Water gate risk real-time evolution prediction and evaluation method

Abstract

The invention discloses a sluice risk real-time evolution prediction and evaluation method, and relates to the technical field of hydraulic engineering, and the method comprises the steps: synchronously collecting real-time multi-physics field data of a target region in a foundation through an in-situ multi-parameter collaborative sensing network; extracting a multi-dimensional spatial-temporal feature sequence from the real-time multi-physics field data, and matching the multi-dimensional spatial-temporal feature sequence with a seepage evolution mode record pre-constructed in a multi-physics field data fusion and knowledge base to identify a collaborative anomaly mode representing a current seepage state; predicting a risk evolution trend based on the identified collaborative abnormal mode; according to the method, the real-time perspective of the three-dimensional structure and dynamic evolution of the seepage network in the foundation can be realized, the real-time perspective of the three-dimensional structure of the seepage network in the foundation and the dynamic evolution of the seepage network in the foundation can be realized, and the real-time perspective of the three-dimensional structure of the seepage network in the foundation can be realized before the occurrence of catastrophe such as piping. And sending out critical state early warning.

Description

Technical Field

[0001] The present invention relates to the technical field of water conservancy projects, and in particular to a method for predicting and evaluating the real-time evolution of sluice risks. Background Art

[0002] As a key water conservancy hub that controls water flow and ensures flood control, water supply, and navigation, the safety of its foundation is paramount. Among the many factors that can cause sluice gate instability, hidden seepage in deep foundations has long been recognized as a major safety hazard. Under high water heads, seepage within the foundation is not static. It can gradually evolve from a diffuse Darcy flow, which poses no threat, to a non-Darcy rapid flow dominated by dominant channels. This can eventually deteriorate rapidly within a short period of time, forming a continuous piping channel, leading to structural damage to the foundation soil and even catastrophic dam failures such as "tower sweeping."

[0003] However, how to effectively monitor and provide early warning for this dynamic evolution from quantitative change to qualitative change is a major, long-standing, and unresolved technical bottleneck in this field. Existing technical means have fundamental limitations, and their inability is mainly reflected in: Traditional monitoring methods fail to reveal the full picture of the network. Currently used techniques, such as embedding a small number of pressure gauges within the sluice foundation, are essentially "point-based" monitoring. Data from these discrete measurement points can only reflect pore water pressure at a localized location, completely failing to capture the complex morphology, connectivity, and dynamic evolution of seepage channels in three-dimensional space. The vast areas between measurement points become vast monitoring "blind spots," where advantageous seepage channels can develop and breed, rendering existing monitoring systems ineffective.

[0004] Existing geophysical exploration techniques are unable to achieve dynamic and accurate internal perspectives. Although geophysical exploration techniques such as the high-density resistivity method (ERT) have been tried for seepage detection, these techniques have serious shortcomings: First, their extremely low resolution makes it difficult to identify the early development of micro-fractures or dominant channels; second, their static or low-frequency measurements cannot capture the instantaneous and rapid changes in seepage conditions during critical moments such as floods; more importantly, their measurements are a comprehensive reflection of multiple geological factors and cannot accurately and independently infer the dynamic evolution of hydraulic conductivity parameters (such as the permeability field) that characterize the core characteristics of seepage, thus failing to provide accurate information for decision-making.

[0005] Models are disconnected from reality and lack deep coupling with physical mechanisms. Existing analytical methods typically rely on simplified, static seepage models, completely ignoring the complex dynamic coupling between seepage fields, stress fields, rock and soil structural damage, and multi-physics responses. This severe disconnect between these models and reality renders them unable to explain observed anomalies and make reliable predictions about future evolutionary trends. Summary of the Invention

[0006] The present invention provides a method for real-time evolution prediction and assessment of sluice risks to solve the technical problems in the prior art that there are blind spots in the monitoring of the hidden seepage risks of the deep foundation of sluices, the dynamic evolution process of the seepage network cannot be revealed, the early warning is seriously lagged, and accurate pre-event prediction and assessment cannot be achieved.

[0007] In view of the above problems, the present invention provides a method for real-time evolution prediction and assessment of sluice risks, including: Synchronously collecting real-time multi-physical field data of the target area inside the foundation through an in-situ multi-parameter collaborative sensing network arranged inside the deep foundation of the sluice; Extracting a multi-dimensional spatio-temporal feature sequence from the real-time multi-physical field data, and matching the multi-dimensional spatio-temporal feature sequence with the seepage evolution pattern records pre-constructed in the multi-physical field data fusion and knowledge base to identify a collaborative abnormal pattern characterizing the current seepage state; Predicting its risk evolution trend based on the identified collaborative abnormal pattern; Comprehensively considering the collaborative abnormal pattern and the risk evolution trend, and integrating a preset physical instability criterion, to conduct real-time assessment of the sluice risk, and generating a warning message when the assessment result reaches a preset condition.

[0008] Preferably, the method further includes the step of pre-constructing the seepage evolution pattern records, and the step includes at least one of the following: establishing a multi-physical field coupling mathematical model capable of reflecting the specific geological structure of the target sluice project, and conducting multi-condition dynamic simulations on preset typical seepage evolution paths, recording the whole-process multi-physical field response data of virtual sensors, and summarizing and extracting the seepage evolution pattern records therefrom; or, by conducting a physical model test or in-situ controlled test capable of simulating the specific seepage failure process of the target sluice, arranging a sensor array in the test model and synchronously collecting multi-physical field data, and correlating the response characteristics observed in the test with the actually occurring seepage evolution state to calibrate and generate the seepage evolution pattern records.

[0009] Preferably, the method further includes: before or during the step of collecting the real-time multi-physical field data, actively emitting a controllable elastic wave signal through active excitation and auxiliary observation, and receiving it by the in-situ multi-parameter collaborative sensing network to enhance the detection ability of the dynamic changes of the foundation medium structure.

[0010] Preferably, the method further includes: based on the real-time multi-physical field data and the physical model constraints in the knowledge base, through multi-physical field joint inversion, real-time reconstruction of the three-dimensional effective hydraulic conductivity characteristic field and the dominant seepage network topology structure inside the foundation; and presenting the three-dimensional effective hydraulic conductivity characteristic field and the dominant seepage network topology structure through a three-dimensional dynamic visualization interface.

[0011] The technical solution provided by this application has at least the following technical effects or advantages: The present invention obtains all-round information through an in-situ multi-parameter collaborative sensing network, and through multi-physical field data fusion, knowledge base, and subsequent diagnosis and inversion, realizes the dynamic, three-dimensional, and quantitative characterization of the hidden seepage network inside the deep foundation of the sluice and its evolution process. This enables the safety risk assessment to shift from discrete observations to a global perspective.

[0012] By establishing a collaborative abnormal pattern recognition and risk evolution trend prediction mechanism, the present invention can accurately capture the physical precursor information before the qualitative change indicating catastrophic instability such as piping occurs. This enables the risk control to be upgraded from passive response to proactive prediction, and secures an important emergency response window period for project managers.

[0013] The final output of the present invention is not just a simple alarm signal, but warning information including risk level, spatial location, evolution mode, and development trend. This provides a scientific and accurate decision-making basis for the daily maintenance, danger removal and reinforcement, and emergency rescue of the sluice, can significantly improve the scientific, intelligent, and refined level of safety operation and maintenance, avoid unnecessary engineering investment, and reduce the operation and maintenance costs throughout the life cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 The figure is a flowchart of a method for real-time evolution prediction and assessment of sluice risks according to the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0015] The present invention relates to a method for real-time evolution prediction and assessment of sluice risks, aiming to solve the technical problems in the prior art that there are "blind spots" in the monitoring of the hidden seepage risks of the deep foundation of sluices, the dynamic evolution process of the seepage network cannot be revealed, the early warning is severely lagged, and accurate prediction and assessment before the event cannot be achieved.

[0016] The above technical solutions will be described in detail below in combination with the specification drawings and specific embodiments to better understand the above technical solutions. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all embodiments of the present invention. It should be understood that the present invention is not limited to the example embodiments that only explain the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention. In addition, it should be noted that for the sake of description, only the parts related to the present invention are shown in the drawings, rather than all of them.

[0017] As Figure 1 shown in the flowchart of a real-time evolution prediction and evaluation method for sluice risks, the method includes the following steps: Synchronously collect real-time multi-physical field data of the target area inside the foundation through an in-situ multi-parameter collaborative sensing network arranged inside the deep foundation of the sluice. Specifically, in the key areas inside the foundation, an underground network capable of collecting multi-dimensional physical information in a long-term, stable, synchronous, and high-frequency manner is constructed. The target area is determined according to the engineering geological exploration report of the sluice, historical seepage disease data, and the preliminary numerical simulation results of the seepage field, and usually includes key parts such as under the sluice floor, the bypass seepage areas on both banks, the inside and outside of important anti-seepage structures, and geological defect zones.

[0018] In a specific embodiment, the step of collecting the real-time multi-physical field data includes: Synchronously collect through multiple in-situ multi-parameter collaborative sensing probes arranged in a three-dimensional array in the target area inside the foundation. Among them, the in-situ multi-parameter collaborative sensing probes are all composite detection units, and integrally integrated inside are: pore water pressure sensors, triaxial broadband acceleration sensors, temperature sensors, and resistivity measurement units.

[0019] Specifically, the in-situ multi-parameter collaborative sensing probes are designed to synchronously collect multiple key physical parameters at the same physical position or within a very short distance, so as to ensure the strong correlation of data in time and space and lay a foundation for subsequent collaborative anomaly pattern recognition.

[0020] Furthermore, the pore water pressure sensor is preferably a high-precision microelectromechanical system piezoresistive or capacitive sensor, which is used to accurately monitor the absolute value of the pore water pressure in the foundation and its dynamic changes. This is the most direct indicator for judging the seepage driving force. The triaxial broadband acceleration sensor is preferably a high-sensitivity microelectromechanical system accelerometer, and its frequency band needs to cover the range from passive acoustic emission or microseismic signals (usually generated by soil particle friction and microcrack propagation) to actively excited elastic wave signals, so as to capture the damage of the medium microstructure and detect the changes in the mechanical properties of the medium. The temperature sensor is preferably a platinum resistance with Class A accuracy, which is used to monitor the temperature field of the foundation and indirectly invert the seepage path and flow velocity information through the heat convection effect brought by seepage. The resistivity measurement unit preferably adopts a micro four-electrode arrangement to measure the apparent resistivity of the small-range rock and soil around the probe, and its change can effectively reflect the changes in the water saturation, pore connectivity and pore water properties of the medium.

[0021] In another specific embodiment, the method further includes: Before or during the step of collecting the real-time multi-physical field data, a controllable elastic wave signal is actively emitted through an active excitation and auxiliary observation system and received by the in-situ multi-parameter collaborative sensing network to enhance the detection ability of the dynamic changes of the foundation medium structure.

[0022] Specifically, to further improve the detection accuracy and the reliability of inversion, this method has an active excitation step. By installing a downhole controllable elastic wave excitation source, such as a downhole sparker or a piezoelectric ceramic transducer array, in a dedicated excitation hole on the periphery of the monitoring area or at a specific depth, elastic waves with controllable energy and known waveforms are regularly and on-demand emitted into the foundation medium. The triaxial broadband acceleration sensor in the in-situ multi-parameter collaborative sensing probe is responsible for receiving the elastic wave signals after propagation through the medium. By analyzing the wave travel time, amplitude attenuation and waveform changes, techniques such as tomography can be used to more accurately invert the mechanical parameters such as elastic modulus, density, and Poisson's ratio inside the foundation and their dynamic changes, providing stronger constraints for evaluating the damage of the soil structure.

[0023] Extract a multi-dimensional spatio-temporal feature sequence from the real-time multi-physical field data, and match the multi-dimensional spatio-temporal feature sequence with a record of the seepage evolution pattern pre-constructed in the multi-physical field data fusion and knowledge base to identify the collaborative abnormal pattern characterizing the current seepage state; Specifically, first, for the original data or the preliminarily processed data of each physical field collected, online feature engineering is carried out to extract key feature parameters that can more directly and sensitively reflect the seepage evolution state, and a multi-dimensional feature vector sequence arranged in chronological order is constructed, that is, a multi-dimensional spatio-temporal feature sequence.

[0024] Further, the multi-physical field data fusion and knowledge base store a large number of records of seepage evolution patterns. These records uniquely and quantitatively associate the abstract and invisible seepage evolution stages with a set of measurable and specific multi-physical field co-response characteristic spectra.

[0025] In a specific embodiment, the records of seepage evolution patterns include: A list of key physical field characteristic parameters for association; A defined quantitative reference interval or statistical distribution model for the key characteristic parameters in the list; Rules or association models for describing the co-variation relationships between the characteristic parameters of different physical fields under the seepage evolution pattern.

[0026] Specifically, the records of seepage evolution patterns are all structured digital records, which include: 1. Basic pattern information: including the unique identifier of the pattern and a clear written description of the seepage evolution pattern, such as: "Initial stable pore seepage", "Initial stage of the formation of a linear dominant seepage channel along a specific geological structure plane", "Accelerated period of the development of piping channels in a sandy gravel foundation accompanied by the initiation of fine particles", etc.

[0027] 2. Multi-physical field co-response characteristic spectra: The list of key physical field characteristic parameters for association: clearly list the key characteristic parameters of each physical field that are most indicative under this pattern. Specifically, this list is selected and combined from the following sets: Acoustic characteristic parameters: For example, the cumulative number of acoustic emission events, the average event energy, the logarithmic distribution slope of event energy, the statistical distribution of the peak frequency of event waveforms, and the aggregation density of sound source events in three-dimensional space.

[0028] Electrical characteristic parameters: For example, the average apparent resistivity value, the change rate of apparent resistivity, the apparent resistivity spectrum characteristics under multi-frequency excitation, and the spatial gradient of apparent resistivity between adjacent monitoring units.

[0029] Thermal characteristic parameters: For example, the current temperature value, the deviation value relative to the background temperature, the temperature change rate, and the temperature gradients along the water flow direction and perpendicular to the water flow direction.

[0030] Hydraulic characteristic parameters: For example, the excess pore water pressure value, the fluctuation amplitude and dominant frequency of pore water pressure in the short term, and the response amplitude and phase lag time of pore water pressure to external head changes.

[0031] The quantitative reference interval or statistical distribution model: For each key characteristic parameter in the list, give its typical numerical range or the key parameters of the probability density function under this pattern.

[0032] The rules or correlation models of the co-variation relationship: Describe how the characteristic parameters of different physical fields co-vary under this pattern. Its manifestation forms can be: Rule sets based on logical conditions: For example, "IF (the rate of change of resistivity < -X% / day) AND (the rate of high-energy acoustic emission events > Y events / hour) THEN (highly consistent with the co-variation characteristics of pattern A)".

[0033] Mathematical models based on statistical correlations: For example, describe the functional relationship between the change of resistivity and the change of saturation.

[0034] Synchronization or lag characteristics in time series: For example, "Usually, a significant decrease in resistivity is first observed, and then the acoustic emission activity reaches its peak within XX hours".

[0035] In another specific embodiment, the method further includes the step of pre-constructing the seepage evolution pattern record, and the step includes at least one of the following: By establishing a multi-physical-field coupling mathematical model that can reflect the specific geological structure of the target sluice project, and performing parametric design and multi-condition dynamic simulation on the preset typical seepage evolution paths, recording the full-process multi-physical-field response data of the virtual sensors, and inducing and extracting the seepage evolution pattern record therefrom; Or by conducting a physical model test or in-situ controlled test that can simulate the specific seepage failure process of the target sluice, arranging a sensor array in the test model and synchronously collecting multi-physical-field data, and correlating the response characteristics observed in the test with the actually occurring seepage evolution state to calibrate and generate the seepage evolution pattern record.

[0036] Specifically, to ensure the scientificity and reliability of the knowledge base, its initialization and construction process is extremely rigorous: 1. Based on high-fidelity multi-physical-field coupling numerical simulation: First, according to the detailed geological exploration report of the target sluice, establish a refined hydraulic-geotechnical-acoustic-electric-thermal multi-physical-field coupling mathematical model.

[0037] Second, for several key hidden seepage evolution paths, set a group of key control parameters that can characterize the whole process and their reasonable values.

[0038] Then, perform unsteady multi-physical-field coupling simulation on each combination in the parameter space, and record all the physical-field response time series data obtained by the virtual sensors.

[0039] Finally, signal processing and pattern recognition algorithms are used to analyze a large amount of analog data, and the results with similar multi-physical field collaborative response evolution characteristics and clearly corresponding to specific seepage stages are summarized into a seepage evolution pattern record.

[0040] Calibration and verification based on large-scale or in-situ controlled physical model tests: Design and implement large-scale indoor physical model tests or small-scale in-situ tests that can truly simulate key seepage failure processes.

[0041] Deploy a sensor array similar to the actual project in the test model, precisely control the boundary conditions, and synchronously collect multi-physical field data at high frequency until the model undergoes the expected seepage failure.

[0042] After the test, accurately determine the actual development of the internal seepage network and the failure mechanism through methods such as excavation and dissection, and dye tracer tracking.

[0043] Precisely correlate the multi-physical field response characteristics observed in the test with the actual evolution state, and use it to strictly calibrate and verify the seepage evolution pattern record obtained through numerical simulation.

[0044] Maintenance and optimization mechanism of the knowledge base: After actual operation, use the real case data verified by engineering surveys to finely adjust the parameters and rules of the relevant records in the knowledge base. At the same time, organize domain experts to regularly review and inject knowledge into the knowledge base, and establish a strict version management system to ensure that the quality of the knowledge base keeps pace with the times.

[0045] In a preferred embodiment, the step of matching to identify the collaborative abnormal pattern includes: Convert the collaborative response rule into a set of executable logical judgment rules; According to the current eigenvalue in the multi-dimensional spatio-temporal feature sequence, calculate its membership degree for the logical judgment rule; Adopt weighted rules and fuzzy logic reasoning, synthesize the activation intensity of all logical judgment rules and their preset weights, and calculate the matching confidence score of the current monitoring area with each seepage evolution pattern record to identify the collaborative abnormal pattern.

[0046] The specific steps are as follows: First, turn the "collaborative response rules" in the knowledge base into a computer program. Then, for the feature values collected in real time, calculate the degree of compliance with the "normal range" defined in the rules, that is, the "membership degree" (a value from 0 to 1). Finally, using methods such as fuzzy comprehensive evaluation, comprehensively consider the degree of compliance of all rules and the preset expert weights (for example, the weight of water pressure change may be higher than that of temperature change), and finally calculate the similarity score between the current state and each "seepage evolution mode" (such as "piping incubation period") in the knowledge base, that is, the "matching confidence score". The mode with the highest score is the most likely current collaborative abnormal mode.

[0047] In another preferred embodiment, the step of matching to identify the collaborative abnormal mode may also adopt dynamic state sequence decoding based on a probabilistic graphical model (such as a hidden Markov model). Specifically, a hidden Markov model is pre-constructed, and its set of hidden states corresponds to several key and ordered seepage evolution stages defined in the knowledge base. The emission probability and transition probability of the model are trained through historical data or simulation data. In real-time applications, the Viterbi algorithm is used to dynamically decode the most likely hidden state sequence, that is, the most likely seepage evolution stage, experienced by the current monitoring area in the past period of time according to the input multi-dimensional spatio-temporal feature sequence.

[0048] Based on the identified collaborative abnormal mode, predict its risk evolution trend; Specifically, after identifying the current collaborative abnormal mode, this step uses an advanced time series prediction model (for example, long short-term memory network) to learn the historical time series data of each key risk indicator (such as the matching confidence score with the high-risk mode, the indirect estimated value of local permeability, the acoustic emission energy release rate, etc.) in this mode, so as to predict its possible change trend within a short time window in the future (for example, in the next few hours to several days).

[0049] Further, after the step of matching to identify the collaborative abnormal mode, the method further includes: Regard all monitoring units in the target area inside the foundation as nodes in graph theory, and based on the matching confidence scores of each monitoring unit, construct and dynamically update a seepage activity network that characterizes the spatial distribution and association state of the dominant seepage activity area in real time; Adopt graph theory analysis algorithms to quantitatively analyze the topological structure of the seepage activity network and its evolution characteristics over time.

[0050] Specifically, to evaluate risks from a global perspective, all monitoring units represented by in-situ multi-parameter collaborative sensing probes are regarded as nodes in graph theory. According to the matching confidence scores calculated in real time for each node, an edge representing the association of seepage activities is defined between multiple spatially adjacent nodes whose scores all indicate a high risk of seepage activities. The weight of the edge can be determined based on factors such as the similarity of scores and distance between nodes. Thus, a weighted "seepage activity network" is constructed and dynamically updated in real time. Then, graph theory analysis algorithms (such as connected component identification, shortest path algorithm, critical node identification algorithm, etc.) are used to quantitatively analyze the topological structure of this network (for example, whether there is a highly active, nearly connected path or network cluster extending from upstream to downstream) and its evolution characteristics over time. This analysis from "point" to "net" is the key to evaluating global and structural risks, and its results are also an important basis for predicting future risk evolution trends.

[0051] Based on the comprehensive collaborative anomaly patterns and the risk evolution trends, and by integrating preset physical instability criteria, the risks of the sluice are evaluated in real time, and warning information is generated when the evaluation results meet the preset conditions.

[0052] Specifically, this step is the decision output of this method. The strategy of multi-criterion fusion is adopted to avoid the risk of misjudgment by a single index.

[0053] In a preferred embodiment, the step of evaluating the risks of the sluice in real time includes: Comprehensively evaluating the following criteria: whether the risk status of the local monitoring unit continuously remains or rapidly enters the preset dangerous precursor level, whether the global topological structure of the seepage activity network reaches or approaches the critical connection threshold, and the proximity of the calculated local hydraulic gradient or energy dissipation index to the physical instability criteria; And determining the risk assessment result according to the comprehensive satisfaction of the multiple criteria.

[0054] Specifically, the final risk assessment and early warning trigger incorporate at least three levels of information: First, the risk status of local monitoring units, i.e., whether the confidence score of the collaborative abnormal pattern matching of a single monitoring point continues to be at a high level or is increasing rapidly. Second, the global topological criticality of the "dominant seepage activity area network", i.e., whether the connectivity, penetration, or expansion rate of the network reaches or approaches its critical percolation threshold. Third, the degree of compliance with known physical instability criteria, i.e., comparing parameters such as indirectly deduced hydraulic gradients with classical soil mechanics instability criteria (such as Terzaghi's piping criterion) to evaluate the degree of proximity. Only when the information at multiple levels jointly indicates danger will the system dynamically evaluate and output the corresponding early warning level (for example, divided into four levels: daily attention, general early warning, relatively severe early warning, and severe alarm) through a preset decision logic system that can be adjusted by experts (such as a decision table based on weighted scoring or a simple neural network classifier), and generate detailed early warning information including the risk location, type, and evolution trend.

[0055] In a preferred embodiment, the method further includes: Based on the real-time multi-physical field data and the physical model constraints in the knowledge base, through multi-physical field joint inversion, the three-dimensional effective hydraulic conductivity characteristic field and the dominant seepage network topological structure inside the foundation are reconstructed in real time; And through a three-dimensional dynamic visualization interface, the three-dimensional effective hydraulic conductivity characteristic field and the dominant seepage network topological structure are presented.

[0056] Specifically, using a multi-physical field joint inversion engine, all the collected data are subjected to in-depth fusion calculations to directly invert the distributions of key physical parameters such as the three-dimensional permeability field, saturation field, and porosity field inside the foundation. Then, through a three-dimensional dynamic visualization platform, these invisible geological parameter fields and the identified dominant seepage network are intuitively presented to engineering and technical personnel in the forms of animations, profiles, streamlines, etc., enabling them to clearly "see" the location, shape, and change process of seepage risks over time, providing the most powerful data support for the final risk assessment and disposal decision-making.

[0057] In a specific embodiment: A concrete gravity dam that has been in operation for 40 years. There are known and minor problems of seepage around the dam. Near the dam foundation gallery (observation gallery) on the right bank, there have been a small number of seepage records in history, especially during the high water level period in the flood season. The operation and maintenance team is worried that as the dam ages and the number of extreme climate events increases, concentrated dominant seepage channels may evolve in this weak area, ultimately leading to risks of piping or seepage failure.

[0058] Step 1: Deployment of the perception layer: There is no need to comprehensively cover the entire dam foundation. Based on historical data and geological analysis, only in the identified highest-risk area - around the gallery of the right-bank dam foundation and on the downstream side, 8 "in-situ multi-parameter collaborative perception probes" are arranged in a "pin" shape by means of drilling and implantation. This quantity is sufficient to form a local high-density monitoring network with basic three-dimensional analysis capabilities.

[0059] Temporarily do not deploy the "active excitation and auxiliary observation system" and rely entirely on passive synchronous acquisition.

[0060] Step 2: Construction of the knowledge base: There is no need to construct a huge knowledge base containing dozens of patterns. First, focus on the core issues to be verified this time and only construct 3 of the most critical "seepage evolution pattern records": PEMR-01: Steady seepage pattern (baseline state) PEMR-02: Pattern of uniform increase in saturation (normal response under high water level) PEMR-03: Pattern at the initial stage of the formation of the dominant channel along microfractures (core early warning target) For the construction of the knowledge base, the method of "high-fidelity multi-physical field coupling numerical simulation" is mainly adopted. A refined local numerical model containing only the right-bank dam section and the foundation is established. By setting different boundary conditions and microfracture parameters in the model, several simulations are run, and the "multi-physical field collaborative response characteristic spectra" corresponding to the above 3 patterns are extracted and calibrated.

[0061] Step 3: Implementation of the diagnosis and early warning engine - Diagnosis algorithm: Adopt the relatively easy-to-implement "matching confidence scoring method based on weighted rules and fuzzy logic reasoning".

[0062] Early warning logic: Set a simplified double-layer early warning logic: Attention (blue early warning): For the state detected by any one probe, its matching confidence scoring with "PEMR-03" (initial stage of the formation of the dominant channel) continuously exceeds 40 points within 24 hours.

[0063] Early warning (yellow early warning): For at least 2 spatially adjacent probes, their matching confidence scoring with "PEMR-03" continuously exceeds 70 points within 6 hours.

[0064] Temporarily do not conduct global topological analysis of the complex "seepage activity network", and the early warning is mainly based on the risk status of local monitoring units.

[0065] Step 4: Result presentation: Only a simple Web dashboard is required.

[0066] Dashboard interface: On the left: Display a schematic diagram of the positions of 8 probes.

[0067] Middle: In the form of a line chart, it shows the 24-hour change curves of the key physical parameters (such as excess pore water pressure and resistivity) of each probe in real time.

[0068] Right side: In the form of a dashboard or bar chart, it shows the current system status and the "matching confidence score" of 3 "seepage evolution mode records" in real time.

[0069] Top: A prominent status bar shows "System Normal", "Blue Attention" or "Yellow Warning".

[0070] Notification method: When a yellow warning is triggered, the system sends a warning notice to the dam safety responsible person via email and text message.

[0071] Operation demonstration in the scenario: Deployment and baseline learning: After the system is deployed, it runs for one month during the dry season. The dashboard shows that the matching confidence score of the system status and "PEMR-01: Stable seepage mode" remains above 90 points, and the scores of the other two modes are extremely low. The operation and maintenance team confirmed the normal baseline of the system.

[0072] Event occurrence: Entering the flood season, the upstream water level rises rapidly by 15 meters within 3 days.

[0073] System response and diagnosis: The pore water pressure readings of all 8 probes have increased significantly, and the resistivity has generally decreased.

[0074] On the dashboard, the score of "PEMR-01" drops rapidly, and the score of "PEMR-02: Uniform saturation rising mode" rises sharply and then stabilizes, which is in line with the normal response under high water levels.

[0075] Appearance of key changes: For Probes 3 and 5 located on the downstream side of the gallery, the resistivity drops significantly faster than other probes, and at the same time, their acceleration sensors capture weak but continuous and high-frequency acoustic emission signals.

[0076] Diagnostic engine analysis: The system matches these features with the knowledge base in real time. Due to the abnormal accelerated drop of resistivity (indicating the possible formation of local water passage) and acoustic emission signals (indicating soil particle friction or microfracture activity), the system calculates the status of the areas of Probes 3 and 5, and the matching confidence score with "PEMR-03: Initial mode of formation of dominant microfracture channels" rises rapidly within a few hours and stabilizes above 75 points.

[0077] Warning output: The status bar at the top of the dashboard turns yellow and flashes to prompt "Yellow Warning".

[0078] Probes 3 and 5 become highlighted in red on the schematic diagram.

[0079] The mobile phone of the person in charge of dam safety received a text message: "HYDRA-VIEW warning: The risk status in the downstream monitoring area of the right bank corridor (sensors 3, 5) highly matches the 'initial mode of forming the dominant channel' (score 78 / 100). Please immediately activate the key attention plan." After receiving the warning, the operation and maintenance team was able to take preventive measures such as curtain grouting in a targeted manner, effectively avoiding the further deterioration of potential safety hazards.

[0080] In summary, the real-time evolution prediction and assessment method for the risk of sluice gates provided by the present invention realizes the dynamic, accurate, and forward-looking diagnosis and warning of the hidden seepage risk of the deep foundation of the sluice gate through multi-physical field collaborative perception, pattern matching based on a knowledge base, trend prediction based on data driving, and risk assessment based on multi-criteria fusion. Fundamentally overcoming the limitations of the prior art, it achieves the technical effects of changing the safety monitoring of sluice gates from a "blind area" to a "measurable area" and changing the risk response from "passive response" to "active anticipation", and has significant technological progress and great engineering application value.

[0081] It should be understood that the disclosed embodiments of the present invention and the above descriptions enable those skilled in the art to implement the present invention using the present invention. At the same time, the present invention is not limited to the above-mentioned part of the embodiments. It should be understood that those of ordinary skill in the art can still modify the technical solutions recorded in the foregoing embodiments or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A real-time evolution prediction and evaluation method for sluice risks, characterized in that Including: Synchronously collecting real-time multi-physical field data of a target area inside the foundation through an in-situ multi-parameter collaborative sensing network deployed inside the deep foundation of the sluice; Extracting a multi-dimensional spatio-temporal feature sequence from the real-time multi-physical field data, and matching the multi-dimensional spatio-temporal feature sequence with the seepage evolution pattern records pre-constructed in the multi-physical field data fusion and knowledge base to identify a collaborative anomaly pattern representing the current seepage state; Predicting the risk evolution trend based on the identified collaborative anomaly pattern; Comprehensively considering the collaborative anomaly pattern and the risk evolution trend, and integrating a preset physical instability criterion to conduct real-time assessment of the sluice risk, and generating a warning message when the assessment result reaches the preset condition.

2. The real-time evolution prediction and assessment method for the risk of a sluice as claimed in claim 1, wherein The step of collecting real-time multi-physical field data includes: synchronously collecting through multiple in-situ multi-parameter collaborative sensing probes deployed in a three-dimensional array in the target area inside the foundation; wherein, the in-situ multi-parameter collaborative sensing probes are all composite detection units, and integrally integrated inside are: pore water pressure sensors, triaxial broadband acceleration sensors, temperature sensors, and resistivity measurement units.

3. The real-time evolution prediction and evaluation method for the risk of a sluice as claimed in claim 1, characterized in that, The seepage evolution pattern records include: a list of associated key physical field characteristic parameters; a quantified reference interval or statistical distribution model defined for the key characteristic parameters in the list; and rules or association models for describing the co-variation relationship between the characteristic parameters of different physical fields under the seepage evolution pattern.

4. The real-time evolution prediction and assessment method for the risk of a sluice according to claim 3, wherein The method further includes the step of pre-constructing the seepage evolution pattern records, and the step includes at least one of the following: establishing a multi-physical field coupling mathematical model capable of reflecting the specific geological structure of the target sluice project, and conducting multi-condition dynamic simulations on a preset typical seepage evolution path, recording the whole-process multi-physical field response data of virtual sensors, and inducing and extracting the seepage evolution pattern records therefrom; or, by implementing a physical model test or in-situ controlled test capable of simulating the specific seepage failure process of the target sluice, deploying a sensor array in the test model and synchronously collecting multi-physical field data, and correlating the response characteristics observed in the test with the actually occurring seepage evolution state to calibrate and generate the seepage evolution pattern records.

5. The real-time evolution prediction and evaluation method for the risk of a sluice according to claim 4, characterized in that The step of matching to identify the collaborative anomaly pattern representing the current seepage state includes: Converting the rules or association models of the co-variation relationship defined in the seepage evolution pattern records into a set of executable logical judgment rules; Calculating the membership degree of the current eigenvalue in the multi-dimensional spatio-temporal feature sequence to the logical judgment rules; Adopting a weighted rule and fuzzy logic reasoning, comprehensively considering the activation intensity of all logical judgment rules and their preset weights, and calculating the matching confidence score between the current monitoring area and each seepage evolution pattern record to identify the collaborative anomaly pattern.

6. The real-time evolution prediction and evaluation method for the risk of a sluice as claimed in claim 5, characterized in that After the step of matching to identify the collaborative anomaly pattern characterizing the current seepage state, the method further includes: regarding all monitoring units in the target area inside the foundation as nodes in graph theory, and in real time constructing and dynamically updating a seepage activity network characterizing the spatial distribution and association state of the dominant seepage activity area according to the matching confidence scores of each monitoring unit; and adopting a graph theory analysis algorithm to quantitatively analyze the topological structure of the seepage activity network and its evolution characteristics over time.

7. The real-time evolution prediction and evaluation method for the risk of a sluice as described in claim 6, characterized in that, The step of performing real-time assessment of the sluice risk includes: comprehensively evaluating the following criteria: whether the risk state of local monitoring units continuously remains or rapidly enters a preset dangerous precursor level, whether the global topological structure of the seepage activity network reaches or approaches a critical penetration threshold, and the proximity of the calculated local hydraulic gradient or energy dissipation index to the physical instability criterion; wherein, the risk state assessment dynamically determines the risk assessment result by performing weighted scoring on the matching confidence scores of the collaborative anomaly patterns of local monitoring units, the connectivity, penetration or expansion rate indexes of the dominant seepage activity network, and the comparison results of the local hydraulic gradient or energy dissipation index with the classical soil mechanics instability criterion.

8. The real-time evolution prediction and assessment method for the risk of a sluice as claimed in claim 1, characterized in that, The method further includes: before or during the step of collecting the real-time multi-physical field data, actively emitting a controllable elastic wave signal through active excitation and auxiliary observation, and receiving the signal by the in-situ multi-parameter collaborative sensing network to enhance the detection ability of the dynamic changes of the foundation medium structure.

9. The real-time evolution prediction and evaluation method for the risk of a sluice as claimed in claim 1, wherein The method further includes: based on the real-time multi-physical field data and the physical model constraints in the knowledge base, through multi-physical field joint inversion, in real time reconstructing the three-dimensional effective hydraulic conductivity characteristic field and the topological structure of the dominant seepage network inside the foundation; and presenting the three-dimensional effective hydraulic conductivity characteristic field and the topological structure of the dominant seepage network through a three-dimensional dynamic visualization interface.

Images