaction function between states. This equation can calculate the corrected transition probability considering mode coupling and construct the coupling adjustment probability model CP. It realizes the quantitative description of the interaction between failure modes on state evolution.

[0138] S369. Integrate the failure mode coupling strength matrix C and the coupling adjustment probability model CP into the multi-layer failure mode network M0, establish the inter-layer connection and dynamic adjustment mechanism, and form the complete multi-layer failure mode network M. This network is a multi-layer, weighted, and dynamic complex network structure, which not only describes the internal evolution law of each failure mode but also depicts the interaction relationship between modes. It breaks through the limitation of traditional single failure path analysis, can capture the synergistic effect and competition relationship between failure modes, and provides a new perspective for understanding and predicting complex failure processes.

[0139] According to one aspect of the present application, step S37 is specifically as follows:

[0140] S371. Read the multi-layer failure mode network M, the failure mode coupling strength matrix C, and the initial failure path network Z, analyze the network structure characteristics, including node connectivity, path length distribution, and key node distribution, and construct the network feature statistic set E. This statistic set describes the topological characteristics of the failure path network and provides a reference for subsequent path generation.

[0141] S372. Design the path generation rule library, including physical rationality rules (constraints based on mechanical principles), historical similarity rules (references based on case patterns), and structural integrity rules (requirements based on network topology), to form the path generation rule set R. The rule set is described in a formal language, such as "if state A appears and the environmental conditions meet X, then it may change to state B". Each rule has a corresponding triggering condition and execution action.

[0142] S373. Based on the path generation rule set R, design a rule execution engine, adopt a forward reasoning mechanism, start from the initial state, and iteratively generate possible state transition sequences according to the rule set to construct the rule-driven path set P_rule. This engine can automatically explore the state space allowed by the rules and discover failure paths that may exist but do not appear in historical data.

[0143] S374. Combined with the failure mode coupling mechanism, a cross-layer path generation algorithm is designed. The algorithm can jump between mode layers and generate a composite path containing multiple failure mode interactions based on the failure mode coupling strength matrix C to form a cross-mode path set P_cross. For example, from a certain state of the fracture layer to the relevant state of the seepage layer through the coupling relationship, and then jump back to the fracture layer, a composite failure path of "fracture-seepage-fracture" is formed. It can discover collaborative failure paths that are difficult to identify in single mode analysis.

[0144] S375. Aiming at the special structural characteristics of modern ultra-high dams, a specialized path generation module is designed. Combining finite element analysis and expert knowledge, a special failure path for ultra-high dam structural characteristics is generated, and a special path set P_high for ultra-high dams is constructed. This module takes into account special factors such as temperature control, segmented casting, and stress distribution of ultra-high dams, and can generate failure paths that are more in line with the actual ultra-high dam project.

[0145] S376. Perform physical rationality verification on the newly generated path sets (rule-driven path set P_rule, cross-mode path set P_cross and special path set P_high for ultra-high dams), and use a combination of mechanical models and expert review to evaluate the physical feasibility of each path, filter out unreasonable paths, and retain reasonable paths to form a verification path set P_valid. Ensure that the automatically generated paths conform to physical laws and avoid invalid paths that violate mechanical principles.

[0146] S377. Design a path scoring function, comprehensively consider physical rationality, historical similarity and risk impact, calculate the importance score for each path in the validation path set P_valid, and construct a path importance ranking table S_path. The scoring function adopts a multi-index weighted summation form, which can identify the key failure paths that pose the greatest threat to high dam safety.

[0147] S378. Based on the path importance ranking table S_path, select the paths with scores exceeding the threshold τ as the core extension paths, merge them with the initial invalid path network Z, delete redundant paths and low-importance paths, optimize the network structure, and form a simplified extension path network Z_opt. This network retains the most valuable extension paths to avoid excessive network complexity that affects computational efficiency.

[0148] S379. Estimate the parameters of the reduced extended path network Z_opt. Based on historical data, physical models, and expert knowledge, assign reasonable transition probabilities and time distribution parameters to the newly added paths to complete the quantification of the network, and finally form a complete extended failure path network Z'. This network not only includes the failure paths observed in historical cases but also integrates the potential paths automatically generated based on rules, greatly expanding the coverage of failure path analysis and being able to handle complex and changeable failure scenarios, especially rare path mutation situations. It breaks through the limitations of the preset paths and provides more comprehensive path coverage for the safety assessment of high dams.

[0149] According to one aspect of the present application, step S38 is specifically as follows:

[0150] S381. Read the state transition probability matrix P, the state duration distribution matrix D, the external factor data matrix F, the multi-layer failure mode network M, and the extended failure path network Z', design a data integration framework, convert data from different sources and in different formats into a unified model input format, and construct a comprehensive model input data set I.

[0151] S382. Design a macroscopic-scale state sequence simulation module. Based on the state transition probability matrix P and the state duration distribution matrix D, use the Monte Carlo method to simulate the state transition sequence and time process of high dam failure, realize the overall evolution simulation from the normal state to the final failure state, and generate a macroscopic evolution model M_macro. This model simulates the overall trend of the failure process through random sampling and can predict the approximate path and time frame of failure.

[0152] S383. Develop a mesoscopic-scale network spread model. Based on the multi-layer failure mode network M and percolation theory, simulate the propagation and spread process of failure in the network, consider the interaction between nodes and the cascade effect, and construct a mesoscopic spread model M_meso. This model regards the failure process as the propagation of information or energy in the network and can capture the spatial distribution characteristics and spread dynamics of failure.

[0153] S384. Construct a microscopic-scale physical evolution model. Combining the principles of material mechanics, fracture mechanics, and fluid mechanics, based on the environmental conditions in the external factor data matrix F, simulate the physical field evolution process in the local area of the high dam, including stress distribution, crack propagation, and seepage change, to form a microscopic physical model M_micro. This model describes the spatio-temporal evolution of the physical field through partial differential equations and can provide a detailed physical mechanism explanation for the failure process.

[0154] S385. Design a multi-scale model coupling framework, construct an information transfer and feedback mechanism among the macro, meso, and micro scale models, realize bidirectional coupling between different scales, and form a multi-scale coupling architecture C. For example, the crack propagation results of the micro physical model can be transmitted to the meso network model for node state update; the failure spread of the meso network can be transmitted to the macro state model as the basis for transfer probability adjustment. It breaks through the limitation of a single-scale model and can provide both the overall trend and local details simultaneously.

[0155] S386. Based on the extended failure path network Z', design a dynamic path priority adjustment algorithm. According to the system state and environmental conditions during the simulation process, evaluate the activation probability of each failure path in real time, dynamically adjust the path weights, and construct a path priority model P_prior. This algorithm can adaptively adjust the importance of the failure paths according to the changes in external conditions and the evolution of the system state, realizing the dynamic optimization of the failure paths.

[0156] S387. Integrate the macro evolution model M_macro, the meso spread model M_meso, the micro physical model M_micro, the multi-scale coupling architecture C, and the path priority model P_prior to construct a complete multi-scale time evolution framework and form a comprehensive time evolution model M_evol. This model can simulate the time evolution, spatial distribution, and physical mechanism of the high dam failure process according to the given initial state and external conditions, providing a comprehensive dynamic perspective for safety assessment.

[0157] S388. Design a model solving engine, adopt hybrid numerical algorithms, including Monte Carlo sampling, Markov chain solving, network dynamics calculation, and finite element analysis, etc., to achieve efficient calculation of the comprehensive time evolution model M_evol and generate a simulation result dataset R_sim. This engine selects suitable solving strategies according to the characteristics of different scale models, optimizing the calculation efficiency and result accuracy.

[0158] S389. Develop a result analysis and visualization module to process the simulation result dataset R_sim, extract key time series features, identify critical states and warning indicators, generate a multi-dimensional visual expression of the failure process, and form a time evolution analysis report A_evol. This report includes failure time prediction, path probability distribution, key node identification, and risk level assessment, etc., providing intuitive and comprehensive information for decision support. It realizes the full-scale, dynamic, and quantitative simulation of the high dam failure process, providing strong technical support for high dam safety assessment.

[0159] According to one aspect of the present application, step S42 is specifically as follows:

[0160] S421. Collect the real-time monitoring data of the high dam to be evaluated, including deformation monitoring, seepage monitoring, stress monitoring, and environmental parameter monitoring data, etc. Conduct data cleaning, outlier processing, and standardization to construct the high dam monitoring dataset M data . This dataset reflects the current working state and environmental conditions of the high dam and is the basic input for failure path prediction.

[0161] S422. Based on the high dam monitoring dataset M data , combined with the high dam design parameters and operation history, evaluate the current state of the high dam, identify potential abnormal features and initial disease signs, and locate to the specific state nodes in the state space S to form the initial state information S i nit. This information contains the current state evaluation results of the high dam and the probability distributions of various possible states, reflecting the uncertainty of the initial state.

[0162] S423. Predict the future external environmental change trends of the high dam to be evaluated, including water level changes, temperature fluctuations, rainfall conditions, and possible extreme events (such as floods, earthquakes, etc.), and construct the future environmental condition prediction set E f uture. This prediction set considers various possible environmental scenarios and provides external condition inputs for subsequent path simulations.

[0163] S424. Read the time-series failure path prediction model (i.e., the comprehensive time-series evolution model M_evol generated in step S38), and use the initial state information S i nit and the future environmental condition prediction set E f uture as inputs to perform multi-scenario failure path simulations, generate a large number of possible failure evolution trajectory samples, and construct the failure trajectory sample set T_sample. This sample set contains thousands of simulated trajectories, covering the possible failure development paths of the high dam under different initial states and environmental conditions.

[0164] S425. Conduct cluster analysis on the failure trajectory sample set T_sample, identify typical failure evolution patterns, extract the representative trajectories and key features of each type of failure pattern, and construct the typical failure pattern set T_typical. This set compresses a large number of trajectory samples into a small number of representative patterns, facilitating subsequent analysis and decision-making.

[0165] S426. For each typical failure pattern, calculate the key time features of the failure process, including the expected duration of each state, the time window of state transition, the time points of key nodes, and the overall failure time distribution, to form the time feature analysis table T feat . This table provides a quantitative description of the failure process in the time dimension and is the core basis for the time-series assessment of failure risk.

[0166] S427. Based on the typical failure mode set T_typical and the time feature analysis table T feat , evaluate the occurrence probability and risk level of each failure mode. Combine with the severity assessment of failure consequences to construct a risk assessment matrix R. This matrix combines the failure probability and consequence severity to give a comprehensive risk level assessment result.

[0167] S428. Design a time-series risk visualization scheme. Integrate the failure path, time window, and risk level information into the same view. Use multi-dimensional visualization techniques (such as timeline charts, heatmaps, Sankey diagrams, etc.) to express the time-series characteristics and risk distribution of the failure process, and generate a preliminary failure risk time-series diagram G_risk.

[0168] S429. Optimize and enhance the failure risk time-series diagram G_risk. Add interactive functions, multi-level detailed information, and warning threshold marks to improve readability and usability, and form a final failure risk time-series diagram. This diagram intuitively shows the possible failure evolution path of the high dam to be evaluated, the time window of each stage, key turning points, and risk level distribution, providing a decision-making basis in the time dimension for safety monitoring and risk management. It breaks through the limitations of traditional static risk assessment, can reveal the dynamic characteristics of the failure process from the time dimension, and provides accurate time window guidance for risk warning and intervention decisions.

[0169] According to one aspect of the present application, step S44 is specifically as follows:

[0170] S441. Read the potential risk occurrence modes, failure risk time-series diagrams, and verification analysis results. Design a multi-source information fusion framework to convert and normalize the evaluation results from different sources according to a unified standard, and construct a comprehensive evaluation input set I_comp. This input set integrates the results of similar project analogy analysis and time-series evolution prediction, providing a rich information basis for comprehensive evaluation.

[0171] S442. Based on the comprehensive evaluation input set I_comp, use the analytic hierarchy process (AHP) to construct a safety evaluation index system, set the index weights, and form an evaluation index system H. This index system includes three major categories of indicators: structural safety, operation reliability, and risk controllability. Each category of indicators has multiple secondary and tertiary indicators, forming a complete evaluation framework.

[0172] S443. For each indicator in the evaluation index system H, conduct quantitative scoring based on the relevant information in the comprehensive evaluation input set I_comp. Use the fuzzy comprehensive evaluation method to handle the uncertainty of the scoring, obtain the fuzzy scoring vector of each indicator, and construct an index scoring matrix S. This matrix reflects the performance of the high dam on each evaluation indicator and is the basic data for comprehensive evaluation.

[0173] S444. Combine the weights of the evaluation index system H and the index scoring matrix S to calculate the comprehensive score of the high dam safety status and the scores of each subsystem. Based on the scoring results, determine the overall safety level and the sub-item safety levels to form the safety level assessment result L. The safety level is usually divided into four levels: normal, attention, warning, and danger, which reflects the current overall safety status of the high dam.

[0174] S445. Based on the failure risk time-series diagram and the verification analysis results, identify the most likely failure modes and weak links of the high dam, analyze their physical causes and development trends, and construct the key risk point analysis table K. This table details the locations, natures, severities, and development rates of each potential risk point, providing a clear target for targeted intervention.

[0175] S446. For each risk point identified in the key risk point analysis table K, design a monitoring enhancement plan, including monitoring point layout, monitoring frequency adjustment, new monitoring items, and warning threshold setting, etc., to form a targeted monitoring plan M. It strengthens the monitoring coverage of key risk points and improves the ability to detect anomalies early.

[0176] S447. Based on the safety level assessment result L and the key risk point analysis table K, combined with the actual engineering conditions and technical and economic feasibility, design a hierarchical and phased intervention strategy, including immediate intervention measures, medium-term reinforcement plans, and long-term maintenance plans, to construct the intervention strategy matrix I. This matrix provides differentiated intervention plans for different risk levels and problem types, considering both the rapid response to emergency risks and the systematic solution to long-term safety.

[0177] S448. Design a risk dynamic monitoring framework. Based on the time-series failure path prediction model and the targeted monitoring plan M, establish a real-time data-driven risk assessment update mechanism, determine the regular assessment period and trigger-based assessment conditions, to form a dynamic risk monitoring plan D. This plan establishes a closed-loop feedback between monitoring data and risk assessment, realizing continuous tracking and timely update of the high dam safety status.

[0178] S449. Integrate the safety level assessment result L, the key risk point analysis table K, the targeted monitoring plan M, the intervention strategy matrix I, and the dynamic risk monitoring plan D to compile a structured high dam safety status assessment report.

[0179] Taking a certain arch dam as the research object, conduct the diagnosis of the accident mode and failure path. A certain arch dam is a 300m-class ultra-high arch dam with a relatively short operation time and insufficient accumulation of monitoring information. It is necessary to carry out safety assessment by means of engineering analogy and failure path analysis. The existing medium-weight multiple correspondence analysis and static failure path analysis methods in the technology cannot meet its assessment requirements.

[0180] Step S1. Construction of the high dam accident case database

[0181] Step S11: Collect the basic information and failure records of 235 high dam failure cases at home and abroad from engineering literature, accident reports, and monitoring records, including dam type, dam height, reservoir capacity, construction year, country where located, as well as failure time, failure type, failure location, treatment measures, etc., to form the original case information. The collected cases include 73 arch dams, 89 gravity dams, and 73 earth-rock dams, with the construction year spanning from 1920 to 2015.

[0182] Step S12: Clean the data of the original case information, and process missing values and outliers. For some missing parameters, such as material parameters, the average values of the same type of high dams are used for supplementation; for outliers, the 3σ principle is used for identification and processing. Unify the parameter units, such as unifying the dam height to meters and the reservoir capacity to billions of cubic meters. After processing, standardized case data is formed, containing 235 complete records.

[0183] Step S13: Based on the standardized case data, extract 27 characteristic parameters for each case: structural characteristics, including dam type coefficient (1.0 - 1.5), dam height (30m - 300m), crest length of the dam, bottom width of the dam, dam thickness coefficient (0.1 - 0.5), reservoir capacity (0.15 - 50 billion m³); material characteristics, including concrete strength grade, elastic modulus, Poisson's ratio, permeability coefficient; environmental characteristics, including geological conditions (divided into 5 levels), climate characteristics (divided into 4 categories), seismic region (divided into 4 zones); service characteristics, including service life, historical water level fluctuations, historical temperature changes; all the extracted characteristic parameters form the characteristic parameter matrix X, with the matrix dimension of 235×27, where each row represents a high dam case and each column represents a characteristic parameter.

[0184] Step S14: Analyze the failure situations in the standardized case data and identify three major types of failure modes: crack mode (112 cases): including 5 subtypes such as heel crack of the dam, surface crack of the dam, and crest crack of the dam; abnormal deformation mode (78 cases): including 4 subtypes such as differential settlement, excessive deformation, and top displacement; seepage mode (45 cases): including 6 subtypes such as dam body leakage, dam foundation leakage, and joint leakage; construct a failure mode classification system Y, adopting a two-level classification structure, with the first level being the main failure mode and the second level being the specific failure type.

[0185] Step S15: Construction of the failure path network

[0186] Step S151: Extract the failure event sequence of each case from the standardized case data, including failure phenomena, occurrence time, development status, and influence range, to form the failure event time-series data E. For example, the Kolnbrein arch dam records the cracks that appeared at the heel of the dam during the first impoundment in 1977, as well as the complete process of subsequent crack propagation and increased leakage.

[0187] Step S152: Perform time series decomposition on the time series data E of the accident events to identify the starting event, intermediate development event, and final result event in each case. Taking the Kolnbrein arch dam as an example, the event chain is identified: initial water storage → stress concentration at the dam heel → cracking at the dam heel → crack propagation → increased leakage → reduced structural safety factor. Perform similar analysis on all cases to construct the case-level event chain set L.

[0188] Step S153: Perform semantic normalization on all events in the event chain set L to merge and unify events with different expressions but the same essence. For example, unify "cracks at the dam heel", "bottom cracking", "tensile cracks at the dam heel", etc. into the "cracking at the dam heel" event. After normalization, the standardized event type set T is obtained, which contains 45 standard event types.

[0189] Step S154: Based on the event type set T, analyze the forward and backward dependence relationships between events in the event chain set L, and count the transfer frequencies between event pairs to form the event transfer frequency matrix F. For example, the transfer frequency from "cracking at the dam heel" to "crack propagation" is 56 times, and the transfer frequency from "crack propagation" to "increased leakage" is 42 times.

[0190] Step S155: Perform normalization on the event transfer frequency matrix F, and calculate the conditional probability P(j|i) = f ij / ∑fik; where: P(j|i) is the conditional probability that event j occurs after event i occurs; f ij is the historical occurrence times from event i to event j; ∑fik represents the total occurrence times from event i to all possible events k; i, j, k are event indices. For example, the conditional probability P("crack propagation"|"cracking at the dam heel") = 56 / 75 = 0.747 from "cracking at the dam heel" to "crack propagation". Calculate the conditional probabilities of all event pairs to construct the event transfer probability matrix P event .

[0191] Steps S156 - S159: Based on the event transfer probability matrix P event , use the minimum spanning tree algorithm to extract the main event transfer paths, eliminate the transfer relationships with probabilities lower than the threshold θ = 0.15 to obtain the core event transfer network N core . Combine the high dam engineering mechanics theory and expert knowledge to verify the physical rationality of the network, and add the transfer edges that physically must exist but are missing in the data to form the physically corrected event network N phy . Classify the events in two dimensions according to the accident type and development stage to construct the event classification matrix C, and based on this matrix and the corrected network, construct the final initial failure path network Z. This network is a directed weighted graph structure, with event types as nodes, event transfer relationships as edges, and the edge weights as transfer probabilities.

[0192] Step S16: Integrate the characteristic parameter matrix X, the accident mode classification system Y and the initial failure path network Z into a structured database to form a complete high dam accident case data set. The data set adopts a relational database structure, including a case basic information table, a characteristic parameter table, an accident mode table and a failure path table, and establishes an association relationship through the case ID.

[0193] Implementation step S2: Multi-scale hierarchical adaptive correspondence analysis based on physical mechanism

[0194] Step S21: extract the characteristic parameter matrix X from the high dam accident case data set, and calculate the point bi-column correlation coefficient ρ between each characteristic parameter and the accident mode ij , construct the feature correlation matrix R. Point bi-column correlation coefficient calculation formula ρ ij = (φ ij -φi+φ+j) / √(φi+(1-φi+)φ+j(1-φ+j)); where: ρ ij is the point biserial correlation coefficient between feature j and accident mode i; φ ij is the proportion of cases that simultaneously meet feature j and failure mode i; φi+ is the marginal proportion of failure mode i; φ+j is the marginal proportion of feature j; i is the failure mode index; j is the characteristic parameter index. For example, the correlation coefficient between the dam type coefficient and the failure mode of dam heel cracks is calculated, and ρ11 = 0.78 is obtained, indicating that the two have a strong positive correlation.

[0195] Step S22: Calculation of feature weights

[0196] Steps S221-S222: read the feature correlation matrix R, extract the correlation coefficient between each feature parameter j and all accident patterns, and form a feature-accident correlation coefficient set ρ_set. Calculate the average correlation coefficient ρ_avg of each feature j j = (1 / M)∑ρ ij ; Among them: ρ_avg j is the average correlation coefficient of feature j; ρ ij is the point-wise bi-column correlation coefficient between feature j and accident mode i; M is the total number of accident modes; ∑ represents the summation operation from i=1 to M. For the dam type coefficient (j=1), the average correlation coefficient ρ_avg_1 = 0.76 is calculated; for the dam height ratio (j=2), the average correlation coefficient ρ_avg_2 = 0.82 is calculated; for the elastic modulus ratio (j=3), the average correlation coefficient ρ_avg_3 = 0.65 is calculated. The average correlation coefficients of all features are summarized to form the average correlation coefficient vector ρ_avg.

[0197] Steps S223 - S224: Conduct a significance test on the values in the average correlation coefficient vector ρ_avg, calculate the p - value, screen out the features with p > 0.05, and retain 21 significantly correlated features to form a significant feature index set J_sig. For the features in the significant feature index set J_sig, introduce 10 experts to conduct importance scoring (0 - 1 point), and calculate the average score as the expert weight coefficient e j , and construct an expert weight vector E. For example, the expert weight of the dam type coefficient e_1 = 0.85, and the expert weight of the dam height ratio e_2 = 0.92.

[0198] Steps S225 - S228: Design a fusion function to perform weighted fusion on the data - driven average correlation coefficient vector ρ_avg and the expert weight vector E: The fusion weight calculation formula f j = α·ρ_avg j + (1 - α)·e j ; where: f j is the fusion weight value of feature j; ρ_avg j is the average correlation coefficient of feature j; e j is the expert weight coefficient of feature j; α is an adaptive fusion coefficient, and its value range is [0, 1].

[0199] According to the data quality assessment, determine α = 0.7. For the dam type coefficient (j = 1), calculate the fusion weight f_1 = 0.7×0.76 + 0.3×0.85 = 0.787. Calculate the fusion weights of all features to form a fusion weight value vector F.

[0200] Conduct sensitivity analysis and stability adjustment, and finally perform normalization on the stable - adjusted weight vector G to obtain the initial feature weight vector W0, where the weight of the dam type coefficient w_1 = 0.092, the weight of the dam height ratio w_2 = 0.105, and the weight of the elastic modulus ratio w_3 = 0.078.

[0201] Step S23: Stratified weighted for the failure modes

[0202] According to the failure mode classification system Y, divide 235 high - dam cases into three categories: crack mode (C1, 112 cases), abnormal deformation mode (C2, 78 cases), and seepage mode (C3, 45 cases), and respectively construct a feature parameter sub - matrix X i specific to the mode. For each feature parameter sub - matrix X i , calculate the correlation coefficient between the feature and the sub - mode, and construct a feature correlation sub - matrix R i specific to the mode. For each feature correlation sub - matrix R i, using the feature weight calculation method in step S22, obtain the feature weight vector W for a specific risk occurrence mode i i .

[0203] For example: The feature weight vector W_1 of the crack mode (i = 1) focuses on stress-related features such as the dam type coefficient (w_11 = 0.126), the dam height ratio (w_12 = 0.135), and the bending radius (w_13 = 0.114). The feature weight vector W_3 of the seepage mode (i = 3) focuses on permeability-related features such as the material permeability coefficient (w_31 = 0.152), the anti-seepage measure (w_32 = 0.138), and the geological condition (w_33 = 0.127).

[0204] Analyze the sample characteristics of each mode subset and calculate the sample size n i , the sample coverage rate c i and the sample diversity index d i , construct the mode sample characteristic matrix D. Based on the mode sample characteristic matrix D, evaluate the reliability coefficient r of each mode weight vector i , construct the mode weight reliability vector R_mode. For the mode j with a reliability lower than the threshold λ = 0.6, such as the seepage mode (r_3 = 0.58), adopt the transfer learning strategy to reinforce the weight W j ' = W j + γ·∑(sim(j,k)·(W_k - W j )); where: W j ' is the reinforced weight vector; W j is the original weight vector; sim(j,k) is the similarity between mode j and k; γ is the transfer intensity coefficient; ∑ represents the summation over all similar modes k. For the seepage mode, adopt γ = 0.4, transfer the weight information from the crack mode and the deformation mode, and obtain the reinforced weight vector W_3'. Combine the feature weight vectors of all modes into an m×n-dimensional matrix to form the complete mode feature weight matrix WM. Each row of this matrix represents a risk occurrence mode, and each column represents a feature parameter. The matrix element WM ij represents the weight value of feature j under the risk occurrence mode i. Calculate the coefficient of variation CV of each feature weight among different modes, identify the mode-sensitive features (large CV values) and the mode-stable features (small CV values), and construct the feature mode sensitivity vector S feature .

[0205] Step S24, construction of the risk occurrence mode correlation matrix

[0206] Analyze the relationships between different risk-occurrence modes, calculate the conversion probabilities and co-occurrence frequencies between modes, and construct the risk-occurrence mode correlation matrix R. For example, the correlation coefficient r13 between the crack mode and the seepage mode is 0.65, indicating a strong correlation between the two modes.

[0207] Step S25: Construction of the physical similarity index system

[0208] Analyze the physical mechanism of high dam failure, select four key physical quantities: stress field (σ), deformation field (ε), seepage field (Φ), and damage evolution (D), and construct the set K of key physical quantities. For each physical quantity, design a similarity measurement function to form the set Φ of physical similarity measurement functions. For example, the stress distribution similarity function φ σ , the deformation mode similarity function φ ε etc.

[0209] Identify the key characteristic parameter combinations that affect the distribution of each physical quantity, and establish the mapping relationship set {M σ , M ε , M f , M_d} between the characteristic parameters and the physical quantities. For example:

[0210] Stress mapping M σ : Mainly consider parameters such as dam type, dam height, thickness-height ratio, and bending radius;

[0211] Seepage mapping M f : Mainly consider parameters such as permeability coefficient, anti-seepage facilities, and geological conditions;

[0212] For the stress distribution similarity (σ similarity), design the calculation function P σ (i,j) = exp(-||Γσ(X σi ) - Γσ(X σj )|| 2 / 2θσ 2 ); where: P σ (i,j) is the stress distribution similarity between high dams i and j; X σi is the stress-related characteristic subset of high dam i; Γσ is the stress field characteristic mapping function; ||·|| represents the Euclidean distance; θσ is the scaling parameter.

[0213] For a certain arch dam and the Kolnbrein arch dam, extract the stress-related characteristics: a certain arch dam X σ _1 = [1.2, 294,0.32, 456]; Kolnbrein arch dam X σ_2 = [1.15, 200, 0.28, 425]; which respectively represent [dam type coefficient, dam height (m), thickness-height ratio, bending radius (m)]. The stress field characteristic vector is obtained by applying the stress field characteristic mapping function Γσ, and the stress similarity P is calculated. σ (1,2) = 0.875. Similarly, the deformation similarity P is calculated. ε (1,2) = 0.823, the seepage similarity P f (1,2)= 0.762, the damage similarity Pd(1,2) = 0.798. Combine each physical similarity index to form the physical similarity index vector P_12 = [0.875, 0.823, 0.762, 0.798]. By calculating the physical similarity index for all high dam pairs, a complete physical similarity index matrix P is constructed.

[0214] Analyze the correlation between each index in the physical similarity index matrix P, identify redundant indexes and complementary indexes, and optimize the physical similarity index system. Finally, determine the mapping relationship F between physical indexes and basic characteristic parameters: characteristic parameter matrix X → physical similarity index matrix P.

[0215] In steps S26 to S28, take the characteristic parameter matrix X and the mapping relationship F, and calculate the original physical similarity index matrix P. Analyze the structural differences between a certain arch dam as a modern extra-high arch dam and historical cases, identify three key difference characteristics: extra-high dam shape (d_1), low-heat cement material (d_2), and intelligent temperature control construction (d_3), and construct a structural difference characteristic set D. Evaluate the influence of each difference characteristic on each physical similarity index, and construct a difference-influence mapping table M_diff. For example, the influence coefficient c_11 of the extra-high dam shape (d_1) on the stress similarity is 1.15, and the influence coefficient c_12 on the deformation similarity is 1.23.

[0216] Design a structural difference compensation function P'_k(i,j) = P_k(i,j)·∏(1 + δ_km·d_m); where: P'_k(i,j) is the corrected physical similarity index k; P_k(i,j) is the original physical similarity index k; δ_km is the influence coefficient of the difference characteristic m on the physical index k; d_m is the intensity value of the difference characteristic m; ∏ represents the product operation.

[0217] Compensate the stress similarity between a certain arch dam and the Kolnbrein arch dam: the original stress similarity P σ (1,2) =0.875; the extra-high dam shape difference d_1 = 0.45; the influence coefficient δ σ 1 = 0.15; the corrected stress similarity P' σ(1,2) = 0.875×(1 + 0.15×0.45) = 0.934; Similarly, calculate the correction values of other physical indicators to obtain the corrected physical similarity index vector P'_12 = [0.934, 0.893, 0.762, 0.846]. By performing similar compensation on all high dam pairs, the corrected physical similarity index matrix P' is obtained.

[0218] Extract 27 characteristic parameters of a certain arch dam to construct the characteristic vector X0 of the high dam to be evaluated. Apply the physical mapping relationship F and the compensation function H to obtain the physical similarity index vector P0 of the high dam to be evaluated. In the original feature space, calculate the weighted Euclidean distance between a certain arch dam and historical cases, DX(0,j) = √(∑(wi·(X0i - Xji) 2 )); where: DX(0,j) is the distance between the high dam 0 to be evaluated and the historical case j in the original feature space; wi is the weight coefficient of feature i; X0i is the value of feature i of the high dam to be evaluated; Xji is the value of feature i of the historical case j; ∑ represents the summation over all features i. For example, calculate the feature space distance DX(0,2) between a certain arch dam and the Kolnbrein arch dam = 0.263.

[0219] In the physical similarity space, calculate the weighted Euclidean distance between the high dam to be evaluated and historical cases, that is, the physical space distance calculation formula DP(0,j) = √(∑(vk·(P0k - P'jk) 2 )); where: DP(0,j) is the distance between the high dam 0 to be evaluated and the historical case j in the physical similarity space; vk is the weight coefficient of physical similarity index k; P0k is the value of physical similarity index k of the high dam to be evaluated; P'jk is the corrected physical similarity index k value of the historical case j; ∑ represents the summation over all physical indicators k. Calculate the physical space distance DP(0,2) between a certain arch dam and the Kolnbrein arch dam = 0.118.

[0220] Step S29, multi-scale fusion strategy

[0221] Read the original feature space distance DX and the physical similarity space distance DP, and construct the dual space distance vector V dist = [DX, DP].

[0222] Design a distance-similarity conversion function to convert the distance into similarity SX = exp(-DX 2 / 2σX 2 ) and SP = exp(-DP 2 / 2σP 2); where: SX is the similarity of the original feature space; SP is the similarity of the physical similarity space; DX is the distance of the original feature space; DP is the distance of the physical similarity space; σX and σP are scaling parameters.

[0223] For a certain arch dam and the Kolnbrein arch dam, take σX = 0.5 and σP = 0.3. The calculated feature space similarity SX = exp(-0.263 2 / 0.5) = 0.78; the physical space similarity SP = exp(-0.118 2 / 0.3) = 0.89; evaluate the data quality of the two spaces, determine the credibility QX = 0.65 of the original feature space and the credibility QP = 0.82 of the physical space, and calculate the adaptive weight factor α = QX / (QX + QP); where: α is the fusion weight factor; QX is the credibility evaluation index of the original feature space; QP is the credibility evaluation index of the physical similarity space. Calculate α = 0.65 / (0.65 + 0.82) = 0.44.

[0224] According to the risk probability vector p = [0.65, 0.25, 0.10] (representing the risk probabilities of cracks, deformations, and seepage respectively) of a certain arch dam and the pattern space weight adjustment matrix B, calculate the comprehensively adjusted weight factor α adj = 0.42.

[0225] Calculate the multi-scale fusion similarity S(i,j) = α adj ·SX(i,j) + (1 - α adj )·SP(i,j); where: S(i,j) is the final similarity between high dams i and j; SX(i,j) is the similarity of the original feature space; SP(i,j) is the similarity of the physical similarity space; α adj is the comprehensively adjusted adaptive weight factor.

[0226] For a certain arch dam and the Kolnbrein arch dam, calculate the final similarity S(1,2) = 0.42×0.78 + 0.58×0.89 = 0.843.

[0227] For multi-objective optimization requirements, calculate the multi-dimensional similarity vector S multi = [S struct , S mat , S env , and describe the similarity relationship in detail from three dimensions of structural similarity, material similarity, and environmental similarity.

[0228] Sort the historical cases based on the final similarity to obtain the top 5 most similar projects: Kolnbrein Arch Dam (0.843), Mica Arch Dam (0.812), Ertan Arch Dam (0.795), Hoover Arch Dam (0.766), and Mauvoisin Arch Dam (0.754), and form a weighted list of sorted similar projects.

[0229] Implementation step S3: Temporal adaptive network evolution with multiple failure mode coupling

[0230] Step S31: Extract the initial failure path network Z from the high dam failure case dataset, and define the state space S = {S0, S1, S2, ..., Sf}, where S0 represents the normal state, S1 represents the microcrack at the dam heel, S2 represents crack propagation, S3 represents abnormal seepage, and Sf represents the failure state. A total of 12 key state nodes are identified to form the complete state space S.

[0231] Step S32: Based on the historical cases in the high dam failure case dataset, count the transfer frequencies between states and calculate the state transition probability matrix P. For example, P(S1|S0) = 0.08 (probability of the normal state transitioning to a microcrack at the dam heel), and P(S2|S1) = 0.32 (probability of the microcrack at the dam heel developing into crack propagation).

[0232] Step S33: Fitting of the time distribution function

[0233] Filter the cases with complete time records from the high dam failure case dataset to form a time-sequence complete case set T_cases, which contains 87 cases. For each case, identify the state sequence and transfer process it has experienced, extract the transfer time data for each pair of adjacent states Si→Sj, and construct the state transfer time dataset T data 。

[0234] For the state transfer time dataset T data Perform data cleaning to handle outliers. Use the box plot method to identify abnormal time records and replace the values outside the upper and lower limits with critical values to obtain the cleaned time dataset T clean 。

[0235] For each pair of state transfers, extract its time samples and analyze the data distribution characteristics. Use the Shapiro-Wilk test to judge the distribution type and construct the distribution type mapping table M dist 。For example, the time samples of S1→S2 (the microcrack at the dam heel developing into crack propagation) are found to conform to the lognormal distribution after testing; the time samples of S2→S3 (crack propagation developing into abnormal seepage) conform to the Weibull distribution. Perform parameter estimation on the time samples of each pair of state transfers, and use the maximum likelihood estimation method to fit the distribution parameters to form the distribution parameter table P distFor example, the lognormal distribution parameters for S1→S2 are: μ12 = 1.8, σ12 2 = 0.5; the Weibull distribution parameters for S2→S3 are: shape parameter k23 = 2.1, scale parameter λ23 = 2.8. For state transitions with insufficient sample size (such as S3→Sf), the parameter transfer learning method is adopted to draw on parameter information from similar state transitions and estimate the parameters: μ3f = 2.3, σ3f 2 = 0.7. Based on the physical failure mechanism, the physically reasonable verification and correction of the statistically estimated distribution parameters are carried out to ensure that the time distribution conforms to the physical laws, and the physically corrected parameter set P is obtained phy 。

[0236] For each state transition process, based on its distribution type and parameters, calculate the key time characteristic quantities: mean μ ij (expected transition time); variance σ ij 2 (time volatility); median m ij (typical transition time); quantile q ij ,α (reliability interval); form the time characteristic matrix T feat , comprehensively describing the time characteristics of each state transition process. Integrate the time distribution functions and their parameters of all state transition processes to construct a complete state duration distribution matrix D. Each element D of this matrix ij contains the time distribution function T ij (t) and its parameters.

[0237] Step S34, Construction of the external factor data matrix

[0238] Analyze the main external factors affecting the failure process of high dams, collect historical data on factors such as water level changes, temperature fluctuations, and load history, and construct the external factor data matrix F. This matrix contains the time series data of each external factor, providing a basis for modeling the conditional state transition probability.

[0239] Step S35, Establishment of the conditional state transition probability function

[0240] Read the external factor data matrix F, the state transition probability matrix P, and the state duration distribution matrix D, determine the key state transition processes that need to establish conditional relationships, and form the key transition process set K trans 。Conduct an importance analysis of the external factors, evaluate the influence degree of each factor on the state transition, select water level change (F1), temperature change (F2), and load history (F3) as key factors, and construct the key external factor set F key 。

[0241] Design a data stratification strategy, divide the factor space into multiple intervals to form a multi-dimensional grid, and construct a factor stratification framework G. Statistically analyze the state transition frequencies in each factor combination region, calculate the conditional transition probabilities, and construct a discrete conditional probability table CP discrete For regions with insufficient sampling points, design a conditional probability interpolation function, use a radial basis function network to achieve continuous estimation, and construct a continuous conditional probability model CP continuous Analyze the interaction effects among external factors, design an interaction term modeling strategy to capture the non-linear effects of the combined action of multiple factors, and form an interaction effect model I model 。

[0242] Integrate the aforementioned models to construct a complete conditional state transition probability function P ij (F1,F2,F3) = P ij ·(1+β1·(F1-F1_ref) / F1_max + β2·(F2-F2_ref) / F2_max + β3·(F3-F3_ref) / F3_max + β12·F1·F2 / F1_max / F2_max + β13·F1·F3 / F1_max / F3_max + β23·F2·F3 / F2_max / F3_max); where: P ij (F1,F2,F3) is the conditional transition probability from state i to state j under the action of external factors F1,F2,F3; P ij is the reference transition probability; β1,β2,β3 are the first-order influence coefficients; β12,β13,β23 are the interaction influence coefficients; F1_ref,F2_ref,F3_ref are the reference values; F1_max,F2_max,F3_max are the normalization coefficients

[0243] For the S1→S2 transfer process of a certain arch dam (the development of microcracks at the dam heel to crack propagation), the relevant parameters are: the reference probability P12 = 0.32; the water level influence coefficient β1 = 0.65; the temperature influence coefficient β2 = 0.43; the load influence coefficient β3 = 0.28; the water temperature interaction coefficient β12 = 0.15; when the water level ratio (F1-F1_ref) / F1_max = 0.8 and the temperature deviation (F2-F2_ref) / F2_max = 0.5, calculate the conditional transition probability: P12(0.8,0.5,0) = 0.32×(1+0.65×0.8+0.43×0.5+0.15×0.8×0.5) = 0.32×1.698 = 0.543

[0244] Similarly, establish the mapping relationship between the time distribution parameters and external factors, and the formula is: the time parameter mapping function μ ij(F1,F2,F3) = μ ij ·(1 - γ1·(F1 - F1_ref) / F1_max - γ2·(F2 - F2_ref) / F2_max - γ3·(F3 - F3_ref) / F3_max - γ12·F1·F2 / F1_max / F2_max); where: μ ij (F1,F2,F3) is the expected transition time from state i to state j under the action of external factors F1,F2,F3; μ ij is the reference expected time; γ1, γ2, γ3 are first-order influence coefficients; γ12 is the interaction influence coefficient; other parameters are the same as above.

[0245] For the S1→S2 transition process, the parameters are: reference expected time μ12 = 1.8 years; water level influence coefficient γ1 = 0.45; temperature influence coefficient γ2 = 0.32; water temperature interaction coefficient γ12 = 0.12; when the water level ratio is 0.8 and the temperature deviation is 0.5, calculate the conditional expected time: μ12(0.8,0.5) = 1.8×(1 - 0.45×0.8 - 0.32×0.5 - 0.12×0.8×0.5) = 1.8×0.588 = 1.06 years.

[0246] Verify the model performance to form the final non-stationary Markov model parameter set Θ, which completely describes the variation law of state transition characteristics with external factors.

[0247] Step S36, Construction of multi-layer failure mode network

[0248] Extract the main failure modes from the high dam failure case dataset and the failure mode classification system Y: crack mode (CM), deformation mode (DM), and seepage mode (FM), and construct the failure mode type set MODE.

[0249] For each failure mode, extract the relevant state nodes and transition relationships from the initial failure path network Z, construct a mode sub-network, and form the mode sub-network set Z sub .

[0250] Optimize the structure of each sub-network, supplement possible state nodes and transition relationships, and form the mode enhanced sub-network Z'm.

[0251] Define network parameters for each sub-network, including node attributes and edge attributes, and construct the mode network parameter set Pm.

[0252] Take all enhanced sub-networks as independent layers to construct the initial multi-layer failure mode network M0, which realizes the separated representation of different types of failure modes.

[0253] Analyze the physical interaction mechanism between failure modes, identify the key influence paths between modes, and construct the mode coupling path set L. For example, the influence of crack development on seepage (CM→FM), and the feedback of abnormal seepage on crack development (FM→CM).

[0254] For each coupling path, evaluate the influence strength between modes and construct the failure mode coupling strength matrix C ij = η·φ ij ·(1 + δ ij ·S ij );where: C ij is the coupling strength of failure mode i on mode j; η is the global coupling coefficient; φ ij is the physical mechanism influence coefficient; δ ij is the historical data adjustment coefficient; S ij is the expert scoring weight.

[0255] For the influence of the crack mode on the seepage mode, the parameters are: physical mechanism influence coefficient φCF = 0.75; historical data adjustment coefficient δCF = 0.6; expert scoring weight SCF = 0.8; global coupling coefficient η = 0.5; calculate the coupling strength CCF = 0.5×0.75×(1 + 0.6×0.8) = 0.45. For the influence of the seepage mode on the crack mode, the calculated coupling strength CFC = 0.38.

[0256] Design the coupling dynamic equation to describe the modified transition probability considering mode coupling. Among them, the coupling dynamic equation P' ij = P ij + ∑(Ckm·f(Sk,Sm)); where: P' ij is the modified transition probability considering mode coupling; P ij is the original transition probability; Ckm is the coupling strength of failure mode k on mode m; f(Sk,Sm) is the state interaction function; ∑ represents the summation over all relevant mode pairs (k,m). For the S1→S2 transition (the development of microcracks at the dam heel to crack propagation) in a certain arch dam crack network, the original transition probability P12 = 0.32. Considering the coupling influence of the seepage mode on the crack mode CFC·f(SF,SC) = 0.38×0.25 = 0.095, calculate the modified transition probability P'12 = 0.32 + 0.095 = 0.415.

[0257] Integrate the coupling relationship into the multi-layer network to form a complete multi-layer failure mode network M, which describes the internal evolution law of each failure mode and the interaction relationship between modes.

[0258] Step S37, Adaptive path generation

[0259] Read the multi - layer failure mode network M, the failure mode coupling strength matrix C, and the initial failure path network Z, analyze the network characteristics, and construct the network feature statistical set E. Design a path generation rule library, which contains 30 specific rules, to form the path generation rule set R. For example, the rule "If there is a crack at the heel of the dam and the water level change rate exceeds 0.5, it may develop into a crown - arch crack". Based on the rule set, perform forward reasoning, starting from the initial state, iteratively generate possible state - transition sequences, and construct the rule - driven path set P_rule, which contains 45 newly generated possible paths. Design a cross - layer path generation algorithm to jump between mode layers, generate composite paths that include interactions of multiple failure modes, and form the cross - mode path set P_cross, which contains 28 cross - mode paths. For example, the path "heel - of - dam crack → seepage anomaly → crown - arch crack" describes a complex failure process where a crack triggers seepage and then triggers a new crack. Considering the characteristics of a certain extra - high arch dam, generate dedicated failure paths, taking into account special factors such as temperature control and the combined action of arches and beams, to form the special - high - dam dedicated path set P_high, which contains 12 dedicated paths. Verify the physical rationality of the newly generated path sets, using a method that combines finite - element models and expert reviews to evaluate the physical feasibility of each path, and retain the reasonable paths to form the verified path set P_valid, which totals 63 effective paths.

[0260] Design a path scoring function, comprehensively considering physical rationality, historical similarity, and the degree of risk impact, calculate the importance scores for each path in the verified path set P_valid, and construct the path importance ranking table S_path. Select the paths with scores exceeding the threshold τ = 0.65 as the core extended paths, and fuse them with the initial failure path network Z to form the refined extended path network Z_opt, which contains a total of 75 core paths, including the original and newly added ones. Perform parameter estimation on the refined extended path network Z_opt, assign transition probabilities and time - distribution parameters to the newly added paths, complete the network quantification, and form the complete extended failure path network Z'.

[0261] Step S38: Construction of a multi - scale time - series evolution model

[0262] Integrate various types of data to construct the comprehensive model input data set I as the unified input for the multi - scale model. Design a macro - scale state - sequence simulation module, using the Monte Carlo method to simulate the state - transition sequence and time process of high - dam failure, and generate the macro - evolution model M_macro. Develop a meso - scale network spread model, based on the multi - layer failure mode network M and seepage theory, to simulate the propagation and spread process of failure in the network, and construct the meso - scale spread model M_meso. Construct a micro - scale physical evolution model, combining the principles of material mechanics, fracture mechanics, and fluid mechanics to simulate the physical - field evolution process in the local area of the high dam, and form the micro - physical model M_micro.

[0263] Design a multi-scale model coupling framework, construct an information transfer and feedback mechanism among the macro, meso, and micro scale models, realize bidirectional coupling between the upper and lower scales, and form a multi-scale coupling architecture C. Based on the extended failure path network Z', design a dynamic adjustment algorithm for path priority, evaluate the activation probability of each failure path in real time according to the system state and environmental conditions during the simulation process, and construct a path priority model P_prior. Integrate each sub-model and framework to construct a complete multi-scale time series evolution framework, and form a comprehensive time series evolution model M_evol. This model can simulate the time evolution, spatial distribution, and physical mechanism of the high dam failure process according to the given initial state and external conditions. Design a model solving engine, adopt a hybrid numerical algorithm, realize the efficient calculation of the model, and generate a simulation result dataset simulation result set R_sim. Develop a result analysis and visualization module, extract key time series features, identify critical states and warning indicators, and generate a time series evolution analysis report A_evol.

[0264] Implementation step S4, engineering analogy analysis and safety assessment

[0265] Step S41: Select the 5 projects with the highest similarity from the weighted similar project ranking list as the core analogy project set, and the Kolnbrein arch dam (similarity 0.843) as the most core analogy object. Analyze the historical accident situations of the Kolnbrein arch dam: When the dam was first impounded in 1977, cracks occurred at the dam heel, mainly located at the contact surface of the arch dam foundation, and the cause was the concentration of temperature stress and water pressure stress. Based on the analogy analysis, predict that the potential accident mode that may occur in a certain arch dam is "crack risk caused by stress concentration in the dam heel area".

[0266] Step S42: Collect the real-time monitoring data of a certain arch dam, including deformation monitoring, seepage monitoring, stress monitoring, etc., and construct a high dam monitoring dataset M data . Evaluate the current state of the high dam, identify potential abnormal features, locate specific state nodes, and determine the initial state information S i nit. Currently, a certain arch dam is in the normal state S0, but the stress level in the dam heel area has reached 85% of the design value, and there is a potential risk of developing into S1 (micro-cracks in the dam heel). Predict the future change trends of environmental conditions, including water level changes, temperature fluctuations, etc., and construct a future environmental condition prediction set E f uture, including various scenarios such as normal working conditions and extreme working conditions.

[0267] Input the initial state information S i nit and the future environmental condition prediction set E f uture into the time series failure path prediction model, perform simulation calculations, generate 1000 possible failure evolution trajectories, and construct a failure trajectory sample set T_sample.

[0268] Perform clustering analysis on the trajectory samples to identify three types of typical failure modes: dam heel crack-dominated (probability 65%); dam body deformation-dominated (probability 25%); seepage anomaly-dominated (probability 10%).

[0269] Extract the key time features of each mode in the typical failure mode set T_typical and construct a time feature analysis table T feat . For the most likely dam heel crack-dominated failure mode, the key time features are: from the initial state to the dam heel micro-crack (S0→S1): expected time 4.2 years, 90% confidence interval [2.8 years, 6.1 years]; from the micro-crack to the obvious crack (S1→S2): expected time 1.6 years, 90% confidence interval [0.9 years, 2.5 years]; from the obvious crack to the seepage anomaly (S2→S3): expected time 2.3 years, 90% confidence interval [1.5 years, 3.4 years];

[0270] Evaluate the risk levels of each failure mode, construct a risk assessment matrix R, combine the failure probability and the consequence severity, and give the risk level assessment results. Design a time series risk visualization scheme, integrate the failure path, time window, and risk level information, and generate a preliminary failure risk time series diagram G_risk. Optimize and enhance the time series diagram, add interactive functions, multi-level detailed information, and warning threshold marks to form the final failure risk time series diagram. This diagram intuitively shows the possible failure evolution path, the time window of each stage, and the risk level distribution of a certain arch dam, and marks three key warning points: the warning point for the dam heel stress exceeding the limit (time T1); the warning point for crack initialization (time T2); the warning point for seepage anomaly (time T3);

[0271] Step S43: For the typical projects in the core analog project set, combine the characteristics of a certain arch dam, construct a finite element numerical model, conduct a failure mechanism analysis, verify the rationality of the potential risk occurrence modes and the failure risk time series diagram, and form a verification analysis result. The verification analysis shows that under extreme water level fluctuation conditions, stress concentration may indeed occur in the dam heel area of a certain arch dam, which is basically consistent with the risk occurrence mechanism of the analog project Kolnbrein arch dam, confirming the reliability of the engineering analogy result.

[0272] Step S44: Read the potential risk patterns, the failure risk time series diagram, and the verification analysis results, design a multi-source information fusion framework, and construct a comprehensive evaluation input set I_comp. Use the analytic hierarchy process to construct a safety evaluation index system, set the index weights, and form an evaluation index system H. The index system includes: Structural safety (weight 0.4): stress state, deformation control, seismic resistance, etc.; Operational reliability (weight 0.35): monitoring system, operation management, response speed, etc.; Risk controllability (weight 0.25): warning mechanism, emergency measures, backup plans, etc.; Quantify the score for each index, and use the fuzzy comprehensive evaluation method to handle the uncertainty of the scores, and construct an index score matrix S.

[0273] Calculate the comprehensive score and the scores of each subsystem: Structural safety score: 85 points (safety level A); Operational reliability score: 78 points (safety level B); Risk controllability score: 82 points (safety level A); Comprehensive safety level: A (good); Determine that the safety level assessment result L is level A (good), indicating that the current overall safety state of a certain arch dam is good, but potential problems in terms of operational reliability still need to be concerned.

[0274] Based on the failure risk time series diagram and the verification analysis results, identify the key risk points, and construct a key risk point analysis table K. The main risk points include: Stress concentration points in the heel area of the dam (risk coefficient 0.78); Dam body temperature control area (risk coefficient 0.65); Grouting curtain leakage area (risk coefficient 0.52);

[0275] For the key risk points, design a monitoring enhancement plan, including: Add 5 stress monitoring points in the heel area of the dam; Adjust the existing strain monitoring frequency from once a quarter to once a month; Install an acoustic detection system in the potential crack area; Form a targeted monitoring plan M to strengthen the monitoring coverage of high-risk areas.

[0276] Design a hierarchical and phased intervention strategy, and construct an intervention strategy matrix I: Short-term strategy: Regular ultrasonic detection, once a quarter; Medium-term strategy: Optimize the water level control plan to avoid drastic water level changes; Long-term strategy: Refer to the reinforcement experience of the Kolnbrein arch dam and reserve the design plan for the post-dam support structure;

[0277] Design a risk dynamic monitoring framework, establish a data-driven risk assessment update mechanism, determine a regular assessment cycle of once every six months, and form a dynamic risk monitoring plan D. Integrate all evaluation results and suggestions, and compile a structured high dam safety status evaluation report to provide a scientific basis for the safe operation and management of a certain arch dam.

[0278] The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the specific details in the above embodiments. Within the scope of the technical concept of the present invention, various equivalent transformations can be made to the technical solutions of the present invention, and these equivalent transformations all fall within the protection scope of the present invention.

Claims

1. A method for diagnosing the service behavior of a structure based on the failure mode and failure path of a high dam, characterized in that, It includes the following steps: Collect information on high dam failure cases, construct a high dam failure case database, and extract characteristic parameters, failure modes, and failure paths to form a high dam failure case dataset; Based on the high dam failure case dataset, apply multi-scale hierarchical adaptive correspondence analysis to generate a weighted similar project ranking list; Based on the high dam failure case dataset, construct a time-series adaptive network model with coupled multi-failure modes, and output a time-series failure path prediction model; Combine the weighted similar project ranking list and the time-series failure path prediction model to conduct engineering analogy analysis on the high dam to be evaluated, and generate a service safety diagnosis report; Among them, the steps of generating the weighted similar project ranking list include: Extract the characteristic parameter matrix from the high dam failure case dataset, calculate the characteristic correlation matrix, and calculate the characteristic weight vector accordingly; According to the classification system of failure modes, divide the high dam failure case dataset into at least two mode subsets, calculate their characteristic weights one by one, and construct a mode characteristic weight matrix; Construct a physical similarity index system, establish the mapping relationship between characteristic parameters and physical similarity indexes, calculate the physical similarity index matrix, and obtain the corrected physical similarity index matrix through structural difference compensation; Combine the outputs of the above three processes, calculate the similarity between high dams, and generate a weighted similar project ranking list.

2. The method according to claim 1, wherein The steps of calculating the characteristic weight vector include: Calculate the average correlation coefficient between each characteristic and the failure mode to form an average correlation coefficient vector; Conduct a significance test on the average correlation coefficient vector, screen out significantly correlated characteristics to form a significant characteristic index set, and obtain an expert weight vector through expert scoring; Perform weighted fusion on the average correlation coefficient vector and the expert weight vector to form a fused weight value vector; Conduct stability adjustment and normalization processing on the fused weight value vector to obtain the characteristic weight vector.

3. The method according to claim 1, characterized in that, The steps of constructing the mode characteristic weight matrix include: According to the failure mode classification system, divide the high dam failure case dataset into multiple mode subsets, and extract the characteristic parameter sub-matrix for each mode subset; Calculate the correlation coefficient between the characteristics and the mode in each characteristic parameter sub-matrix, construct a characteristic correlation sub-matrix, and calculate the mode characteristic weight vector; Analyze the sample distribution characteristics of each mode subset, calculate the sample size, coverage rate, and diversity index, and construct a mode sample characteristic matrix; Combine the characteristic weight vectors of all modes to form a mode characteristic weight matrix.

4. The method according to claim 1, characterized in that, The calculation process of the corrected physical similarity index matrix includes: Combine the physical mechanism of high dam failure, determine the key physical quantities, construct a similarity measurement function, and establish the mapping relationship between characteristic parameters and physical quantities; According to the mapping relationship, convert and calculate the physical similarity index values from the characteristic parameter matrix to construct the physical similarity index matrix; Identify the structural difference characteristics between modern super-high dams and historical cases, construct a structural difference characteristic set, evaluate the impact of each difference characteristic on the physical similarity index, and form a structural difference compensation function; According to the structural difference compensation function, compensate and correct the physical similarity index matrix to obtain the corrected physical similarity index matrix.

5. The method according to claim 1, wherein The steps of generating the weighted similar project ranking list include: Calculate the distances of the high dam in the original feature space and the physical similarity space respectively, and construct a dual-space distance vector; Convert the distances into similarities to obtain the similarity in the original feature space and the similarity in the physical similarity space; and evaluate the data quality and credibility of each space one by one, and combine the risk probability of the failure mode to obtain a comprehensively adjusted fusion weight factor; Based on the fusion weight factor, perform weighted fusion on the similarity in the original feature space and the similarity in the physical similarity space to calculate the multi-scale fusion similarity; Rank the historical cases according to the multi-scale fusion similarity to generate a weighted similar project ranking list.

6. The method according to claim 1, characterized in that, The steps to form a time-series failure path prediction model include: Extract the failure path information from the high dam failure case dataset to construct a state space and a state transition probability matrix; Fit a time distribution function to the state transition process to construct a state duration distribution matrix; Analyze the influence of external factors on the state transition to construct a conditional state transition probability function; Construct different failure modes into a multi-layer network, extract the coupling relationship between failure modes, and generate an extended failure path network; Combine the outputs of the above steps to construct a multi-scale time-series evolution model, and output a time-series failure path prediction model through simulation calculation.

7. The method according to claim 6, characterized in that, The steps to construct a state duration distribution matrix include: Select the cases containing complete time records from the high dam failure case dataset and extract the state transition time data; Clean the time data, analyze its distribution characteristics, determine the theoretical distribution type, and fit the distribution parameters; Based on the physical failure mechanism, correct the rationality of the distribution parameters, and calculate the expected transition time, variance, and reliability interval; Integrate the time distribution functions and their parameters of all state transition processes to construct a state duration distribution matrix.

8. The method according to claim 6, wherein The steps to construct a conditional state transition probability function include: Determine the key state transition processes that need to establish conditional relationships, evaluate the influence degree of each external factor, and select the key external factors; Design a factor space stratification framework according to the key external factors, count the state transition frequencies in each factor combination area, and construct a discrete conditional probability table; Design an interpolation function for the area with insufficient sampling points, and analyze the interaction between factors to form a continuous conditional probability model; Integrate the discrete conditional probability table and the continuous conditional probability model to construct a conditional state transition probability function.

9. The method according to claim 6, characterized in that, The steps to construct the coupling relationship between failure modes include: Extract the main failure mode types from the high dam failure case dataset to construct a failure mode type set; For each failure mode, extract the relevant state nodes and transition relationships, and construct and optimize the mode sub-network; Define node attributes and edge attributes for each sub-network, and construct a multi-layer network with all sub-networks as independent layers; Analyze the physical interaction mechanism between failure modes, identify the influence paths between modes, construct a mode coupling path set and evaluate the influence strength between modes, construct a failure mode coupling strength matrix, obtain the coupling relationship, and form a multi-layer failure mode network.

Images