A simulation analysis method for stability of dam foundation surface under osmotic pressure

Abstract

The present application relates to dam foundation data analysis technical field, specifically to a kind of dam foundation surface stability simulation analysis method under the action of seepage pressure, first to the upstream and downstream temperature signal in each period of time Continuous wavelet transform, temperature sequence is decomposed into discrete event with time sequence, energy and waveform characteristics, composite cost function including waveform distortion index and energy-time attenuation is constructed, the preferred input-output event pair corresponding to the most credible advantage transport path in physics is dynamically identified by combining optimization algorithm Then, combined with the transport duration of the through melting channel that has greater influence on stability is short, energy retention rate is high, and signal waveform fidelity is high characteristics, the preferred input-output event is analyzed, so that the stability determination coefficient obtained by the dam foundation surface stability characterized is more accurate, improve the accuracy of dam foundation surface stability analysis according to stability determination coefficient.

Description

Technical Field

[0001] This invention relates to the field of dam foundation data analysis technology, specifically to a simulation analysis method for the stability of dam foundation surfaces under seepage pressure. Background Technology

[0002] The stability of the dam foundation is central to the safety of hydraulic engineering projects. For dam foundations built in permafrost regions, stability analysis presents unique challenges because the frozen state of the soil and rock mass has a decisive impact on its permeability. Permafrost, due to the freezing of pore water, has a permeability coefficient far lower than that of soil and rock in their thawed state. Therefore, the 0°C isothermal surface within the dam foundation, determined by the temperature field, constitutes the de facto internal dynamic boundary controlling the seepage path and seepage pressure distribution. The accuracy of the analysis in addressing such problems heavily depends on a precise understanding of the morphology of this internal thaw zone. However, this freeze-thaw boundary is located within the dam foundation structure, and its spatiotemporal evolution cannot be directly and continuously obtained through conventional surveying methods.

[0003] Existing methods for stability simulation analysis based on static or empirical assumptions about the internal melting zone morphology of dam foundations cannot capture and distinguish dynamic seepage path changes caused by different physical mechanisms (such as diffuse melting and through-cracks) in real time and accurately. This leads to a serious disconnect between the simulation model and the actual engineering situation, which may seriously underestimate the risk of dam foundation instability caused by through-cracks with high risk. In other words, the accuracy of existing methods for dam foundation surface stability analysis based on static or empirical assumptions about the internal melting zone morphology of dam foundations is low. Summary of the Invention

[0004] To address the low accuracy of existing technologies that rely on static or empirical assumptions about the internal fusion zone morphology of dam foundations for stability analysis, this application aims to provide a simulation analysis method for dam foundation stability under seepage pressure. The specific technical solution adopted is as follows: The first aspect of this application provides a simulation analysis method for the stability of dam foundation surfaces under seepage pressure, including: The upstream boundary temperature sequence and downstream seepage water temperature sequence of the dam foundation are obtained in each periodic time. Based on the upstream boundary temperature sequence, the corresponding input event set is determined by wavelet transform. Based on the principle of obtaining the input event set, the corresponding output event set is determined based on the downstream seepage water temperature sequence. Based on the causal relationship between the input events in the input event set and the output events in the output event set in terms of timing and energy, as well as the thermal response time constant for each period, all preferred input-output event pairs are selected; wherein, the thermal response time constant is determined based on the temporal and energy distribution of the preferred input-output event pairs in the previous period. Based on the distribution of the relative deviations in energy and timing between the output events and input events among all preferred input and output events in the current period, the stability determination coefficient is determined; the stability analysis of the dam foundation surface is then performed based on the stability determination coefficient.

[0005] Furthermore, the process of obtaining the input event set includes: The upstream boundary temperature sequence is subjected to wavelet transform to determine the corresponding complex wavelet coefficient spectrum; all local maxima points on the modulus of the complex wavelet coefficient spectrum are statistically analyzed; the corresponding fluctuation peak time is determined according to the time corresponding to each local maxima point on the time axis; the corresponding fluctuation peak amplitude is determined according to the wavelet coefficient value of each local maxima point at the scale with the most concentrated energy; the temperature waveform feature vector of each local maxima point is determined by arranging the energy values ​​of the fluctuation peak time at all scales in ascending order of scale and normalizing them. The set of fluctuation peak time, fluctuation peak amplitude, and temperature waveform feature vector corresponding to each local maximum point of the upstream boundary temperature sequence is taken as an input event; the set of input events is determined based on the input events of all local maximum points corresponding to the upstream boundary temperature sequence.

[0006] Furthermore, the process of acquiring the preferred input-output event pair includes: The output event set includes the peak fluctuation time, peak fluctuation amplitude, and temperature waveform feature vector of each output event corresponding to the downstream seepage water temperature sequence; element matching traversal is performed on the input event set and the output event set to determine all input-output event pairs; The corresponding cost element value is determined based on the deviations between the input and output events of each input-output event pair in terms of temperature waveform feature vector, peak fluctuation time, and peak fluctuation amplitude. Construct a cost matrix where rows represent input events and columns represent output events. Each element of the cost matrix is ​​the cost element value of an input-output event pair consisting of the input event in the corresponding row and the output event in the corresponding column. Input the cost matrix into the Hungarian algorithm and output the set of optimal matching pairs with the minimum total cost. The set of optimal matching pairs includes all preferred input-output event pairs.

[0007] Furthermore, the process of acquiring the input / output event pair includes: Within each periodic time interval, an input event and an output event are randomly selected to form an input-output event pair; the set of input events and the set of output events are traversed to determine all input-output event pairs for each periodic time interval.

[0008] Furthermore, the process of obtaining the cost element value includes: Filter out the ideal input-output event pairs from all input-output event pairs; in each ideal input-output event pair, determine the corresponding waveform evolution cost based on the Earth's movement distance between the temperature waveform feature vector of the input event and the temperature waveform feature vector of the output event; The energy decay performance value is determined by positively mapping the ratio between the peak amplitude of the output event and the peak amplitude of the input event; the time interval between the peak moments of the output event and the peak moments of the input event is calculated; the time interval is weighted by the negative correlation mapping value of the thermal response time constant of each ideal input-output event to its periodic time interval to determine the time decay coefficient; and the comprehensive decay cost is determined based on the absolute value of the sum of the energy decay performance value and the time decay coefficient. The cost element value for each ideal input-output event pair is determined by weighted summation of the waveform evolution cost and the comprehensive attenuation cost. Set the cost element values ​​of all other input / output event pairs of the ideal input / output event to the preset cost constant.

[0009] Furthermore, the process of acquiring the ideal input-output event pair includes: An ideal input-output event pair is one where the peak value of the input event precedes the peak value of the output event, and the amplitude of the peak value of the input event is greater than that of the peak value of the output event.

[0010] Furthermore, the process of obtaining the thermal response time constant includes: In each preferred input-output event pair within each cycle time period, the corresponding transport duration is determined based on the time interval between the peak fluctuation time of the output event and the peak fluctuation time of the input event; the corresponding energy retention rate is determined based on the ratio between the peak fluctuation amplitude of the output event and the peak fluctuation amplitude of the input event. The optimal characteristic time constant is determined based on the concentrated distribution of waveform evolution cost, transport duration, and energy retention rate of all preferred input and output events; Each period outside the first period is taken as the target period, and the period preceding the target period is taken as the reference period. The thermal response time constant of the target period is determined by weighted summation of the optimal characteristic time constant of the reference period and the thermal response time constant of the reference period. The thermal response time constant of the reference period of the first period is a priori thermal response constant value.

[0011] Furthermore, the process of obtaining the optimal characteristic time constant includes: Density clustering analysis is performed based on the waveform evolution cost, transport duration, and energy retention rate of all preferred input-output event pairs to determine at least two event pair clusters. In the event pair cluster with the most preferred input-output event pairs, the optimal characteristic duration is determined based on the mean transport duration of all preferred input-output event pairs. The optimal retention rate is determined based on the mean energy retention rate of all preferred input-output event pairs. The optimal characteristic time constant is determined by performing a negative correlation mapping between the negative correlation mapping value of the optimal retention rate and the product of the optimal characteristic duration.

[0012] Furthermore, the process of obtaining the stability determination coefficient includes: In the current period, the preferred input-output event pair that satisfies the following conditions is that the transport duration is less than the preset duration threshold, the energy retention rate is greater than the preset retention rate threshold, and the waveform evolution cost is less than the preset evolution threshold. Density clustering analysis is performed on the waveform evolution cost, transport duration, and energy retention rate of all reference input-output event pairs to determine all density clusters; the number of reference input-output event pairs in the density cluster with the most reference input-output event pairs is used as the stability determination coefficient.

[0013] Furthermore, the process of performing stability analysis of the dam foundation surface based on the stability determination coefficient includes: When the stability determination coefficient is greater than the preset stability threshold, the dam foundation is determined to be unstable; when the stability determination coefficient is less than or equal to the preset stability threshold, the dam foundation is determined to be stable.

[0014] Secondly, this application provides a simulation analysis system for the stability of a dam foundation surface under seepage pressure, the system comprising: The data acquisition and preprocessing module is used to acquire the upstream boundary temperature sequence and the downstream seepage water temperature sequence of the dam foundation in each cycle time period; based on the upstream boundary temperature sequence, the corresponding input event set is determined by wavelet transform method; and based on the principle of acquiring the input event set, the corresponding output event set is determined based on the downstream seepage water temperature sequence. The event pair filtering module is used to filter out all preferred input-output event pairs based on the causal relationship between input events in the input event set and output events in the output event set in terms of timing and energy, as well as the thermal response time constant for each period; wherein, the thermal response time constant is determined based on the time and energy distribution of the preferred input-output event pairs in the previous period. The stability analysis module is used to determine the stability determination coefficient based on the distribution of the relative deviations in energy and timing between the output events and input events among all preferred input and output events in the current period; and to perform stability analysis of the dam foundation surface based on the stability determination coefficient.

[0015] Thirdly, this application provides a computer device including a memory and a processor. The memory is used to store computer program code, and the processor is used to call and run the computer program code from the memory to perform the method as described in the first aspect of this application or any embodiment of the first aspect.

[0016] Fourthly, this application provides a computer program product comprising computer program code, which, when executed, performs the method as described in the first aspect of this application or any embodiment thereof.

[0017] Fifthly, this application provides a computer-readable storage medium that stores computer program code, which, when executed, performs the method as described in the first aspect of this application or any embodiment thereof.

[0018] This application has the following beneficial effects: This application performs continuous wavelet transform on the upstream and downstream temperature signals in each period, decomposing the temperature sequence into discrete events with temporal, energy, and waveform characteristics. A composite cost function including waveform distortion index and energy-time decay is constructed. Combined with optimization algorithms, the preferred input-output event pairs corresponding to the most physically reliable advantageous transport path are dynamically identified. Then, considering the characteristics of the through-flow melting channel, which has a significant impact on stability, such as short transport duration, high energy retention rate, and high signal waveform fidelity, feature analysis is performed on the preferred input-output events. This makes the stability determination coefficient obtained more accurate in representing the stability of the dam foundation surface, thus improving the accuracy of dam foundation surface stability analysis based on the stability determination coefficient. Attached Figure Description

[0019] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart of a simulation analysis method for the stability of a dam foundation surface under seepage pressure, provided in one embodiment of the present invention. Figure 2This is a structural diagram of a dam foundation surface stability simulation analysis system under seepage pressure provided in one embodiment of the present invention; Figure 3 This is a schematic diagram of a computer device structure provided in one embodiment of the present invention. Detailed Implementation

[0021] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a dam foundation stability simulation analysis method under seepage pressure proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment, and specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0023] The following description, in conjunction with the accompanying drawings, details the specific scheme of the dam foundation surface stability simulation analysis method under seepage pressure provided by this invention.

[0024] This application provides a simulation analysis method for the stability of dam foundation surfaces under seepage pressure. Please refer to [link / reference]. Figure 1 The diagram illustrates a flowchart of a simulation analysis method for dam foundation surface stability under seepage pressure, provided by an embodiment of the present invention. The method includes: Step S101: Obtain the upstream boundary temperature sequence and downstream seepage water temperature sequence of the dam foundation in each period; determine the corresponding input event set based on the upstream boundary temperature sequence using wavelet transform; determine the corresponding output event set based on the downstream seepage water temperature sequence according to the principle of obtaining the input event set.

[0025] In one specific implementation of this invention, the following time series are extracted from the dam foundation index monitoring database: upstream water level time series, downstream seepage flow time series, dam area air temperature time series, downstream seepage water temperature time series, and upstream boundary temperature time series during the sampling period. The sampling period is divided into at least two periodic time series of equal length. The corresponding downstream seepage water temperature series is determined based on the subsequence corresponding to each periodic time series in the downstream seepage water temperature time series. The corresponding upstream boundary temperature series is also determined based on the subsequence corresponding to each periodic time series in the upstream boundary temperature time series. The sampling period is set to one calendar year prior to the current moment, the sampling frequency is set to once every 15 minutes, and the periodic time series length is set to 7 days. The length of the sampling period, the periodic time series length, and the sampling frequency can be adjusted according to the specific implementation environment. The purpose of introducing the periodic time series is to enable periodic simulation analysis of dam foundation stability. That is, in this invention, a dam foundation stability simulation analysis process is performed after each periodic time series to respond promptly to the dam foundation stability status.

[0026] When analyzing seepage problems within dam foundations, focusing solely on signal transmission time delays and amplitude attenuation neglects the morphological distortion of the signal waveform during transport. This distortion contains crucial information for distinguishing transport mechanisms such as through-flow convection or matrix diffusion. As temperature fluctuations from the upstream boundary propagate through the dam foundation, their signal waveforms undergo significant changes due to the properties of the medium they pass through (e.g., open melting fissures or saturated porous soil). To capture and quantify this change, the continuous temperature time series needs to be decomposed into a series of discrete, analyzable independent events. Therefore, this embodiment of the invention uses wavelet transform to determine the corresponding set of input events based on the upstream boundary temperature sequence; and based on the causal relationship of the transport process, determines the corresponding set of output events based on the downstream seepage water temperature sequence according to the principle of obtaining the input event set.

[0027] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the input event set includes: A wavelet transform is performed on the upstream boundary temperature sequence to determine the corresponding complex wavelet coefficient spectrum. All local maxima points on the modulus of the complex wavelet coefficient spectrum are statistically analyzed. The corresponding peak fluctuation time is determined based on the time corresponding to each local maxima point on the time axis. The corresponding peak fluctuation amplitude is determined based on the wavelet coefficient value of each local maxima point at the scale with the highest energy concentration. The temperature waveform feature vector for each local maxima point is determined by arranging the energy values ​​at all scales in ascending order of scale and normalizing them. In a specific implementation of this invention, the method for normalizing the energy values ​​corresponding to each scale in the temperature waveform feature vector uses the softmax function, making the sum of the energy values ​​corresponding to all scales in the temperature waveform feature vector equal to 1, thus more clearly showing the relative distribution of energy at different scales.

[0028] The modulus represents the energy intensity of a signal at a specific time and scale. Analyzing the modulus of the wavelet coefficient spectrum can more sensitively extract the event features corresponding to temperature fluctuations. Local maxima usually correspond to abrupt changes in fluctuation characteristics, so analyzing the features at local maxima can capture the events corresponding to temperature fluctuations. Therefore, it is necessary to first count all local maxima. The peak time of the fluctuation is equivalent to marking the event in the time dimension, that is, representing the temporal characteristics of the event. The amplitude of the peak fluctuation is a direct measure of the energy contained in the event, which quantifies the intensity of the temperature fluctuation. For the temperature waveform feature vector, the small scale in the wavelet transform corresponds to the high-frequency component, and the large scale corresponds to the low-frequency component. Therefore, the temperature waveform feature vector essentially represents the multi-scale energy distribution structure of a single temperature fluctuation event. This is reflected in the fact that it uniquely defines the waveform features of the event by describing the relative proportion of energy in the fluctuation components at different scales, thus enabling the comparison of waveforms of different energy events. This feature is closely related to the physical properties of the transport path, such as low dispersion caused by penetrating fractures or high dispersion caused by saturated soil, thus becoming the core basis for analyzing the internal transport mechanism. Therefore, the set of fluctuation peak time, fluctuation peak amplitude and temperature waveform feature vector corresponding to each local maximum point of the upstream boundary temperature sequence is further taken as an input event; the set of input events is determined based on the input events of all local maximum points corresponding to the upstream boundary temperature sequence.

[0029] Based on the principle of obtaining the input event set, the corresponding output event set is determined according to the downstream seepage water temperature sequence. Specifically, the upstream boundary temperature sequence in the process of obtaining the input event set is replaced with the downstream seepage water temperature sequence. The set of fluctuation peak time, fluctuation peak amplitude, and temperature waveform feature vector corresponding to each local maximum point of the downstream seepage water temperature sequence is taken as an output event. The input event set is determined based on the input events of all local maximum points corresponding to the downstream seepage water temperature sequence. The principle of obtaining the output event set is exactly the same as the principle of obtaining the input event set, and will not be further limited or elaborated here. Among them, the upstream boundary temperature sequence belongs to the upstream data and is used as input for analysis, and the downstream seepage water temperature sequence belongs to the downstream data and is used as output for analysis, so that the input and output conform to the correspondence of transport.

[0030] Step S102: Based on the causal relationship between the input events in the input event set and the output events in the output event set in terms of timing and energy, and the thermal response time constant for each period, select all preferred input-output event pairs; wherein, the thermal response time constant is determined based on the time and energy distribution of the preferred input-output event pairs in the previous period.

[0031] The input and output event sets only reflect the independent temperature fluctuation characteristics of the upstream and downstream boundaries. In order to establish a causal relationship between the two that conforms to physical laws and to diagnose the physical mechanism of the internal seepage path, it is necessary to screen out the most physically credible event pairs that can characterize the real transport path from all possible combinations. Therefore, this embodiment of the invention further screens out all preferred input-output event pairs based on the causal relationship between the input events in the input event set and the output events in the output event set in terms of time sequence and energy, as well as the thermal response time constant of each period. This ensures that the obtained preferred input-output event pairs can reflect the real, physically meaningful, and most credible dominant thermal transport path inside the dam foundation, thereby improving the accuracy of subsequent risk mode diagnosis and stability assessment.

[0032] Preferably, in some possible implementations of the embodiments of the present invention, the process of acquiring input-output event pairs preferably includes: The output event set includes the peak time, peak amplitude, and temperature waveform feature vector for each output event corresponding to the downstream seepage water temperature sequence. Element-wise matching is performed on the input and output event sets to determine all input-output event pairs. Specifically, within each period, one input event and one output event are randomly selected to form an input-output event pair. The input and output event sets are traversed to determine all input-output event pairs for each period. Since an upstream event (input event) may have potential correlations with multiple downstream events (output events), to identify the optimal input-output event pair that best conforms to the heat transfer law, all possible matches between input and output events need to be pre-screened, allowing for further filtering and analysis among all input-output event pairs.

[0033] The cost element value is determined based on the deviations between the input and output events of each input-output event pair in terms of temperature waveform feature vector, peak fluctuation time, and peak fluctuation amplitude. For each input-output event pair, the more significant the deviation between the input and output events in the three feature dimensions of temperature waveform feature vector, peak fluctuation time, and peak fluctuation amplitude, the lower the probability that the corresponding event pair represents a real physical transport process. The higher the possibility that its waveform distortion, time delay, or energy decay is inconsistent with the overall thermodynamic law of the system, the larger the cost element value obtained, the lower the probability that the corresponding input and output events originate from the same heat transport path, and the less likely it is to belong to the preferred input-output event pair.

[0034] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the cost element value includes: Ideal input-output event pairs are selected from all input-output event pairs. In a specific implementation of this invention, the process of obtaining ideal input-output event pairs includes: selecting input-output event pairs whose peak fluctuation time of the input event precedes the peak fluctuation time of the output event, and whose peak fluctuation amplitude of the input event is greater than that of the output event, as ideal input-output event pairs. First, if an input event and an output event correspond in objective reality, then the timing of the input event must precede that of the output event to create a causal effect in time; and the peak fluctuation amplitude representing energy at the input event will objectively experience energy attenuation after transmission, so the peak fluctuation amplitude representing energy at the obtained output event must be relatively small; therefore, based on these characteristics, ideal input-output event pairs are first selected, and the transmission cost of these ideal input-output event pairs is quantitatively calculated. Other input-output event pairs are based on a larger preset transmission cost, making the accuracy of the subsequently selected preferred input-output event pairs higher.

[0035] In each ideal input-output event pair, the waveform evolution cost is determined based on the Earth's movement distance between the temperature waveform feature vectors of the input and output events. According to the definition of Earth's movement distance, the smaller this distance, the lower the minimum cost required to transform the temperature waveform feature vector of the input event into that of the output event. Therefore, from the perspective of characterizing the physical mechanism of the transport path, a lower transport cost corresponds to a higher probability of belonging to the preferred input-output event pair, and the smaller the cost element value assigned to characterize the path matching quality should be. Furthermore, for subsequently selected preferred input-output events, a lower waveform evolution cost indicates that the signal waveform is better preserved during transmission. This directly points to a through-flow transport path that is dominated by convection and has extremely weak diffusion effects, posing a threat to dam foundation stability.

[0036] The energy decay performance value is determined by positively mapping the ratio between the peak amplitude of the output event and the peak amplitude of the input event; the time interval between the peak moments of the output event and the peak moments of the input event is calculated; the time interval is weighted by the negative correlation mapping value of the thermal response time constant of each ideal input-output event to its periodic time interval to determine the time decay coefficient; and the comprehensive decay cost is determined based on the absolute value of the sum of the energy decay cost and the time decay cost.

[0037] First, energy decay follows an exponential decay law. In the scenario of this invention embodiment, the formula corresponding to the ideal energy decay model should be: ;in, For the output energy under ideal conditions, For the ideal input energy, For the duration of transportation, The characteristic time constant under ideal conditions, This corresponds to the decay rate in the exponential decay law. The formula for the energy decay model under this ideal condition can be obtained by transforming it: However, this is an ideal energy decay model. In real-world environments, multiple factors inevitably exist, such as measurement errors, noise interference, medium inhomogeneity, and heat loss. This causes the observed output energy, input energy, transport duration, and characteristic time constant in real-world environments to not perfectly satisfy this ideal equation, resulting in a non-zero deviation. ,in, The output energy observed in the actual environment. The input energy observed in the actual environment. The transport duration observed in the actual environment. The characteristic time constant observed in the actual environment is denoted as . The smaller the corresponding deviation, the higher the degree of agreement between the observed energy decay and transport duration and the physical model prediction value of the characteristic time constant.

[0038] Therefore, for each ideal input-output event pair in this embodiment of the invention: the peak amplitude of the corresponding output event is the observed output energy, the peak amplitude of the input event is the observed input energy, the time interval between the peak moment of the output event and the peak moment of the input event is the observed transport duration, and the thermal response time constant is the observed characteristic time constant; the process of obtaining the comprehensive attenuation cost is expressed by the formula: ;in, Periodic time period Ideal Input / Output Event Pair The overall attenuation cost; Periodic time period Ideal Input / Output Event Pair The peak amplitude of the fluctuation in the output event; Periodic time period Ideal Input / Output Event Pair The peak amplitude of the fluctuation of the input event; Periodic time period Ideal Input / Output Event Pair The time interval between the peak value of the output event and the peak value of the input event; Periodic time period The lower the overall attenuation cost, the better the coupling relationship between energy attenuation and transport duration exhibited by the ideal input-output event pair in the actual environment matches the predicted value of the physical model based on the current characteristic time constant. This indicates that the transport path represented by the ideal input-output event pair is more consistent with the overall heat dissipation characteristics of the dam foundation system, and is more likely to be a physically reliable and real transport path that conforms to the mainstream response law of the system. The corresponding ideal input-output event pair is more likely to be the preferred input-output event pair, and the smaller the cost element value assigned to characterize the path matching quality should be.

[0039] Furthermore, based on the characteristic that both waveform evolution cost and comprehensive attenuation cost correspond to cost element values, a weighted sum is performed on the waveform evolution cost and comprehensive attenuation cost to determine the cost element value for each ideal input-output event pair. In a specific implementation of this invention, the corresponding weighted attenuation cost is determined by multiplying the normalized value of the comprehensive attenuation cost of each ideal input-output event pair with a preset first weight; the corresponding weighted evolution cost is determined by multiplying the waveform evolution cost of each ideal input-output event pair with a preset second weight; and the cost element value for each ideal input-output event pair is determined by the sum of the weighted attenuation cost and the weighted evolution cost; wherein the preset first weight and the preset second weight are both real numbers greater than 0, and the sum of the preset first weight and the preset second weight is 1.

[0040] In one specific implementation of this invention, the process of obtaining the cost element value is expressed by the following formula: ;in, Periodic time period Ideal Input / Output Event Pair Cost element value; Periodic time period Ideal Input / Output Event Pair The overall attenuation cost; Periodic time period Ideal Input / Output Event Pair Waveform evolution cost; It is a minimum-maximum normalization function, and the normalization range is the periodic time interval. All ideal input / output event pairs; The first weight is preset; The second weight is preset; Periodic time period Ideal Input / Output Event Pair The weighted attenuation cost; Periodic time period Ideal Input / Output Event Pair The weighted evolution cost; in a specific implementation of this invention, the first weight is preset to 0.5 and the second weight is preset to 0.5, which indicates that the attention to the comprehensive attenuation cost and the waveform evolution cost is consistent. It can be adjusted according to the specific implementation environment, and the parameter with higher attention is given a greater weight. This will not be elaborated further here.

[0041] The cost element values ​​of all other input / output event pairs exposed to the ideal input / output event are set to a preset cost constant. In a specific implementation of this invention, the preset cost constant is set to 10, which can be adjusted according to the specific implementation environment, but it is recommended to take a value greater than 1. Therefore, the value range of the cost element values ​​of all ideal input / output event pairs is between 0 and 1. If the preset cost constant is less than 1, it may cause other input / output events exposed to the ideal input / output event to affect the subsequent selection of the optimal input / output event, because other input / output events exposed to the ideal input / output event are not input / output events that can occur in the actual environment. Therefore, a larger cost element value is assigned to reduce the impact on the selection process.

[0042] Further construct a cost matrix, where rows represent input events and columns represent output events. Each element of the cost matrix is ​​the cost element value of an input-output event pair consisting of the input event in the corresponding row and the output event in the corresponding column. Input the cost matrix into the Hungarian algorithm and output the set of optimal matching pairs with the minimum total cost. The set of optimal matching pairs includes all preferred input-output event pairs.

[0043] By constructing a cost matrix, the physical rationality of all potential input-output event pairs can be systematically and structurally evaluated. This allows the set of optimal matching pairs with the minimum total cost output after combining with the Hungarian algorithm to represent the optimal transport path combination in terms of both physical mechanism (waveform distortion) and macroscopic dynamics (energy time decay) within the corresponding periodic time. This makes the paths represented by the selected input-output event pairs more realistic and reliable, and can keenly capture the high-risk transport patterns that are crucial for stability assessment.

[0044] In the process of calculating the comprehensive attenuation cost, a thermal response time constant, which characterizes the characteristic time constant, is introduced. This thermal response time constant essentially represents the dynamic state parameter of the overall heat dissipation characteristics of the dam foundation. However, the thermal-hydraulic characteristics of the dam foundation are not constant and need to be iteratively updated in conjunction with the transport characteristics of the latest identified preferred input-output time pairs representing the main healthy path cluster in each analysis cycle. This determines the thermal response time constant for the next cycle, providing a physical basis for continuous calibration of the dynamic and adaptive stability analysis of the dam foundation. This ensures that the simulation model can track the changes in the actual state of the dam foundation in real time and avoids evaluation deviations caused by model parameter lag. Therefore, in this embodiment of the invention, the thermal response time constant is determined based on the time and energy distribution of the preferred input-output event pairs in the previous cycle.

[0045] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the thermal response time constant includes: For each preferred input-output event pair within each cycle time period, the corresponding transport duration is determined based on the time interval between the peak fluctuation times of the output event and the input event; the corresponding energy retention rate is determined based on the ratio between the peak fluctuation amplitude of the output event and the peak fluctuation amplitude of the input event. It should be noted that if the peak fluctuation amplitude of either the output or input event of a preferred input-output event pair is 0 or less than 0, that preferred input-output event pair is discarded, as it has lost its physical meaning for path analysis and is therefore no longer considered a preferred input-output event for analysis.

[0046] The optimal characteristic time constant is determined based on the concentrated distribution of waveform evolution cost, transport duration, and energy retention rate of all preferred input-output events. In a specific implementation of this invention, the process of obtaining the optimal characteristic time constant includes: performing density clustering analysis based on the waveform evolution cost, transport duration, and energy retention rate of all preferred input-output event pairs to determine at least two event pair clusters; determining the optimal characteristic duration based on the mean transport duration of all preferred input-output event pairs in the event pair cluster with the most preferred input-output event pairs; determining the optimal retention rate based on the mean energy retention rate of all preferred input-output event pairs; and determining the optimal characteristic time constant by performing a negative correlation mapping between the negative correlation mapping value of the optimal retention rate and the product of the optimal characteristic duration. Among them, the clustering analysis method for density clustering analysis based on waveform evolution cost, transport duration and energy retention rate of all preferred input-output event pairs adopts DBSCAN clustering analysis, with the minimum number of points set to 7. The neighborhood radius is determined by analyzing the distribution of the distance between all points in the cluster sample and their k-th nearest neighbor. The distance value at the inflection point of the distance distribution curve is selected as the value of the neighborhood radius. It can be subjectively set according to the specific implementation environment, and will not be elaborated further here.

[0047] The cluster of event pairs with the most preferred input-output event pairs represents the transport pattern followed by the dominant and conventional healthy seepage path in the dam foundation system within each cycle time period. Therefore, the optimal characteristic duration and optimal retention rate determined by this cluster essentially belong to the macroscopic average transport characteristics of the mainstream healthy state of the system. Furthermore, based on the optimal characteristic duration and optimal retention rate, the optimal characteristic time constant corresponding to the system characteristic time scale representing this macroscopic average transport characteristic can be determined by back-calculating through the energy decay model. This optimal characteristic time constant is an objective, data-driven instantaneous estimate of the overall heat dissipation characteristics of the dam foundation under the current time window.

[0048] In one specific implementation of this invention, the process of obtaining the optimal characteristic time constant is expressed by the following formula: ;in, Periodic time period The optimal characteristic time constant; Periodic time period The mean energy retention rate of all preferred input-output event pairs in the cluster with the most preferred input-output event pairs, which is also the optimal retention rate; Periodic time period The optimal characteristic time constant is the average transport duration of all optimal input-output event pairs in the cluster with the most optimal input-output event pairs, which is also the optimal characteristic duration. Firstly, in the objective reality, due to energy loss, the energy retention rate of optimal input-output event pairs is always less than 1, and greater than 0; therefore, the formula is not meaningless. The formula for obtaining the optimal characteristic time constant is essentially the inverse derivation of the energy decay model, which will not be further limited or elaborated upon here.

[0049] Furthermore, considering that directly using the optimal characteristic time constant as the thermal response time constant for the next period would cause the state estimate to jump drastically between different periods, since the observation data of a single window may contain random noise or short-term disturbances and cannot smoothly reflect the continuous and gradual evolution trend of the thermal characteristics of the dam foundation, it is also necessary to combine the thermal response time constant of each period to comprehensively determine the thermal response time constant of the next period. This ensures that the thermal response time constant of the next period incorporates the information of the latest observation data and inherits the stability of the historical state, thereby achieving smooth, robust and adaptive tracking of the dynamic characteristics of the entire dam foundation system.

[0050] In this embodiment of the invention, each periodic time interval outside the first periodic time interval is taken as the target period interval, and the period interval preceding the target period interval is taken as the reference period interval. The thermal response time constant of the target period interval is determined by a weighted sum of the optimal characteristic time constant and the thermal response time constant of the reference period interval. In a specific implementation of this embodiment of the invention, the target period interval... The process of obtaining the thermal response time constant can be expressed by the formula: ;in, Target time period The thermal response time constant; Preset weighting parameters; For the target time period The optimal characteristic time constant of the preceding time period, i.e., the reference time period; For the target time period The thermal response time constant of the preceding time period, i.e., the reference time period; in a specific implementation of this invention... Setting it to 0.1 adjusts the degree of attention given to the optimal characteristic time constant. This can be adjusted according to the specific implementation environment, and will not be elaborated further here.

[0051] The thermal response time constant of the reference time period in the first cycle is a priori thermal response constant value. Furthermore, considering that there is no preceding cycle period in the first cycle, its thermal response time constant should essentially be obtained based on long-term historical monitoring data of the dam foundation through a macroscopic benchmark parameter extraction process. This provides an objective and reliable initial state benchmark for subsequent dynamic iterative analysis, improving the accuracy of the entire dynamic monitoring process. In a specific implementation of this invention, the process of obtaining the priori thermal response constant value includes: A Fast Fourier Transform (FFT) was performed on the time series of upstream water level to determine the corresponding complex spectrum of upstream water level; a FFT was also performed on the time series of downstream seepage flow to determine the corresponding complex spectrum of downstream seepage flow; the ratio between the mapping value of the daily fluctuation frequency point on the complex spectrum of downstream seepage flow and the mapping value on the complex spectrum of upstream water level was used as the daily periodic hydraulic admittance; where the angular frequency of the daily fluctuation frequency point is taken as... rad / s, which represents the angular frequency with a period of one natural day; the obtained diurnal periodic hydraulic admittance is essentially a complex value, which contains the system's response amplitude and phase delay information to diurnal periodic water level fluctuations; the diurnal periodic hydraulic admittance quantifies the dam foundation's flow response capability to external water level fluctuations.

[0052] Then, the thermal attenuation coefficient is determined based on the ratio between the variance of all elements in the dam area temperature time series and the variance of all elements in the downstream seepage water temperature time series. When the variance of all elements in the downstream seepage water temperature time series is 0, it is set to one-twentieth of the variance of all elements in the dam area temperature time series, resulting in a thermal attenuation coefficient of 20. This reflects a strong thermal damping effect of the dam foundation and better thermal stability of the system. If both the variances of all elements in the dam area temperature time series and the variances of all elements in the downstream seepage water temperature time series are 0, the thermal attenuation coefficient is set to 1, indicating that the temperature fluctuation has not changed. It should be noted that this situation is unlikely to occur in most real-world environments; this limitation is to prevent the formula from becoming meaningless and will not be elaborated further here.

[0053] The thermal attenuation coefficient quantifies the dam foundation's thermal damping capacity to external temperature fluctuations. If the dam foundation had no energy attenuation effect, temperature fluctuations would be transmitted directly to the seepage water, and the two variances should be approximately equal, resulting in a lower thermal attenuation coefficient. In reality, the dam foundation, as a massive rock and soil mass, acts as a highly efficient thermal buffer and damper with a large heat capacity. When drastic temperature fluctuations are transmitted through the dam foundation, their high-frequency components are strongly filtered out, and the fluctuation amplitude is significantly smoothed. This results in downstream seepage water temperature fluctuations typically being much smaller than those in the dam area. Therefore, a larger calculated thermal attenuation coefficient indicates a stronger thermal damping effect of the dam foundation and better system thermal stability.

[0054] Further, the hydraulic response characteristic time is determined by the ratio between the absolute value of the phase angle of the daily periodic hydraulic admittance and the angular frequency of the daily fluctuation frequency point; the a priori thermal response constant value is determined by the product of the positive correlation mapping value of the thermal attenuation coefficient, the preset proportional coefficient, and the combing response characteristic time. In a specific implementation of this invention, the process of obtaining the a priori thermal response constant value includes: ;in, This is the a priori thermal response constant value; As a preset scaling factor, this embodiment of the invention sets it to 1, which can be calibrated according to the geological data of the dam foundation in the specific implementation environment. It represents the empirical adjustment factor in the coupling relationship between hydraulic and thermal processes, and is used to make the macroscopic model more consistent with the geological structural characteristics of the specific dam foundation. The characteristic time of the hydraulic response; The thermal attenuation coefficient; It is a logarithmic function with the natural constant as its base.

[0055] In this context, the phase angle of the diurnal periodic hydraulic admittance in the hydraulic response characteristic time is measured in radians (rad), and the angular frequency of the diurnal fluctuation frequency point is measured in rad / s. Therefore, the unit of the obtained hydraulic response characteristic time is seconds (s), which characterizes the average time delay required for the system to generate a seepage response to diurnal periodic water level fluctuations. This is because, from a frequency domain perspective, the ratio of phase delay (radians) to angular frequency (radians / second) is directly converted to time (seconds). This definition is a standard method in signal and system theory for converting the phase difference in the frequency domain into a time delay in the time domain. The reason for using a logarithmic function with the natural constant as the base to perform a positive correlation mapping on the thermal attenuation coefficient is to convert the proportionality coefficient F, which characterizes the energy attenuation factor, into a dimensionless quantity that can be multiplied by the time dimension and characterizes the attenuation intensity by taking the logarithm. This operation originates from an ideal energy attenuation model, thus ensuring the consistency between the formula and the physical mechanism in mathematical form. It should be noted that the unit of hydraulic response characteristic time is seconds. In order to avoid the influence of the unit on the calculation of the a priori thermal response constant value, the corresponding unit of seconds is not considered in the calculation. Only the value before the second is used in the calculation of the a priori thermal response constant value. Alternatively, other methods can be used to avoid the influence of the unit on the calculation, such as by assigning a unit to the preset proportional coefficient to eliminate the influence of the unit of seconds in the calculation. This will not be elaborated further here.

[0056] This formula couples the measurable hydraulic response with the thermal decay, two independent but related physical processes, to deduce the system's intrinsic property, namely the thermal response time constant. This provides a clear, data-driven, and objective initial state for subsequent dynamic analysis, ensuring that the entire monitoring process does not start from an arbitrary guess but is based on a reliable benchmark calibrated with historical data. Therefore, the a priori thermal response constant value, which characterizes the thermal response time constant in the initial state, i.e., the first period, can be determined according to the above formula.

[0057] Step S103: Determine the stability determination coefficient based on the distribution of the relative deviations in energy and timing between the output events and input events among all preferred input and output events in the current period; perform dam foundation stability analysis based on the stability determination coefficient.

[0058] The preferred input-output event pairs characterize the physical essence of the most likely real and diagnostically valuable dominant heat transport paths within the dam foundation. To further analyze the stability of the dam foundation surface, it is necessary to combine the characteristics of the poorly stable through-flow melting channels and perform quantitative feature analysis on all preferred input-output event pairs. This allows for a comprehensive evaluation of the overall stability of the dam foundation surface in terms of feature matching with the through-flow melting channels. Therefore, this embodiment of the invention further determines the stability determination coefficient based on the distribution of the relative energy and temporal deviations between the output and input events in all preferred input-output events within the current period. In a specific implementation of this embodiment, the process of obtaining the stability determination coefficient includes: A continuous melting channel refers to a continuous flow path of ice-water mixture or liquid water with high water conductivity formed inside the dam foundation, which has a significant impact on the stability of the dam foundation surface. Therefore, the preferred input-output events corresponding to paths that conform to the characteristics of a continuous melting channel have the characteristics of short transport duration, high energy retention rate, and high signal waveform fidelity. Therefore, in the current cycle time period, the preferred input-output event pairs that meet the following conditions are used as reference input-output event pairs: transport duration less than a preset duration threshold, energy retention rate greater than a preset retention rate threshold, and waveform evolution cost less than a preset evolution threshold. In a specific implementation of the present invention, the preset duration threshold is set to the 10th percentile of the transport duration of all preferred input-output event pairs in all cycle time periods, the preset retention rate threshold is set to the 90th percentile of the energy retention rate of all preferred input-output event pairs in all cycle time periods, and the preset evolution threshold is set to the 10th percentile of the waveform evolution cost of all preferred input-output event pairs in all cycle time periods, so that the selected reference input-output event pairs are more consistent with the characteristics of a continuous melting channel.

[0059] Density clustering analysis is performed on the waveform evolution cost, transport duration, and energy retention rate of all reference input-output event pairs to determine all density clusters. The number of reference input-output event pairs in the density cluster with the most pairs is used as the stability determination coefficient. A large stability determination coefficient indicates that the density cluster belongs to a stable mode rather than random noise. In this case, a high-risk structural anomaly is considered to have occurred inside the dam foundation, and the corresponding stability is usually poor. Therefore, in this embodiment of the invention, when the stability determination coefficient is greater than a preset stability threshold, the dam foundation surface is determined to be unstable; when the stability determination coefficient is less than or equal to the preset stability threshold, the dam foundation surface is determined to be stable. In a specific implementation of this embodiment, the preset stability threshold is set to 10, which can be adjusted according to the number of sampling times in the specific implementation environment. The more sampling times, i.e., the more samples, the larger the preset stability threshold should be. This will not be elaborated further here. In one specific implementation of this invention, the density clustering method uses the DBSCAN clustering algorithm. The minimum number of points is set to 7 in this embodiment. The neighborhood radius is determined by analyzing the distribution of the distances between all points in the cluster sample and their k-th nearest neighbors. The distance value at the inflection point of the distance distribution curve is selected as the neighborhood radius value. This value can be subjectively set according to the specific implementation environment, and will not be elaborated further here.

[0060] In another specific implementation of this invention, when the dam foundation surface is determined to be unstable, the following steps are performed: First, a standard three-dimensional finite element model reflecting the general geological structure of the dam foundation is constructed as the basic data for the stability simulation analysis of the seepage model. Based on the geological survey data and engineering experience of the dam foundation, weak structural surfaces or regions most likely to form through-flow are identified in the model, such as fault fracture zones and the contact surface between the dam body and the bedrock, which are designated as potential risk channels. Then, the permeability coefficient of these potential risk channels is calibrated. In the model, an initial high permeability coefficient value is assigned to the potential risk channel region. A unit head pulse is applied upstream of the model to perform a transient seepage analysis, simulating the pressure response at the corresponding downstream location. Then, the transport duration of the simulated downstream pressure response sequence is calculated using the same method as in S102 of this scheme, obtaining the simulated transport duration. The energy retention rate of the simulated downstream pressure response sequence is then calculated using the same method as in S102 of this scheme, obtaining the simulated head fluctuation amplitude retention rate. Then, the reference input-output event pair corresponding to the cluster center point of the density cluster with the most reference input-output event pairs is taken as the feature input-output event pair. In this embodiment of the invention, the cluster center point is selected as the medoid point, which can be adjusted according to the specific implementation environment. The transport duration of the feature input-output event pair is taken as the feature duration. The energy retention rate of the feature input event pair is taken as the feature retention rate.

[0061] Further, the permeability coefficient value is iteratively adjusted until the simulated transport duration and simulated head fluctuation amplitude retention rate obtained from the simulation match the observed characteristic duration and characteristic retention rate, thus satisfying the minimization objective function. ;in, Permeability coefficient value The objective function value is as follows; Permeability coefficient value The simulated transport duration is as follows; For characteristic duration; Permeability coefficient value The simulation head fluctuation amplitude retention rate; Permeability coefficient value The characteristic retention rate is calculated. The permeability coefficient value that is minimized when the objective function value is minimized is taken as the final permeability coefficient value.

[0062] After calculating the final permeability coefficient, a targeted correction seepage model is obtained. This model is consistent with the initial model in macroscopic structure, but in key weak areas, the permeability coefficient values ​​are calibrated to the final permeability coefficient values, which can accurately reflect the current high-risk through-flow state. Finally, steady-state seepage analysis is performed based on the targeted correction seepage model to calculate the pore water pressure field and phreatic line position inside the dam foundation under the influence of the through-flow channel, generating a time-varying seepage field. Then, this pore water pressure field is used as a load condition and applied to the limit equilibrium method or finite element strength reduction method module to calculate the overall stability safety factor of the dam foundation along the preset sliding surface. Finally, the output overall stability safety factor is input into the database as the result of the dam foundation surface stability analysis.

[0063] In summary, a simulation analysis method for dam foundation stability under seepage pressure performs continuous wavelet transform on upstream and downstream temperature signals in each period, decomposing the temperature sequence into discrete events with temporal, energy, and waveform characteristics. A composite cost function including waveform distortion index and energy-time decay is constructed. Combined with an optimization algorithm, the optimal input-output event pairs corresponding to the most physically reliable advantageous transport path are dynamically identified. Then, considering the characteristics of the through-flow melting channel, which has a significant impact on stability, such as short transport duration, high energy retention rate, and high signal waveform fidelity, feature analysis is performed on the optimal input-output events. This makes the stability determination coefficient of the dam foundation more accurate, thus improving the accuracy of dam foundation stability analysis based on the stability determination coefficient.

[0064] This application also provides a simulation analysis system for the stability of dam foundation surfaces under seepage pressure. Please refer to [link / reference]. Figure 2The diagram shows a structural diagram of a dam foundation stability simulation analysis system under seepage pressure according to an embodiment of the present invention. The system includes: a data acquisition and preprocessing module 201, an event pair filtering module 202, and a stability analysis module 203.

[0065] The data acquisition and preprocessing module 201 is used to acquire the upstream boundary temperature sequence and the downstream seepage water temperature sequence of the dam foundation in each period; based on the upstream boundary temperature sequence, the corresponding input event set is determined by wavelet transform method; and based on the principle of acquiring the input event set, the corresponding output event set is determined based on the downstream seepage water temperature sequence. The event pair filtering module 202 is used to filter out all preferred input-output event pairs based on the causal relationship between input events in the input event set and output events in the output event set in terms of timing and energy, as well as the thermal response time constant for each period; wherein, the thermal response time constant is determined based on the time and energy distribution of the preferred input-output event pairs in the previous period. The stability analysis module 203 is used to determine the stability determination coefficient based on the distribution of the relative deviations in energy and timing between the output events and input events among all preferred input and output events in the current period; and to perform stability analysis of the dam foundation surface based on the stability determination coefficient.

[0066] It should be noted that the system provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer equipment can be divided into different functional modules to complete all or part of the functions described above. In addition, the dam foundation surface stability simulation analysis system under seepage pressure provided in the above embodiments and the dam foundation surface stability simulation analysis method embodiment under seepage pressure belong to the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0067] This application also provides a computer device; please refer to [link / reference]. Figure 3 The diagram illustrates a computer device structure according to an embodiment of the present invention. The computer device includes a memory 301, a processor 302, and a computer program 303 stored in the memory 301 and running on the processor 302. When the processor 302 executes the computer program 303, the computer device can perform any of the aforementioned simulation analysis methods for dam foundation stability under seepage pressure.

[0068] This application also provides a computer program product that, when run on a computer device, enables the computer device to execute any of the aforementioned simulation analysis methods for dam foundation stability under seepage pressure.

[0069] This application also provides a computer-readable storage medium storing computer program code. When the computer program code is run on a computer device, the computer device can execute any of the aforementioned simulation analysis methods for dam foundation stability under seepage pressure.

[0070] In the embodiments provided in this application, it should be understood that the computer device, computer program product and computer-readable storage medium provided are all used to perform the corresponding methods provided above, and therefore the beneficial effects they can achieve can be referred to the beneficial effects of the methods provided above, which will not be repeated here.

[0071] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0072] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A simulation analysis method for the stability of dam foundation surface under seepage pressure, characterized in that, The method includes: The upstream boundary temperature sequence and downstream seepage water temperature sequence of the dam foundation are obtained in each periodic time. Based on the upstream boundary temperature sequence, the corresponding input event set is determined by wavelet transform. Based on the principle of obtaining the input event set, the corresponding output event set is determined based on the downstream seepage water temperature sequence. Based on the causal relationship between the input events in the input event set and the output events in the output event set in terms of timing and energy, as well as the thermal response time constant for each period, all preferred input-output event pairs are selected; wherein, the thermal response time constant is determined based on the temporal and energy distribution of the preferred input-output event pairs in the previous period. Based on the distribution of the relative deviations in energy and timing between the output events and input events among all preferred input and output events in the current period, the stability determination coefficient is determined; the stability analysis of the dam foundation surface is then performed based on the stability determination coefficient. The process of obtaining the input event set includes: The upstream boundary temperature sequence is subjected to wavelet transform to determine the corresponding complex wavelet coefficient spectrum; all local maxima points on the modulus of the complex wavelet coefficient spectrum are statistically analyzed; the corresponding fluctuation peak time is determined according to the time corresponding to each local maxima point on the time axis; the corresponding fluctuation peak amplitude is determined according to the wavelet coefficient value of each local maxima point at the scale with the most concentrated energy; the temperature waveform feature vector of each local maxima point is determined by arranging the energy values ​​of the fluctuation peak time at all scales in ascending order of scale and normalizing them. The set of fluctuation peak time, fluctuation peak amplitude, and temperature waveform feature vector corresponding to each local maximum point of the upstream boundary temperature sequence is taken as an input event; the set of input events is determined based on the input events of all local maximum points corresponding to the upstream boundary temperature sequence.

2. The method for simulation analysis of dam foundation stability under seepage pressure according to claim 1, characterized in that, The process of acquiring the preferred input-output event pair includes: The output event set includes the peak fluctuation time, peak fluctuation amplitude, and temperature waveform feature vector of each output event corresponding to the downstream seepage water temperature sequence; element matching traversal is performed on the input event set and the output event set to determine all input-output event pairs; The corresponding cost element value is determined based on the deviations between the input and output events of each input-output event pair in terms of temperature waveform feature vector, peak fluctuation time, and peak fluctuation amplitude. Construct a cost matrix where rows represent input events and columns represent output events. Each element of the cost matrix is ​​the cost element value of an input-output event pair consisting of the input event in the corresponding row and the output event in the corresponding column. Input the cost matrix into the Hungarian algorithm and output the set of optimal matching pairs with the minimum total cost. The set of optimal matching pairs includes all preferred input-output event pairs.

3. The method for simulation analysis of dam foundation stability under seepage pressure according to claim 2, characterized in that, The process of acquiring the input / output event pair includes: Within each periodic time interval, an input event and an output event are randomly selected to form an input-output event pair; the set of input events and the set of output events are traversed to determine all input-output event pairs for each periodic time interval.

4. The method for simulation analysis of dam foundation stability under seepage pressure according to claim 2, characterized in that, The process of obtaining the cost element value includes: Filter out the ideal input-output event pairs from all input-output event pairs; in each ideal input-output event pair, determine the corresponding waveform evolution cost based on the Earth's movement distance between the temperature waveform feature vector of the input event and the temperature waveform feature vector of the output event; The energy decay performance value is determined by positively mapping the ratio between the peak amplitude of the output event and the peak amplitude of the input event; the time interval between the peak moments of the output event and the peak moments of the input event is calculated; the time interval is weighted by the negative correlation mapping value of the thermal response time constant of each ideal input-output event to its periodic time interval to determine the time decay coefficient; and the comprehensive decay cost is determined based on the absolute value of the sum of the energy decay performance value and the time decay coefficient. The cost element value for each ideal input-output event pair is determined by weighted summation of the waveform evolution cost and the comprehensive attenuation cost. Set the cost element values ​​of all other input / output event pairs of the ideal input / output event to the preset cost constant.

5. The method for simulation analysis of dam foundation stability under seepage pressure according to claim 4, characterized in that, The process of acquiring the ideal input-output event pair includes: An ideal input-output event pair is one where the peak value of the input event precedes the peak value of the output event, and the amplitude of the peak value of the input event is greater than that of the peak value of the output event.

6. The method for simulation analysis of dam foundation stability under seepage pressure according to claim 2, characterized in that, The process of obtaining the thermal response time constant includes: In each preferred input-output event pair within each cycle time period, the corresponding transport duration is determined based on the time interval between the peak fluctuation time of the output event and the peak fluctuation time of the input event; the corresponding energy retention rate is determined based on the ratio between the peak fluctuation amplitude of the output event and the peak fluctuation amplitude of the input event. The optimal characteristic time constant is determined based on the concentrated distribution of waveform evolution cost, transport duration, and energy retention rate of all preferred input and output events. Each period outside the first period is taken as the target period, and the period preceding the target period is taken as the reference period. The thermal response time constant of the target period is determined by weighted summation of the optimal characteristic time constant of the reference period and the thermal response time constant of the reference period. The thermal response time constant of the reference period of the first period is a priori thermal response constant value.

7. The method for simulation analysis of dam foundation stability under seepage pressure according to claim 6, characterized in that, The process of obtaining the optimal characteristic time constant includes: Density clustering analysis is performed based on the waveform evolution cost, transport duration, and energy retention rate of all preferred input-output event pairs to determine at least two event pair clusters. In the event pair cluster with the most preferred input-output event pairs, the optimal characteristic duration is determined based on the mean transport duration of all preferred input-output event pairs. The optimal retention rate is determined based on the mean energy retention rate of all preferred input-output event pairs. The optimal characteristic time constant is determined by performing a negative correlation mapping between the negative correlation mapping value of the optimal retention rate and the product of the optimal characteristic duration.

8. The method for simulation analysis of dam foundation stability under seepage pressure according to claim 6, characterized in that, The process of obtaining the stability determination coefficient includes: In the current period, the preferred input-output event pair that satisfies the following conditions is that the transport duration is less than the preset duration threshold, the energy retention rate is greater than the preset retention rate threshold, and the waveform evolution cost is less than the preset evolution threshold. Density clustering analysis is performed on the waveform evolution cost, transport duration, and energy retention rate of all reference input-output event pairs to determine all density clusters; the number of reference input-output event pairs in the density cluster with the most reference input-output event pairs is used as the stability determination coefficient.

9. The method for simulation analysis of dam foundation stability under seepage pressure according to claim 1, characterized in that, The process of performing stability analysis of the dam foundation surface based on the stability determination coefficient includes: When the stability determination coefficient is greater than the preset stability threshold, the dam foundation is determined to be unstable; when the stability determination coefficient is less than or equal to the preset stability threshold, the dam foundation is determined to be stable.

Images