A method and system for formulating spatiotemporal joint early warning indicators for concrete dam deformation
By collecting engineering data of concrete dams and building 3D digital terrain models, combining structural mechanics and fluid mechanics simulations, potential deformation hot spots are identified and real-time monitoring is carried out, and timeliness and scientific problems in traditional methods are solved, achieving high-precision Concrete dam deformation warning and monitoring.
Patent Information
- Application Number
- CN202410727069.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-06
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-06-06
AI Technical Summary
The method for formulating traditional concrete dam deformation joint warning indicators lacks timeliness and scientificity, and cannot detect potential safety hazards in a timely manner. It mainly relies on engineering experience and expert judgment, resulting in low credibility and operability of warning indicators.
By collecting engineering data of concrete dams, building a 3D digital terrain model, combining structural mechanics and fluid mechanics to simulate the hydrological-dam body deformation dynamic model, identify potential deformation hotspots, optimize monitoring point layout, monitor and analyze deformation trends in real time, and perform abnormalities Detection and finite element mechanical response solution, extract key impact indicators, evaluate risks and issue early warnings.
It realizes high-precision and time-efficient deformation monitoring and early warning of concrete dams, improves monitoring efficiency and targetedness, enhances the credibility and operability of early warning indicators, and ensures the safe operation of the dam body.
Smart Images

Figure CN118536200B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of water conservancy engineering, and in particular to a method and system for formulating a spatiotemporal combined early warning index for concrete dam deformation. Background Art
[0002] As an important infrastructure in water conservancy projects, the safe operation of concrete dams is crucial to regional economic and social development. However, concrete dams will undergo various deformations during long-term operation, such as temperature deformation, seepage deformation, foundation settlement, etc. If these deformations exceed the safety range, they will pose a serious threat to the dam structure. Therefore, it is very necessary to monitor the deformation of concrete dams and warn of deformation trends. However, the traditional method of formulating joint spatiotemporal early warning indicators for concrete dam deformation often obtains data through periodic inspections, which cannot detect potential safety hazards in a timely manner and lacks timeliness; at the same time, the indicators are mainly based on engineering experience or expert judgment, lack scientific quantitative analysis and data support, which is easy to cause subjectivity and one-sidedness, reducing the credibility and operability of the early warning indicators. Summary of the invention
[0003] Based on this, the present invention provides a method and system for formulating a spatiotemporal joint early warning indicator for concrete dam deformation to solve at least one of the above technical problems.
[0004] To achieve the above purpose, a method for formulating a spatiotemporal joint early warning index for concrete dam deformation is provided, comprising the following steps:
[0005] Step S1: collecting dam body engineering data of the concrete dam to generate dam body engineering data; constructing a 3D digital terrain model based on the dam body engineering data, optimizing the dynamic model, and generating a hydrological-dam body deformation dynamic model; performing hydrological-dynamic numerical simulation through the hydrological-dam body deformation dynamic model to generate hydrological-dynamic action point data;
[0006] Step S2: Identify the dam body boundary contour according to the dam body engineering data to generate dam body boundary contour data; process the dam body boundary contour data for potential deformation hotspots through the hydrodynamic action point data to generate potential deformation hotspot data; construct a dam body observation network according to the potential deformation hotspot data, and perform real-time dam body deformation monitoring to generate real-time dam body deformation monitoring data;
[0007] Step S3: extracting trend components according to the real-time dam deformation monitoring data to generate real-time dam trend component data; judging the real-time dam deformation mode according to the real-time dam trend component data to generate real-time dam deformation mode data; processing abnormal detection events according to the real-time dam deformation mode data to generate abnormal detection event data;
[0008] Step S4: screening dam deformation influencing events according to the abnormal detection event data to obtain dam deformation influencing event data; performing geographic space fitting processing on the dam deformation influencing event data to generate abnormal event geographic fitting data; performing spatiotemporal clustering processing on the abnormal event geographic fitting data to generate abnormal event regional data; performing key influencing indicator processing on the abnormal event regional data to generate spatiotemporal joint early warning indicator data;
[0009] Step S5: using the abnormal event area data to solve the finite element mechanical response of the hydrological-dam deformation dynamic model to generate dam body response force data; using the dam body response force data to iteratively process the deformation stress of the concrete dam to generate predicted dam body deformation trend data;
[0010] Step S6: Perform risk weight factor analysis on the spatiotemporal joint early warning index data to generate index weight factor data; perform risk factor scoring processing on the predicted dam deformation trend data using the index weight factor data to generate risk factor scoring data; perform early warning level classification based on the risk factor scoring data to generate dam early warning data.
[0011] The present invention can obtain high-precision and high-resolution dam engineering data, such as dam body geometric dimensions, material parameters, construction quality, etc., through on-site collection. The collected dam engineering data is used to construct a three-dimensional digital model of the concrete dam and its surrounding terrain using a geographic information system (GIS) and three-dimensional modeling software. This model can not only intuitively display the spatial structure of the dam body, but also intuitively display the geometric form, spatial position relationship and other information of the dam body. Based on the 3D digital terrain model, the dynamic response characteristics of the concrete dam are analyzed and simulated in combination with structural mechanics and fluid mechanics, and hydrological factors such as water level, flow rate, and flow velocity are incorporated into the model, which can more comprehensively reflect the influence of hydrological dynamics on the deformation of the dam body. The optimized dynamic model is combined with the hydrological model, and the water flow dynamics, water pressure distribution and its effect on the dam body are considered. Through simulation, the key action point where the hydrological dynamics has the greatest impact on the dam body can be determined. Through image processing and pattern recognition technology, the boundary contour of the dam body in the image or point cloud data can be automatically identified to avoid the subjectivity and errors caused by manual identification. Combined with the data of hydrodynamic action points, the dam areas that are greatly affected by hydrodynamics can be identified. These areas are often potential deformation hotspots and need to be paid special attention. According to the potential deformation hotspot data, the layout of monitoring points can be optimized to ensure that the monitoring points can effectively capture the deformation of the dam body, concentrate limited monitoring resources in key areas, and improve monitoring efficiency and pertinence. By extracting trend components, the influence of short-term factors such as temperature changes and wind on deformation monitoring data can be eliminated, and the long-term deformation trend of the dam body can be more clearly reflected. By analyzing trend component data, the deformation mode of the dam body can be identified, such as overall tilt, local settlement, etc. By monitoring and analyzing the real-time dam deformation mode data, combined with the preset safety thresholds and rules, abnormal detection events can be automatically identified and processed. Through screening, events related to dam deformation can be extracted from all abnormal detection events, eliminating interference from other factors, and focusing on key events that have potential impact on dam safety. Through geographic spatial fitting, dam deformation impact events can be visualized on the map to intuitively understand the spatial distribution characteristics of events. Through spatiotemporal clustering, we can identify groups of events that are clustered in space and time. These areas are areas with a higher risk of dam deformation and need to be paid special attention to. By extracting key influencing indicators, such as the frequency of events, the scope of impact, and the degree of deformation, we can quantify the impact of events on the safety of the dam. By solving the finite element mechanical response of the hydrological-dam deformation dynamic model based on the data of the abnormal event area, we can simulate the mechanical response of the dam under specific abnormal events, including force distribution and stress concentration. The deformation and stress of the concrete dam are iteratively processed to predict the deformation trend of the dam in the future. Identify the factors that have a significant impact on the stability and safety of the dam, and assign weights to them to more accurately consider the impact of various risk factors.According to the weight factors, the predicted dam deformation trend data is scored and processed to quantify the contribution of different factors to the risk, so as to more accurately assess the risk status of the dam. According to the risk score data, the risk level of the dam is divided into different warning levels, and warning information is promptly issued to relevant departments and personnel. Therefore, a method for formulating a spatiotemporal joint warning index for concrete dam deformation of the present invention obtains the engineering data of the concrete dam, constructs a 3D digital simulation model, uses historical hydrological data and deformation data, and mines the influence relationship therein, thereby performing hydrological simulation and determining the action point of hydrological deformation of the dam. Combined with the hydrodynamic action point data, the dam area that is greatly affected by the hydrodynamics can be identified, thereby optimizing the layout of the monitoring points, ensuring that the monitoring points can effectively capture the deformation of the dam, and improving the monitoring efficiency and pertinence. According to the real-time monitoring data, a spatiotemporal joint clustering analysis is performed to extract the dam deformation mode and the abnormal monitoring data, and a joint warning index is formulated according to the dam deformation mode, and the dam trend behavior is predicted using the abnormal monitoring data, and the risk factors are combined for scoring, so as to accurately realize the deformation warning of the concrete dam.
[0012] Preferably, step S1 comprises the following steps:
[0013] Step S11: collecting dam body engineering data of the concrete dam, and performing data cleaning processing to generate dam body engineering data, wherein the dam body engineering data includes dam body surface graphic data, dam body historical deformation data, dam body historical hydrological data and dam body geological structure data;
[0014] Step S12: constructing a 3D digital terrain model based on the dam body surface graphic data, and performing dam body gridding processing to generate a dam body surface terrain model;
[0015] Step S13: Analyze the hydrological deformation characteristics according to the historical hydrological data and historical deformation data of the dam body to generate hydrological-deformation relationship data;
[0016] Step S14: optimizing the hydrological-deformation dynamic model of the dam surface terrain model using the hydrological-deformation relationship data to generate a hydrological-dam deformation dynamic model;
[0017] Step S15: Perform hydrological dynamic numerical simulation through the hydrological-dam deformation dynamic model, and process the water body action point to generate hydrological dynamic action point data.
[0018] The present invention collects dam engineering data for concrete dams, and the collection and cleaning of these data can obtain accurate and reliable dam engineering information. A 3D digital terrain model is constructed according to the dam surface graphic data. By constructing the digital terrain model, the terrain characteristics of the dam body, including its shape, elevation and other information, can be accurately described, and the model can be made more refined through grid processing. The hydrological deformation characteristics are analyzed according to the historical hydrological data of the dam body and the historical deformation data of the dam body, so as to understand the influence of hydrological factors on the deformation of the dam body. The hydrological-deformation dynamic model of the dam surface terrain model is optimized by using the hydrological-deformation relationship data. By optimizing the terrain model and combining the hydrological-deformation relationship data, a more accurate hydrological-dam body deformation dynamic model is established, so as to better simulate the deformation of the dam body under hydrological conditions. The hydrological-dam body deformation dynamic model is used to perform hydrological dynamic numerical simulation, and the dam body response under different hydrological conditions can be simulated through numerical simulation, so as to predict potential risks. At the same time, processing the water body action point can accurately determine the position and strength of the hydraulic action point.
[0019] Preferably, step S13 comprises the following steps:
[0020] Step S131: performing time series decomposition processing on the historical hydrological data of the dam body and the historical deformation data of the dam body to obtain sequence dam hydrological data and sequence dam deformation data respectively;
[0021] Step S132: extracting nonlinear features from the sequence dam body hydrological data to generate hydrological dynamic characteristic data;
[0022] Step S133: performing spatial entropy correlation analysis based on the hydrological dynamic characteristic data to obtain hydrological spatial entropy correlation data;
[0023] Step S134: calculating the deformation rate of the sequence dam body deformation data to generate deformation rate data;
[0024] Step S135: performing deformation pattern clustering processing on the sequence dam body deformation data through the deformation rate data to generate deformation pattern clustering data;
[0025] Step S136: using the hydrological spatial entropy related data to perform multivariate time series fitting processing on the deformation pattern clustering data to obtain hydrological-deformation fitting data;
[0026] Step S137: Perform principal component analysis based on the hydrological-deformation fitting data, and perform coupled regression processing to generate hydrological-deformation spatial relationship data.
[0027] The present invention performs time series decomposition processing on the historical hydrological data and historical deformation data of the dam body, and can decompose the original data according to the time series, so as to better understand the periodic characteristics of hydrology and deformation. The nonlinear characteristics of the sequence dam body hydrological data are extracted, and the regularity of hydrological changes can be described more comprehensively by extracting the nonlinear characteristics of the hydrological data. The spatial entropy correlation analysis is performed according to the hydrological dynamic characteristic data, and the spatial distribution law of hydrology can be revealed. The deformation rate is calculated for the sequence dam body deformation data, and the speed and trend of the dam body deformation can be understood by calculating the deformation rate. The deformation pattern clustering processing is performed on the sequence dam body deformation data through the deformation rate data, and the deformation data can be classified according to different patterns, thereby revealing different characteristics of the dam body deformation. The deformation pattern clustering data is subjected to multivariate time series fitting processing using the hydrological spatial entropy related data, and the hydrological and deformation data are associated to reveal the internal connection between them. The principal component analysis is performed according to the hydrological-deformation fitting data, and the correlation law between hydrology and deformation is further revealed through the principal component analysis and coupled regression processing, which provides a scientific basis for the monitoring and prediction of the dam body project.
[0028] Preferably, step S2 comprises the following steps:
[0029] Step S21: identifying the boundary contour of the dam body according to the dam body surface terrain model, and generating dam body boundary contour data;
[0030] Step S22: extracting key geological features according to the dam body geological structure data to generate key geological feature data;
[0031] Step S23: using the dam surface terrain model to perform geological structure analysis processing on the key geological feature data to generate geological structure analysis array data;
[0032] Step S24: processing the dam body boundary contour data for potential deformation hotspots of the dam body through the geological structure analysis array data and the hydrological dynamic action point data to generate potential deformation hotspot data;
[0033] Step S25: Optimize the observation points according to the potential deformation hotspot data and deploy adaptive observation equipment to build a dam body observation network;
[0034] Step S26: Using the dam body observation network to perform real-time dam body deformation monitoring on the concrete dam, and generate real-time dam body deformation monitoring data.
[0035] The present invention identifies the boundary contour of the dam body according to the dam body surface terrain model, determines the boundary of the dam body, and helps to accurately locate the dam body structure. Extracting key geological features according to the dam body geological structure data can deeply understand the geological structure of the dam body. The dam body surface terrain model is used to perform geological structure analysis processing on key geological feature data. By combining the terrain model and the geological feature data, the geological structure of the dam body can be analyzed more comprehensively. The dam body boundary contour data is processed by the geological structure analysis array data and the hydrodynamic action point data to process the potential deformation hotspots of the dam body, and the area with deformation risk can be identified. According to the potential deformation hotspot data, the observation point is selected, and the adaptive observation equipment is deployed to build a dam body observation network. Through site selection and equipment deployment, the potential deformation hotspot area can be covered to the greatest extent, and a comprehensive dam body monitoring network can be built to realize real-time monitoring and early warning of dam body deformation. The dam body observation network is used to perform real-time dam body deformation monitoring on the concrete dam, thereby improving the timeliness of the concrete dam early warning.
[0036] Preferably, step S24 includes the following steps:
[0037] Step S241: Calculate the surface curvature according to the dam body boundary contour data to generate dam body surface curve data;
[0038] Step S242: performing dam body slope analysis according to the dam body surface curve data to generate dam body slope area data; performing flat area analysis according to the dam body surface curve data to generate dam body flat area data;
[0039] Step S243: marking the dam body corners of the dam body flat area data by using the dam body slope area data, thereby obtaining the dam body corner data; marking the dam body corner data as easily deformable area data;
[0040] Step S244: performing geological weak point processing on the dam body boundary contour data through the geological structure analysis array data to generate geological weak point data;
[0041] Step S245: performing surface action processing on the easily deformable area data through the geological weak point data to generate high deformation hot spot area data;
[0042] Step S246: supplementing the high deformation hotspot area data with potential deformation hotspots through the hydrodynamic action point data to obtain potential deformation hotspot data.
[0043] The present invention calculates the surface curvature according to the dam body boundary contour data, and the generation of the surface curvature data can reveal the curvature change of the dam body surface. The dam body slope analysis is performed according to the dam body surface curve data, and the flat area analysis is performed according to the dam body surface curve data. The analysis of the dam body slope area and the flat area can identify the different terrain features on the dam body surface, and then accurately locate the easily deformed area. The dam body corners are marked on the dam body flat area data through the dam body slope area data, and the marking of the corner data and the identification of the easily deformed area are helpful to determine the structural characteristics of the dam body and the potential deformation risk area. The dam body boundary contour data is processed by the geological structure analysis array data, and the geological weak points around the dam body boundary can be accurately identified. The easily deformed area data is processed by the geological weak point data. The surface action processing of the easily deformed area can further determine the high deformation hotspot area. The high deformation hotspot area data is supplemented with potential deformation hotspots through the hydrodynamic action point data, and the influence of hydrodynamic factors on the high deformation hotspot area can be comprehensively considered, and the accuracy and reliability of the potential deformation hotspot data can be further improved.
[0044] Preferably, step S3 comprises the following steps:
[0045] Step S31: performing multi-scale data preprocessing according to the real-time dam deformation monitoring data to generate multi-scale dam monitoring data, wherein the multi-scale dam monitoring data includes real-time water pressure data, dam slope inclinometer readings, thermocouple temperature readings and precipitation data;
[0046] Step S32: performing data cutting processing on the multi-scale dam body monitoring data through a preset time window, and performing wavelet analysis processing to generate dam body monitoring time series data;
[0047] Step S33: extracting trend components according to the dam body monitoring time series data to generate real-time dam body trend component data;
[0048] Step S34: performing seasonal component processing on the dam body monitoring time series data based on the real-time dam body trend component data to generate real-time dam body seasonal component data;
[0049] Step S35: using the dam body monitoring time series data to calculate residual data items for the real-time dam body trend component data and the real-time dam body seasonal component data, to generate real-time dam body residual data;
[0050] Step S36: using the preset standard dam body monitoring threshold data to perform abnormal component detection on the real-time dam body seasonal component data and the real-time dam body trend component data, and perform periodic pattern judgment to generate real-time dam body deformation pattern data, wherein the real-time dam body deformation pattern data includes periodic dam body deformation data and non-periodic dam body deformation data;
[0051] Step S37: when the real-time dam body deformation mode data is periodic dam body deformation data, performing detrending detection processing on the periodic dam body deformation data to generate detrending detection data;
[0052] Step S38: when the real-time dam body deformation mode data is non-periodic dam body deformation data, abnormal data points are marked on the non-periodic dam body deformation data to generate abnormal data point data;
[0053] Step S39: Perform abnormal detection event processing on the real-time dam residual data by removing the trend detection data and the abnormal data point data to generate abnormal detection event data.
[0054] The present invention performs multi-scale data preprocessing based on real-time dam deformation monitoring data, and integrates monitoring data of different scales to obtain more comprehensive and accurate dam status information. The dam monitoring data is processed in a time series to reveal the time and frequency characteristics in the data. Trend component extraction based on the dam monitoring time series data helps to capture the long-term trend in the dam monitoring data, thereby separating the changes related to the trend. The extraction of seasonal components helps to identify the periodic change pattern in the data, so as to better understand the seasonal change law of the dam monitoring data. The residual data items are calculated for the real-time dam trend component data and the real-time dam seasonal component data using the dam monitoring time series data. The residual data reflects the random fluctuations in the monitoring data other than the trend and seasonal components, and is a key indicator for identifying abnormal changes. The real-time dam seasonal component data and the real-time dam trend component data are detected for abnormal components using the preset standard dam monitoring threshold data, which helps to identify the abnormal change pattern in the monitoring data and better perform abnormal detection and event processing. When the real-time dam deformation pattern data is periodic dam deformation data, detrending detection helps to eliminate the trend component in the periodic change so as to more accurately identify abnormal changes. When the real-time dam deformation pattern data is non-periodic dam deformation data, the identification of abnormal data points helps to accurately locate and identify abnormal events in the monitoring data. By detrending the detection data and the abnormal data point data, the real-time dam residual data is processed for abnormal detection events, which can comprehensively identify and process abnormal events in the monitoring data.
[0055] Preferably, step S4 comprises the following steps:
[0056] Step S41: performing abnormal event impact assessment processing according to the abnormal detection event data to generate abnormal event assessment data;
[0057] Step S42: screening dam body deformation influencing events according to the abnormal event assessment data to obtain dam body deformation influencing event data;
[0058] Step S43: performing geographic orientation fitting processing on the concrete dam through the dam body deformation impact event data to generate abnormal event geographic fitting data;
[0059] Step S44: performing space-time distance calculation on the geographic fitting data of the abnormal event to generate space-time distance data;
[0060] Step S45: using a preset spatiotemporal clustering algorithm and spatiotemporal distance data to perform spatiotemporal clustering processing on the geographic fitting data of abnormal events, and perform abnormal event area identification to generate abnormal event area data;
[0061] Step S46: Process the key impact indexes of the potential abnormal event area data through the dam deformation impact event data to generate spatiotemporal joint early warning index data.
[0062] The present invention performs abnormal event impact assessment processing based on abnormal detection event data, and can systematically assess the degree of impact of abnormal events on the dam body, including the scope of damage, degree of impact, etc. By screening abnormal events, events that have a substantial impact on the deformation of the dam body can be determined. By performing geographic orientation fitting processing on the concrete dam through the dam body deformation impact event data, the abnormal event can be associated with the specific geographical location, and the occurrence location and distribution law of the abnormal event can be better understood. The spatial and temporal distance calculation is performed on the abnormal event geographic fitting data, and the calculation of the spatial and temporal distance data can quantify the degree of spatial and temporal correlation between abnormal events. Through spatial and temporal clustering processing, the spatial and temporal clustering areas of abnormal events can be identified. By performing key impact indicator processing on the potential abnormal event area data through the dam body deformation impact event data, the key factors and indicators affecting the abnormal event area can be identified.
[0063] Preferably, step S5 comprises the following steps:
[0064] Step S51: constructing a finite element relationship model for the dam body segmentation unit data through the abnormal event area data to obtain a finite element relationship model of the dam body;
[0065] Step S52: solving the mechanical response of the finite element relationship model of the dam body to generate dam body response force data;
[0066] Step S53: using the dam body response force data to perform spatial correlation processing on the hydrological-dam body deformation dynamic model to generate spatial deformation stress data;
[0067] Step S54: constructing a chaotic attractor according to the spatial deformation stress data, and defining the attractor characteristics to generate attractor characteristic data;
[0068] Step S55: Perform iterative calculation of deformation stress of the concrete dam through attractor characteristic data, and perform chaos simulation prediction to generate predicted dam deformation trend data.
[0069] The present invention constructs a finite element relationship model for the data of the dam body segmentation unit through the data of the abnormal event area, establishes a finite element model that describes the mechanical behavior of the dam body, and can simulate the stress and deformation of the dam body when it is subjected to external forces. The mechanical response of the finite element relationship model of the dam body is solved, and the stress distribution of each segmentation unit of the dam body under force can be calculated through the mechanical response solution, so as to fully understand the mechanical response characteristics of the dam body structure. The hydrological-dam body deformation dynamic model is spatially correlated using the dam body response force data, and the mechanical response of the dam body is associated with the hydrological factors, revealing the influence of hydrology on the spatial distribution of dam body deformation. Through the construction and feature definition of chaotic attractors, the chaotic characteristics in the dam body deformation process can be identified, so as to better understand the complex dynamic behavior of the dam body. The deformation stress of the concrete dam is iteratively calculated through the attractor feature data, and the deformation trend of the dam body can be predicted based on the attractor feature data, which provides an important prediction basis for the safety assessment and management of the dam body project, and helps to take measures to prevent potential risks in a timely manner.
[0070] Preferably, step S6 comprises the following steps:
[0071] Step S61: Perform risk weight factor analysis on the spatiotemporal joint warning indicator data to generate indicator weight factor data;
[0072] Step S62: Formulate deformation-scoring fuzzy rules according to the indicator weight factor data to generate deformation-scoring fuzzy rule data;
[0073] Step S63: constructing a rule reasoning model through deformation-scoring fuzzy rule data;
[0074] Step S64: using the rule inference model to process the predicted dam deformation trend data with risk factor scores to generate risk factor score data;
[0075] Step S65: Calculate the warning value according to the risk factor score data, classify the warning level, and generate dam body warning data.
[0076] The present invention performs risk weight factor analysis on the spatiotemporal joint early warning index data, and identifies the key factors affecting the dam body risk by performing weight analysis on the early warning index. Deformation-scoring fuzzy rules are formulated according to the index weight factor data. By formulating fuzzy rules, the index weight factors can be combined with the dam body deformation situation, and a fuzzy rule framework for evaluating the dam body risk is established. A rule reasoning model is constructed through deformation-scoring fuzzy rule data. The rule reasoning model can perform reasoning analysis on the dam body deformation trend data according to the set fuzzy rules, thereby evaluating the risk level of the dam body and providing a scientific basis for early warning decision-making. By scoring the deformation trend data, the degree of influence of risk factors on the dam body deformation can be quantified. The threshold of the early warning value is determined according to the scoring data, and the early warning status of the dam body is divided into different levels, providing timely early warning information for relevant departments and personnel, so as to take corresponding measures to deal with potential risks.
[0077] The present invention also provides a system for formulating a spatiotemporal joint early warning index for deformation of a concrete dam, which executes the method for formulating a spatiotemporal joint early warning index for deformation of a concrete dam as described above. The system for formulating a spatiotemporal joint early warning index for deformation of a concrete dam comprises:
[0078] The dam hydrological simulation module is used to collect dam engineering data for concrete dams and generate dam engineering data; construct a 3D digital terrain model based on the dam engineering data, optimize the dynamic model, and generate a hydrological-dam deformation dynamic model; perform hydrological dynamic numerical simulation through the hydrological-dam deformation dynamic model to generate hydrological dynamic action point data;
[0079] The dam deformation monitoring module is used to identify the dam boundary contour according to the dam engineering data and generate the dam boundary contour data; to process the dam boundary contour data for potential deformation hotspots through the hydrological dynamic action point data and generate potential deformation hotspot data; to construct a dam observation network according to the potential deformation hotspot data and perform real-time dam deformation monitoring to generate real-time dam deformation monitoring data;
[0080] The dam body anomaly detection module is used to extract trend components based on real-time dam body deformation monitoring data to generate real-time dam body trend component data; to judge the real-time dam body deformation mode based on the real-time dam body trend component data to generate real-time dam body deformation mode data; to process anomaly detection events based on the real-time dam body deformation mode data to generate anomaly detection event data;
[0081] The early warning indicator formulation module is used to screen dam deformation impact events based on abnormal detection event data to obtain dam deformation impact event data; perform geographic space fitting processing on dam deformation impact event data to generate abnormal event geographic fitting data; perform spatiotemporal clustering processing based on abnormal event geographic fitting data to generate abnormal event area data; perform key impact indicator processing based on abnormal event area data to generate spatiotemporal joint early warning indicator data;
[0082] The deformation trend prediction module is used to solve the finite element mechanical response of the hydrological-dam deformation dynamic model through the abnormal event area data to generate the dam body response force data; the deformation stress of the concrete dam is iteratively processed through the dam body response force data to generate the predicted dam body deformation trend data;
[0083] The dam body early warning assessment module is used to perform risk weight factor analysis on the spatiotemporal joint early warning index data to generate index weight factor data; use the index weight factor data to perform risk factor scoring on the predicted dam body deformation trend data to generate risk factor scoring data; classify the early warning level according to the risk factor scoring data to generate dam body early warning data. BRIEF DESCRIPTION OF THE DRAWINGS
[0084] Figure 1 The present invention is a schematic diagram of the steps of a method and system for formulating spatiotemporal joint early warning indicators for concrete dam deformation.
[0085] Figure 2 for Figure 1 Detailed implementation steps of step S5;
[0086] Figure 3 for Figure 1 Detailed implementation steps of step S6 in FIG.
[0087] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION
[0088] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.
[0089] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.
[0090] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.
[0091] To achieve this, please refer to Figures 1 to 3 The present invention provides a method for formulating a spatiotemporal joint early warning index for concrete dam deformation, comprising the following steps:
[0092] Step S1: collecting dam body engineering data of the concrete dam to generate dam body engineering data; constructing a 3D digital terrain model based on the dam body engineering data, optimizing the dynamic model, and generating a hydrological-dam body deformation dynamic model; performing hydrological-dynamic numerical simulation through the hydrological-dam body deformation dynamic model to generate hydrological-dynamic action point data;
[0093] Step S2: Identify the dam body boundary contour according to the dam body engineering data to generate dam body boundary contour data; process the dam body boundary contour data for potential deformation hotspots through the hydrodynamic action point data to generate potential deformation hotspot data; construct a dam body observation network according to the potential deformation hotspot data, and perform real-time dam body deformation monitoring to generate real-time dam body deformation monitoring data;
[0094] Step S3: extracting trend components according to the real-time dam deformation monitoring data to generate real-time dam trend component data; judging the real-time dam deformation mode according to the real-time dam trend component data to generate real-time dam deformation mode data; processing abnormal detection events according to the real-time dam deformation mode data to generate abnormal detection event data;
[0095] Step S4: screening dam deformation influencing events according to the abnormal detection event data to obtain dam deformation influencing event data; performing geographic space fitting processing on the dam deformation influencing event data to generate abnormal event geographic fitting data; performing spatiotemporal clustering processing on the abnormal event geographic fitting data to generate abnormal event regional data; performing key influencing indicator processing on the abnormal event regional data to generate spatiotemporal joint early warning indicator data;
[0096] Step S5: using the abnormal event area data to solve the finite element mechanical response of the hydrological-dam deformation dynamic model to generate dam body response force data; using the dam body response force data to iteratively process the deformation stress of the concrete dam to generate predicted dam body deformation trend data;
[0097] Step S6: Perform risk weight factor analysis on the spatiotemporal joint early warning index data to generate index weight factor data; perform risk factor scoring processing on the predicted dam deformation trend data using the index weight factor data to generate risk factor scoring data; perform early warning level classification based on the risk factor scoring data to generate dam early warning data.
[0098] The present invention can obtain high-precision and high-resolution dam engineering data, such as dam body geometric dimensions, material parameters, construction quality, etc., through on-site collection. The collected dam engineering data is used to construct a three-dimensional digital model of the concrete dam and its surrounding terrain using a geographic information system (GIS) and three-dimensional modeling software. This model can not only intuitively display the spatial structure of the dam body, but also intuitively display the geometric form, spatial position relationship and other information of the dam body. Based on the 3D digital terrain model, the dynamic response characteristics of the concrete dam are analyzed and simulated in combination with structural mechanics and fluid mechanics, and hydrological factors such as water level, flow rate, and flow velocity are incorporated into the model, which can more comprehensively reflect the influence of hydrological dynamics on the deformation of the dam body. The optimized dynamic model is combined with the hydrological model, and the water flow dynamics, water pressure distribution and its effect on the dam body are considered. Through simulation, the key action point where the hydrological dynamics has the greatest impact on the dam body can be determined. Through image processing and pattern recognition technology, the boundary contour of the dam body in the image or point cloud data can be automatically identified to avoid the subjectivity and errors caused by manual identification. Combined with the data of hydrodynamic action points, the dam areas that are greatly affected by hydrodynamics can be identified. These areas are often potential deformation hotspots and need to be paid special attention. According to the potential deformation hotspot data, the layout of monitoring points can be optimized to ensure that the monitoring points can effectively capture the deformation of the dam body, concentrate limited monitoring resources in key areas, and improve monitoring efficiency and pertinence. By extracting trend components, the influence of short-term factors such as temperature changes and wind on deformation monitoring data can be eliminated, and the long-term deformation trend of the dam body can be more clearly reflected. By analyzing trend component data, the deformation mode of the dam body can be identified, such as overall tilt, local settlement, etc. By monitoring and analyzing the real-time dam deformation mode data, combined with the preset safety thresholds and rules, abnormal detection events can be automatically identified and processed. Through screening, events related to dam deformation can be extracted from all abnormal detection events, eliminating interference from other factors, and focusing on key events that have potential impact on dam safety. Through geographic spatial fitting, dam deformation impact events can be visualized on the map to intuitively understand the spatial distribution characteristics of events. Through spatiotemporal clustering, we can identify groups of events that are clustered in space and time. These areas are areas with a higher risk of dam deformation and need to be paid special attention to. By extracting key influencing indicators, such as the frequency of events, the scope of impact, and the degree of deformation, we can quantify the impact of events on the safety of the dam. By solving the finite element mechanical response of the hydrological-dam deformation dynamic model based on the data of the abnormal event area, we can simulate the mechanical response of the dam under specific abnormal events, including force distribution and stress concentration. The deformation and stress of the concrete dam are iteratively processed to predict the deformation trend of the dam in the future. Identify the factors that have a significant impact on the stability and safety of the dam, and assign weights to them to more accurately consider the impact of various risk factors.According to the weight factors, the predicted dam deformation trend data is scored and processed to quantify the contribution of different factors to the risk, so as to more accurately assess the risk status of the dam. According to the risk score data, the risk level of the dam is divided into different warning levels, and warning information is promptly issued to relevant departments and personnel. Therefore, a method for formulating a spatiotemporal joint warning index for concrete dam deformation of the present invention obtains the engineering data of the concrete dam, constructs a 3D digital simulation model, uses historical hydrological data and deformation data, and mines the influence relationship therein, thereby performing hydrological simulation and determining the action point of hydrological deformation of the dam. Combined with the hydrodynamic action point data, the dam area that is greatly affected by the hydrodynamics can be identified, thereby optimizing the layout of the monitoring points, ensuring that the monitoring points can effectively capture the deformation of the dam, and improving the monitoring efficiency and pertinence. According to the real-time monitoring data, a spatiotemporal joint clustering analysis is performed to extract the dam deformation mode and the abnormal monitoring data, and a joint warning index is formulated according to the dam deformation mode, and the dam trend behavior is predicted using the abnormal monitoring data, and the risk factors are combined for scoring, so as to accurately realize the deformation warning of the concrete dam.
[0099] In the embodiment of the present invention, reference Figure 1 The above is a schematic diagram of the steps of a method for formulating a spatiotemporal joint early warning indicator for deformation of a concrete dam according to the present invention. In this embodiment, the method for formulating a spatiotemporal joint early warning indicator for deformation of a concrete dam comprises the following steps:
[0100] Step S1: collecting dam body engineering data of the concrete dam to generate dam body engineering data; constructing a 3D digital terrain model based on the dam body engineering data, optimizing the dynamic model, and generating a hydrological-dam body deformation dynamic model; performing hydrological-dynamic numerical simulation through the hydrological-dam body deformation dynamic model to generate hydrological-dynamic action point data;
[0101] In the embodiment of the present invention, instruments such as total stations and laser scanners are used to measure the concrete dam to obtain surface shape and geological structure data; at the same time, hydrological data such as historical rainfall and river flow, as well as historical deformation data such as displacement and strain of dam deformation monitoring points are collected. Clean and process the data to ensure accuracy and availability. Use GIS software to construct a 3D digital terrain model of the dam body to reflect the terrain characteristics. Analyze the relationship between hydrological data and dam deformation data, establish a hydrological-dam deformation dynamic model, and optimize it. Use computer simulation software to perform hydrological dynamic numerical simulation to determine the hydrological action point and force size, including water pressure, water flow force, etc.
[0102] Step S2: Identify the dam body boundary contour according to the dam body engineering data to generate dam body boundary contour data; process the dam body boundary contour data for potential deformation hotspots through the hydrodynamic action point data to generate potential deformation hotspot data; construct a dam body observation network according to the potential deformation hotspot data, and perform real-time dam body deformation monitoring to generate real-time dam body deformation monitoring data;
[0103] In an embodiment of the present invention, the dam body boundary contour is identified based on the terrain model of the concrete dam to generate boundary contour data; then, these boundary contours are analyzed in combination with the hydrodynamic action point data to identify potential deformation hot spots and generate corresponding hot spot data; based on the hot spot data, a dam body observation network is selected and established, and adaptive observation equipment is deployed; finally, the established observation network is used to perform real-time deformation monitoring of the concrete dam to obtain real-time dam body deformation monitoring data.
[0104] Step S3: extracting trend components according to the real-time dam deformation monitoring data to generate real-time dam trend component data; judging the real-time dam deformation mode according to the real-time dam trend component data to generate real-time dam deformation mode data; processing abnormal detection events according to the real-time dam deformation mode data to generate abnormal detection event data;
[0105] In an embodiment of the present invention, first, the monitoring data such as water pressure, slope inclination, temperature and precipitation are cleaned and corrected, and then integrated into multi-scale dam monitoring data. A time window is set, the data is cut at certain intervals, and wavelet analysis is performed to extract frequency domain and time domain features. The moving average method is used to extract long-term trend components, such as seasonal changes in water levels. The seasonal periodic components, such as annual water level fluctuations, are extracted by the periodic decomposition method. The trend component and the seasonal component are subtracted to obtain residual data to describe non-trend and non-periodic changes. A monitoring threshold is set, and a water pressure rise rate exceeding 0.5 meters / hour is judged to be abnormal. Abnormal component detection is performed on real-time trend and seasonal data, and periodic detection is performed on abnormal points. Detrending is performed on periodic abnormal data to obtain non-periodic abnormal data. Record abnormal detection events and generate abnormal detection event data.
[0106] Step S4: screening dam deformation influencing events according to the abnormal detection event data to obtain dam deformation influencing event data; performing geographic space fitting processing on the dam deformation influencing event data to generate abnormal event geographic fitting data; performing spatiotemporal clustering processing on the abnormal event geographic fitting data to generate abnormal event regional data; performing key influencing indicator processing on the abnormal event regional data to generate spatiotemporal joint early warning indicator data;
[0107] In an embodiment of the present invention, the nature, duration, intensity and other information of the abnormal detection event data are analyzed, and the impact on the safety of the dam body and the surrounding environment is evaluated according to the preset evaluation criteria. The abnormal events that affect the deformation of the dam body are screened, and the specific location of each event is determined according to the geographic information. The time-space distance is calculated to measure the correlation between events. The ST-DBSCAN algorithm is used for time-space clustering to obtain the abnormal event area. The key influencing indicators of each area are determined, such as displacement rate and stress distribution. The time-space joint early warning indicators are calculated to reflect the comprehensive impact of abnormal events on the deformation of the dam body.
[0108] Step S5: using the abnormal event area data to solve the finite element mechanical response of the hydrological-dam deformation dynamic model to generate dam body response force data; using the dam body response force data to iteratively process the deformation stress of the concrete dam to generate predicted dam body deformation trend data;
[0109] In an embodiment of the present invention, a finite element model of the dam body is established using data from the abnormal event area, and the relationship between the abnormal event area and the segmented units of the dam body, including the connection method and the degree of influence, is described. Preprocessing work includes defining material properties and loading conditions. An appropriate finite element solver, such as ABAQUS, is selected. According to the characteristics of the model and the solution requirements, the mechanical response is solved to calculate the mechanical response of each part of the dam body. Post-processing tools are used to extract response data of interest, such as stress and displacement. According to the water level changes, spatial deformation stress data is generated to describe the stress conditions of various parts of the dam body under different water level conditions. The spatial deformation stress data is reconstructed in phase space and estimated by chaotic attractors to extract key features. According to the attractor characteristic data, the deformation stress is iteratively calculated to obtain the deformation stress data of the dam body at different time points. Chaos theory is used for prediction to predict the deformation trend of the dam body in the future.
[0110] Step S6: Perform risk weight factor analysis on the spatiotemporal joint early warning index data to generate index weight factor data; perform risk factor scoring processing on the predicted dam deformation trend data using the index weight factor data to generate risk factor scoring data; perform early warning level classification based on the risk factor scoring data to generate dam early warning data.
[0111] In the embodiment of the present invention, the AHP method is used to determine the relative importance of each factor to the safety of the dam body according to the hierarchical structure of the factors related to the safety of the dam body, and to form the index weight factor data. According to the pre-set deformation-scoring fuzzy rules, combined with the changes in the monitoring indicators, the safety status of the dam body is evaluated. This includes judging the safety level of the dam body according to the changes in the monitoring indicators such as water pressure, temperature, displacement, etc., combined with their weight factors, so as to generate risk factor scoring data. Next, a fuzzy logic rule reasoning model is established, and the predicted dam deformation trend data is input into it. Through the reasoning and calculation of the model, the risk factor scoring data of the dam body is obtained, which reflects the degree of influence of the deformation trend on the safety of the dam body. Finally, according to the scoring data, the warning level of the dam body is classified, usually into three levels: high, medium and low, to form the dam body warning data.
[0112] Preferably, step S1 comprises the following steps:
[0113] Step S11: collecting dam body engineering data of the concrete dam, and performing data cleaning processing to generate dam body engineering data, wherein the dam body engineering data includes dam body surface graphic data, dam body historical deformation data, dam body historical hydrological data and dam body geological structure data;
[0114] Step S12: constructing a 3D digital terrain model based on the dam body surface graphic data, and performing dam body gridding processing to generate a dam body surface terrain model;
[0115] Step S13: Analyze the hydrological deformation characteristics according to the historical hydrological data and historical deformation data of the dam body to generate hydrological-deformation relationship data;
[0116] Step S14: optimizing the hydrological-deformation dynamic model of the dam surface terrain model using the hydrological-deformation relationship data to generate a hydrological-dam deformation dynamic model;
[0117] Step S15: Perform hydrological dynamic numerical simulation through the hydrological-dam deformation dynamic model, and process the water body action point to generate hydrological dynamic action point data.
[0118] In the embodiment of the present invention, a professional engineering team is dispatched to the concrete dam site, and a measuring instrument (such as a total station, a laser scanner, etc.) is used to perform detailed measurements on the dam body, obtain the surface shape data of the dam body, and record the geological structure of the dam body at the same time. The historical hydrological data of the concrete dam in the past, including information such as rainfall and river flow, are collected, and the historical deformation data of the dam body, such as the displacement and strain data of the dam body deformation monitoring points, are obtained. The collected data are cleaned and processed, including the unification of data formats, the elimination of abnormal data, etc., to ensure the accuracy and availability of the data. A 3D digital terrain model of the dam body is constructed based on the collected dam body surface graphic data using professional geographic information system (GIS) software to accurately reflect the terrain characteristics of the dam body. The relationship between historical hydrological data and dam body deformation data is analyzed using statistical analysis, regression analysis, and other methods to explore the influence of hydrological factors on dam body deformation. Based on the hydrological-deformation relationship data, a hydrological-dam body deformation dynamic model is constructed using a mathematical modeling method, and the relationship between hydrological conditions and dam body deformation is converted into a mathematical expression. Using computer simulation software, based on the established hydrological-dam deformation dynamic model, hydrological dynamic numerical simulation is carried out to simulate the deformation of the dam under different hydrological conditions. According to the simulation results, the point of action and force of the water body on the dam body are determined, including water pressure, water flow force, etc., as well as the spatial position of the point of action.
[0119] Preferably, step S13 comprises the following steps:
[0120] Step S131: performing time series decomposition processing on the historical hydrological data of the dam body and the historical deformation data of the dam body to obtain sequence dam hydrological data and sequence dam deformation data respectively;
[0121] Step S132: extracting nonlinear features from the sequence dam body hydrological data to generate hydrological dynamic characteristic data;
[0122] Step S133: performing spatial entropy correlation analysis based on the hydrological dynamic characteristic data to obtain hydrological spatial entropy correlation data;
[0123] Step S134: calculating the deformation rate of the sequence dam body deformation data to generate deformation rate data;
[0124] Step S135: performing deformation pattern clustering processing on the sequence dam body deformation data through the deformation rate data to generate deformation pattern clustering data;
[0125] Step S136: using the hydrological spatial entropy related data to perform multivariate time series fitting processing on the deformation pattern clustering data to obtain hydrological-deformation fitting data;
[0126] Step S137: Perform principal component analysis based on the hydrological-deformation fitting data, and perform coupled regression processing to generate hydrological-deformation spatial relationship data.
[0127] In an embodiment of the present invention, a time series analysis method, such as a moving average method, an exponential smoothing method, etc., is used to decompose the historical hydrological data and historical deformation data of the dam body, and decompose them into trend items, seasonal items, and residual items. For example, the hydrological data for the past 10 years include monthly rainfall and river flow. First, a time series decomposition method (such as STL decomposition) is used to decompose the hydrological data into three parts: trend, seasonality, and residual. After the time series decomposition, the trend, seasonal changes (such as seasonal rainfall) of monthly rainfall and river flow, and the residual part (i.e., random fluctuations that cannot be explained by trends and seasonality) are obtained. For deformation data, a fractal analysis method is used to extract the fractal dimension of the hydrological data as a feature. For each month in the historical hydrological data, its fractal dimension is calculated to obtain a time series that describes the nonlinear dynamic characteristics of the hydrological data. Information entropy is used as an indicator for spatial entropy correlation analysis. For hydrological dynamic characteristic data, the information entropy of each monitoring point is calculated, and its spatial distribution characteristics are analyzed. In this way, hydrological spatial entropy related data is obtained. For the historical displacement monitoring data, the monthly displacement change is calculated and divided by the time interval to obtain the monthly displacement change rate. Clustering analysis methods, such as K-means clustering, are used to cluster the deformation rate data. For example, three cluster centers, i.e., three types of deformation modes, are set. The clustering algorithm divides the deformation rate data into three categories, each representing a deformation mode. The clustering results are analyzed. Assume that the first type of deformation mode is characterized by slow deformation, with slow changes in displacement and strain; the second type of deformation mode is characterized by rapid deformation, with drastic changes in displacement and strain; and the third type of deformation mode is characterized by stable deformation, with relatively stable changes. Multivariate time series fitting is performed on the hydrological data and deformation mode clustering data. Methods such as multivariate linear regression analysis can be used to establish a relationship model between hydrology and different deformation modes. Principal component analysis is performed on the hydrological-deformation fitting data to extract the main influencing factors. For example, the influencing factors include rainfall, river flow, etc. Coupled regression analysis is performed on the hydrological data and deformation data. By establishing a relationship model between hydrology and deformation and considering the influence of the principal components, more accurate regression results can be obtained. Based on the regression results, the hydrological-deformation spatial relationship data are generated. These data describe the spatial distribution law and influence degree of dam deformation under different hydrological conditions.
[0128] Preferably, step S2 comprises the following steps:
[0129] Step S21: identifying the boundary contour of the dam body according to the dam body surface terrain model, and generating dam body boundary contour data;
[0130] Step S22: extract key geological features according to the dam body geological structure data to generate key geological feature data;
[0131] Step S23: using the dam surface terrain model to perform geological structure analysis processing on the key geological feature data to generate geological structure analysis array data;
[0132] Step S24: processing the dam body boundary contour data for potential deformation hotspots of the dam body through the geological structure analysis array data and the hydrological dynamic action point data to generate potential deformation hotspot data;
[0133] Step S25: Optimize the observation points according to the potential deformation hotspot data and deploy adaptive observation equipment to build a dam body observation network;
[0134] Step S26: Using the dam body observation network to perform real-time dam body deformation monitoring on the concrete dam, and generate real-time dam body deformation monitoring data.
[0135] In an embodiment of the present invention, the boundary contour is identified on the 3D terrain model. The boundary contour of the dam body can be determined by computer vision technology or manual drawing. According to the identification result, the contour data of the dam body boundary is extracted and recorded in digital form, including information such as the coordinate points of the dam top, dam bottom and both sides of the boundary. According to geological principles, key geological features that affect the stability and deformation of the dam body, such as the rock property change profile, the distribution of fault zones, etc. are extracted. The key geological feature data are superimposed on the terrain model to perform geological structure analysis processing. For example, according to the terrain model and the fault zone data, the spatial trend and extension of the fault zone in the dam body are analyzed, and a spatial distribution map of the fault zone is generated. The geological structure analysis array data is integrated with the hydrodynamic action point data, and combined with the dam body boundary contour data, data analysis is performed to identify hot spots with potential deformation risks, that is, areas that are greatly affected by geological structure and hydrodynamics. According to the potential deformation hot spot data, the appropriate observation point site selection location is determined to cover the potential deformation hot spot area. According to the selected observation point location, the observation equipment is deployed, including but not limited to the installation of inclinometers, thermocouples, GPS sensors and rain gauges. The monitoring data collected in real time is transmitted to the data processing center through the network, and is processed and analyzed in real time to generate real-time dam deformation monitoring data.
[0136] Preferably, step S24 comprises the following steps:
[0137] Step S241: Calculate the surface curvature according to the dam body boundary contour data to generate dam body surface curve data;
[0138] Step S242: performing dam body slope analysis according to the dam body surface curve data to generate dam body slope area data; performing flat area analysis according to the dam body surface curve data to generate dam body flat area data;
[0139] Step S243: marking the dam body corners of the dam body flat area data by using the dam body slope area data, thereby obtaining the dam body corner data; marking the dam body corner data as easily deformable area data;
[0140] Step S244: performing geological weak point processing on the dam body boundary contour data through the geological structure analysis array data to generate geological weak point data;
[0141] Step S245: performing surface action processing on the easily deformable area data through the geological weak point data to generate high deformation hot spot area data;
[0142] Step S246: supplementing the high deformation hotspot area data with potential deformation hotspots through the hydrodynamic action point data to obtain potential deformation hotspot data.
[0143] In an embodiment of the present invention, a mathematical calculation method is used to calculate the curvature of the boundary contour data. Commonly used methods include fitting method, numerical differentiation method, etc. For example, through mathematical calculation, the curvature value of each point on the boundary contour of the dam body can be calculated, thereby generating the surface curve data of the dam body. According to the surface curve data, the slope information of each point is calculated, that is, the slope direction of the dam body at the location of the point. The calculation can be performed using numerical differentiation method or geographic information system (GIS) tools. The slope information is spatially segmented to divide areas with different slope directions, and the dam body slope area data is generated. The surface curve data is analyzed in flat areas to identify the flat areas on the dam body surface. Based on the flat area data of the dam body, the dam body slope area data is used to identify the areas with corners on the dam body surface. These corner areas are key areas where the dam body is prone to deformation. The corner data is marked as deformable area data to indicate areas with a higher risk of deformation. Through the geological structure analysis array data, the geological weak points on the boundary contour of the concrete dam can be identified, such as areas with fault zones or rock formation anomalies, and these areas can be marked as geological weak points, and the corresponding geological weak point data can be generated. The geological weak point data is analyzed with the deformable area data to determine the degree of influence of surface effects on the deformable area. This includes surface effects such as gravity, surface settlement, and rock landslides. According to the influence of geological weak points and the degree of surface effects, high deformation hotspot areas in the deformable area that are greatly affected by surface effects are identified. Through the hydrodynamic action point data, the high deformation hotspot area data is further analyzed to identify potential deformation hotspots that are greatly affected by hydrodynamic effects. In this way, deformation hotspots are supplemented and the corresponding potential deformation hotspots are identified.
[0144] Preferably, step S3 comprises the following steps:
[0145] Step S31: performing multi-scale data preprocessing according to the real-time dam deformation monitoring data to generate multi-scale dam monitoring data, wherein the multi-scale dam monitoring data includes real-time water pressure data, dam slope inclinometer readings, thermocouple temperature readings and precipitation data;
[0146] Step S32: performing data cutting processing on the multi-scale dam body monitoring data through a preset time window, and performing wavelet analysis processing to generate dam body monitoring time series data;
[0147] Step S33: extracting trend components according to the dam body monitoring time series data to generate real-time dam body trend component data;
[0148] Step S34: performing seasonal component processing on the dam body monitoring time series data based on the real-time dam body trend component data to generate real-time dam body seasonal component data;
[0149] Step S35: using the dam body monitoring time series data to calculate residual data items for the real-time dam body trend component data and the real-time dam body seasonal component data, to generate real-time dam body residual data;
[0150] Step S36: using the preset standard dam body monitoring threshold data to perform abnormal component detection on the real-time dam body seasonal component data and the real-time dam body trend component data, and perform periodic pattern judgment to generate real-time dam body deformation pattern data, wherein the real-time dam body deformation pattern data includes periodic dam body deformation data and non-periodic dam body deformation data;
[0151] Step S37: when the real-time dam body deformation mode data is periodic dam body deformation data, performing detrending detection processing on the periodic dam body deformation data to generate detrending detection data;
[0152] Step S38: when the real-time dam body deformation mode data is non-periodic dam body deformation data, abnormal data points are marked on the non-periodic dam body deformation data to generate abnormal data point data;
[0153] Step S39: Perform abnormal detection event processing on the real-time dam residual data by removing the trend detection data and the abnormal data point data to generate abnormal detection event data.
[0154] In an embodiment of the present invention, real-time monitoring data from water pressure sensors, slope inclinometers, thermocouple temperature sensors and precipitation monitoring equipment are collected. First, these data are cleaned and corrected, and then they are integrated together to form multi-scale dam monitoring data. A predetermined time window is set, which is usually divided into certain time intervals, such as hourly, daily or weekly. The multi-scale dam monitoring data is cut according to the set time window to form data segments within the corresponding time period. Wavelet analysis is performed on the data segments within each time window to extract the frequency domain and time domain characteristics of the data. By applying the moving average method, long-term trend components, such as the seasonal change trend of the water level, are extracted from the dam monitoring time series data. By integrating these trend components together, real-time dam trend component data can be obtained. By using the periodic decomposition method, seasonal periodic components, such as the periodic fluctuation of the water level at the same time every year, are extracted from the dam monitoring time series data. By integrating these seasonal components together, real-time dam seasonal component data can be obtained. The real-time dam trend component data and the real-time dam seasonal component data are subtracted from the dam monitoring time series data to obtain the residual data item, which is used to describe the non-trend and non-periodic changes of the data. For example, the real-time dam water pressure trend component data and the real-time dam water pressure seasonal component data are subtracted from the daily water pressure data to obtain the daily residual data item. In this way, the real-time dam water pressure residual data is obtained. For example, the standard monitoring threshold is set as an abnormality when the water pressure rise rate exceeds 0.5 meters per hour, and then the real-time dam seasonal component data and the real-time dam trend component data are subjected to abnormal component detection. If the water pressure rise rate at a certain time point exceeds 0.5 meters per hour, the point is determined to be an abnormal data point. Then, through the periodicity detection algorithm, if the autocorrelation function of the abnormal data point shows obvious periodicity, the abnormal data point is classified as periodic dam deformation data, otherwise it is classified as non-periodic dam deformation data. The periodic dam deformation data is subjected to detrending detection processing, and the trend component is removed by fitting curves or sliding windows to obtain detrending detection data. For example, statistical methods or jump detection algorithms are used for non-periodic water pressure data over a period of time to determine whether each data point has obvious abnormal changes compared with the surrounding data. If so, it is marked as an abnormal data point. If the rate of increase of water pressure exceeds the threshold in a specific time period and there are abnormal water pressure data points at the same time, it can be determined that this is an abnormal detection event. The event is recorded and abnormal detection event data is generated.
[0155] Preferably, step S4 comprises the following steps:
[0156] Step S41: performing abnormal event impact assessment processing according to the abnormal detection event data to generate abnormal event assessment data;
[0157] Step S42: screening dam body deformation influencing events according to the abnormal event assessment data to obtain dam body deformation influencing event data;
[0158] Step S43: performing geographic orientation fitting processing on the concrete dam through the dam body deformation impact event data to generate abnormal event geographic fitting data;
[0159] Step S44: performing space-time distance calculation on the geographic fitting data of the abnormal event to generate space-time distance data;
[0160] Step S45: using a preset spatiotemporal clustering algorithm and spatiotemporal distance data to perform spatiotemporal clustering processing on the geographic fitting data of abnormal events, and perform abnormal event area identification to generate abnormal event area data;
[0161] Step S46: Process the key impact indexes of the potential abnormal event area data through the dam deformation impact event data to generate spatiotemporal joint early warning index data.
[0162] In the embodiment of the present invention, the nature, duration, intensity and other information of each abnormal event are analyzed for the abnormal detection event data. According to the characteristics of the abnormal event, combined with the preset evaluation criteria, the impact of the abnormal event is evaluated, including its impact on the safety and stability of the dam body, the surrounding environment and facilities. For example, a certain abnormal event is a sudden increase in water pressure, which lasts for 1 hour and has a strong intensity, which will affect the stability of the dam body. According to the preset evaluation criteria, it is evaluated as a high-risk event, and its impact degree and duration and other information are recorded to form abnormal event evaluation data. According to the abnormal event evaluation data, abnormal events that affect the deformation of the dam body are screened out, such as events that pose potential risks to the stability of the dam body. The dam body deformation impact event data is matched with the geographic information of the concrete dam to determine the specific geographical location and orientation involved in each event. For example, the sudden increase in water pressure event is matched to the downstream area on the right side of the dam body, and the abnormal fluctuation event of the inclination is matched to the upstream area on the left side of the dam body. The spatiotemporal distance is calculated for each pair of abnormal event geographic fitting data. Spatiotemporal distance can be divided into two dimensions: temporal distance and spatial distance: Temporal distance: usually measured in time units (such as hours, days), calculating the time difference between events. Spatial distance: measured in geographical coordinates or distances between regional ranges, which can be calculated using Euclidean distance, Manhattan distance or Hammanton distance. For example, there is a set of abnormal event geographic fitting data, which includes the geographical locations of sudden water pressure rise events and abnormal inclination fluctuation events, as well as the spatiotemporal distance data between them. The ST-DBSCAN algorithm is selected for clustering, and a certain spatiotemporal density threshold is set. The spatiotemporal distance matrix is clustered using the ST-DBSCAN algorithm. According to the algorithm's calculation results, two clusters are obtained, representing water pressure events and inclination events, respectively. According to the clustering results, the center or important representative point of each cluster is determined, and the average value of the geographical location in each cluster can be selected as the representative point, which is regarded as the representative of the abnormal event area. These areas are the geographical areas involved in a group of abnormal events with similar spatiotemporal characteristics. According to the data of dam deformation impact events, the key impact indicators corresponding to each abnormal event area are determined. These indicators usually reflect the impact of dam deformation on safety and stability, such as displacement rate, stress distribution, inclination change, etc. The selected key impact indicators are matched with the data of potential abnormal event areas, and the spatiotemporal joint early warning indicators are calculated based on the integrated data, combined with time and space dimensions. These indicators are the weighted average, standardized value, change rate, etc. of the key impact indicators in the abnormal event area.
[0163] Preferably, step S5 comprises the following steps:
[0164] Step S51: constructing a finite element relationship model for the dam body segmentation unit data through the abnormal event area data to obtain a finite element relationship model of the dam body;
[0165] Step S52: solving the mechanical response of the finite element relationship model of the dam body to generate dam body response force data;
[0166] Step S53: using the dam body response force data to perform spatial correlation processing on the hydrological-dam body deformation dynamic model to generate spatial deformation stress data;
[0167] Step S54: constructing a chaotic attractor according to the spatial deformation stress data, and defining the attractor characteristics to generate attractor characteristic data;
[0168] Step S55: Perform iterative calculation of deformation stress of the concrete dam through attractor characteristic data, and perform chaos simulation prediction to generate predicted dam deformation trend data.
[0169] As an example of the present invention, refer to Figure 2 As shown, Figure 1 Detailed implementation steps of step S5 in the flowchart, in this example, step S5 includes:
[0170] Step S51: constructing a finite element relationship model according to the abnormal event area data to obtain a finite element relationship model of the dam body;
[0171] In the embodiment of the present invention, the concrete dam has divided the dam body into a plurality of small blocks or grid units, each unit representing a local area of the dam body. These units can be two-dimensional plane units or three-dimensional volume units. A finite element model of the dam body is established based on the abnormal event area data and the dam body segmentation unit data. This model describes the relationship between the abnormal event area and the dam body segmentation unit, and may include information such as the connection mode, constraints and boundary conditions between units. For example, based on the abnormal event area data, it is found that the left upstream area of the dam body is affected by the inclination abnormal event, while the right downstream area is affected by the water pressure abnormal event. The affected units are marked as potential abnormal event areas. Then, based on the spatial position relationship between the units and the impact range of the abnormal event, the connection relationship and the degree of influence between the units are determined. For example, for the inclination abnormal event area, the grid units nearby are affected, while the water pressure abnormal event area has a wider impact on the adjacent units.
[0172] Step S52: solving the mechanical response of the finite element relationship model of the dam body to generate dam body response force data;
[0173] In an embodiment of the present invention, before solving the mechanical response, it is necessary to preprocess the model. This includes defining material properties, loading conditions, boundary conditions, etc. The finite element relationship model of the dam body is converted into a computer-recognizable form. According to the characteristics of the model and the solution requirements, a suitable finite element solver is selected. Common solvers include ABAQUS, ANSYS, LS-DYNA, etc. For example, in ABAQUS, the geometric model of the dam body is first imported, and the material properties such as the elastic modulus and Poisson's ratio of the material are defined. Then, the loading conditions are set, such as applying water pressure loads or inclination loads, and boundary conditions, such as fixed boundaries or applied displacement boundaries. Run the ABAQUS solver to solve the mechanical response of the finite element relationship model of the dam body. The solver will consider the set loading conditions and boundary conditions and calculate the mechanical response of each part of the dam body. Use the post-processing tools provided by ABAQUS to view the solution results and extract the response data of interest, such as stress, strain, displacement and other information at each unit or node.
[0174] Step S53: using the dam body response force data to perform spatial correlation processing on the hydrological-dam body deformation dynamic model to generate spatial deformation stress data;
[0175] In an embodiment of the present invention, for example, the dam body response force data obtained by finite element simulation, such as the stress value at each unit. The size and position of the water level change are associated with the stress value at each unit of the dam body. For example, the water pressure corresponding to a water level rise of 100 meters is matched with the stress value at the unit in the upstream area of the dam body. Based on the results of the spatial association processing, spatial deformation stress data is generated, which describes the stress conditions of various parts of the dam body under different water level conditions. For example, the stress value of the upstream area of the dam body under different water level conditions can be obtained, so as to understand the stress impact of water level changes on this area.
[0176] Step S54: constructing a chaotic attractor according to the spatial deformation stress data, and defining the attractor characteristics to generate attractor characteristic data;
[0177] In an embodiment of the present invention, first, a suitable data set is selected from the spatial deformation stress data, such as data within a specific time period. The selected data set is reconstructed in phase space, that is, multidimensional data is mapped into a high-dimensional space. Commonly used methods include delayed embedding method and phase space interpolation method. The chaotic attractor is estimated for the data in the phase space, and the commonly used methods include Grassberger-Procaccia algorithm, False Nearest Neighbors algorithm, etc. The estimated chaotic attractor is subjected to locality analysis, that is, the local structural characteristics of the attractor surface are determined. This can be done by methods such as Lyapunov index and bifurcation diagram. The parameters in the attractor construction process are adjusted to further optimize the characteristics of the attractor. This involves adjusting the delay parameters, embedding dimensions, etc. of the phase space reconstruction. According to the construction results and feature definition results of the attractor, the key features of the attractor are extracted. This includes the dimension, fractal dimension, Lyapunov index, etc. of the attractor.
[0178] Step S55: Perform iterative calculation of deformation stress of the concrete dam through attractor characteristic data, and perform chaos simulation prediction to generate predicted dam deformation trend data.
[0179] In the embodiment of the present invention, the attractor characteristic data is used as the initial condition to perform iterative calculation of deformation stress. During the iterative calculation process, appropriate mathematical models and algorithms can be used to convert the attractor characteristic data into specific dam body deformation stress values. Through continuous iteration, the dam body deformation stress data at different time points are obtained. Based on the deformation stress data obtained by iterative calculation, chaos simulation prediction is performed. The prediction methods in chaos theory, such as phase space reconstruction, delayed reconstruction, etc., can be used to predict the deformation trend of the dam body in the future. The prediction results include information such as the trend of dam body deformation and the fluctuation range.
[0180] Preferably, step S6 comprises the following steps:
[0181] Step S61: Perform risk weight factor analysis on the spatiotemporal joint warning indicator data to generate indicator weight factor data;
[0182] Step S62: Formulate deformation-scoring fuzzy rules according to the indicator weight factor data to generate deformation-scoring fuzzy rule data;
[0183] Step S63: constructing a rule reasoning model through deformation-scoring fuzzy rule data;
[0184] Step S64: using the rule inference model to process the predicted dam deformation trend data with risk factor scores to generate risk factor score data;
[0185] Step S65: Calculate the warning value according to the risk factor score data, classify the warning level, and generate dam body warning data.
[0186] As an example of the present invention, refer to Figure 3 As shown, Figure 1 Detailed implementation steps of step S6 in the flowchart, in this example, step S6 includes:
[0187] Step S61: Perform risk weight factor analysis on the spatiotemporal joint warning indicator data to generate indicator weight factor data;
[0188] In the embodiment of the present invention, the analytic hierarchy process (AHP) is used to perform risk weight factor analysis to determine the degree of influence of each factor on the safety of the dam body. For each factor, a hierarchical structure can be established, and then the weights of indicators at each level can be determined by expert scoring or data analysis methods. These weight factors are applied to the spatiotemporal joint early warning indicator data to determine the relative importance of each indicator in the safety of the dam body, and generate indicator weight factor data.
[0189] Step S62: Formulate deformation-scoring fuzzy rules according to the indicator weight factor data to generate deformation-scoring fuzzy rule data;
[0190] In the embodiment of the present invention, based on the indicator weight factor data, a deformation-scoring fuzzy rule is formulated to evaluate the safety status of the dam body according to the changes in the monitoring indicators. For example, there are three monitoring indicators: water pressure, temperature and displacement. According to expert opinions and indicator weight factor data, the following fuzzy rules are formulated: if the water pressure increases and the weight factor is high, the score is high-risk; if the temperature increases and the weight factor is low, the score is average; if the displacement increases and the weight factor is high, the score is high-risk; in other cases, the score is safe.
[0191] Step S63: constructing a rule reasoning model through deformation-scoring fuzzy rule data;
[0192] In the embodiment of the present invention, for example, a rule reasoning model is constructed based on deformation-scoring fuzzy rule data. The rule reasoning model can use fuzzy logic, fuzzy reasoning, fuzzy association rules and other technologies to match the input monitoring data with the rules, and derive the safety score of the dam body based on the matching results.
[0193] Step S64: using the rule inference model to process the predicted dam deformation trend data with risk factor scores to generate risk factor score data;
[0194] In an embodiment of the present invention, the predicted dam deformation trend data is input into a rule reasoning model for rule reasoning processing. The rule reasoning model derives risk factor scoring data based on the predicted dam deformation trend data and the deformation-scoring fuzzy rule data. Based on the result of the rule reasoning processing, risk factor scoring data is generated, which reflects the degree of influence of the dam deformation trend on the risk. The risk factor scoring data can be divided into different levels, such as high risk, medium risk, low risk and the like. For example, the deformation trend data of the dam body in the future period of time, such as the trend of gradually increasing deformation, is obtained by simulating and predicting the attractor characteristic data. The predicted dam deformation trend data is input into a rule reasoning model for rule reasoning processing. Based on the result of the rule reasoning processing, risk factor scoring data is generated, for example, the output is "high risk", indicating that the predicted dam deformation trend poses a greater risk to the safety of the dam body.
[0195] Step S65: Calculate the warning value according to the risk factor score data, classify the warning level, and generate dam body warning data.
[0196] In an embodiment of the present invention, the early warning value is calculated based on the risk factor scoring data. The calculation of the early warning value can be based on statistical methods, expert experience or model analysis. Based on the risk factor scoring data, the early warning value is calculated using a mathematical model, for example, the early warning value is 70. The calculated early warning value is classified into early warning levels. The early warning values are divided into three levels: high, medium and low. According to the size of the early warning value 70, it is determined to be a medium-level early warning. According to the early warning level classification, dam body early warning data is generated. The early warning data includes information such as the early warning level, early warning value, and early warning time, which is used to guide relevant personnel to take corresponding safety measures and countermeasures.
[0197] The present invention also provides a system for formulating a spatiotemporal joint early warning index for deformation of a concrete dam, which executes the method for formulating a spatiotemporal joint early warning index for deformation of a concrete dam as described above. The system for formulating a spatiotemporal joint early warning index for deformation of a concrete dam comprises:
[0198] The dam hydrological simulation module is used to collect dam engineering data for concrete dams and generate dam engineering data; construct a 3D digital terrain model based on the dam engineering data, optimize the dynamic model, and generate a hydrological-dam deformation dynamic model; perform hydrological dynamic numerical simulation through the hydrological-dam deformation dynamic model to generate hydrological dynamic action point data;
[0199] The dam deformation monitoring module is used to identify the dam boundary contour according to the dam engineering data and generate the dam boundary contour data; to process the dam boundary contour data for potential deformation hotspots through the hydrological dynamic action point data and generate potential deformation hotspot data; to construct a dam observation network according to the potential deformation hotspot data and perform real-time dam deformation monitoring to generate real-time dam deformation monitoring data;
[0200] The dam body anomaly detection module is used to extract trend components based on real-time dam body deformation monitoring data to generate real-time dam body trend component data; to judge the real-time dam body deformation mode based on the real-time dam body trend component data to generate real-time dam body deformation mode data; to process anomaly detection events based on the real-time dam body deformation mode data to generate anomaly detection event data;
[0201] The early warning indicator formulation module is used to screen dam deformation impact events based on abnormal detection event data to obtain dam deformation impact event data; perform geographic space fitting processing on dam deformation impact event data to generate abnormal event geographic fitting data; perform spatiotemporal clustering processing based on abnormal event geographic fitting data to generate abnormal event area data; perform key impact indicator processing based on abnormal event area data to generate spatiotemporal joint early warning indicator data;
[0202] The deformation trend prediction module is used to solve the finite element mechanical response of the hydrological-dam deformation dynamic model through the abnormal event area data to generate the dam body response force data; the deformation stress of the concrete dam is iteratively processed through the dam body response force data to generate the predicted dam body deformation trend data;
[0203] The dam body early warning assessment module is used to perform risk weight factor analysis on the spatiotemporal joint early warning index data to generate index weight factor data; use the index weight factor data to perform risk factor scoring on the predicted dam body deformation trend data to generate risk factor scoring data; classify the early warning level according to the risk factor scoring data to generate dam body early warning data.
[0204] The beneficial effect of this application is that by collecting high-precision and high-resolution dam engineering data on site, combining geographic information system (GIS) and three-dimensional modeling software, a three-dimensional digital model of the concrete dam and the surrounding terrain is constructed. The model intuitively displays the spatial structure of the dam, including information such as geometric shape and spatial position relationship. Based on this model, the dynamic response characteristics of the concrete dam are analyzed by combining structural mechanics and fluid mechanics, and the influence of hydrological factors such as water level, flow rate, and flow velocity on the deformation of the dam body is considered. Image processing and pattern recognition technology are used to automatically identify the boundary contour of the dam body, and combined with the data of hydrological and dynamic action points, potential deformation hot spots that are greatly affected by hydrological and dynamic forces are identified, the location of monitoring points is optimized, and the monitoring efficiency and pertinence are improved. By extracting trend components, the influence of short-term factors on deformation monitoring data is eliminated, deformation patterns are analyzed, real-time deformation data is monitored, and abnormal detection events are automatically processed. Combined with geographic space fitting and spatiotemporal clustering, the spatial distribution characteristics of events are visualized and areas with higher risks are identified. Key influencing indicators are extracted to quantify the degree of influence of events on dam safety, and finite element mechanical response solutions are performed to predict future dam deformation trends. Dam risks are graded according to risk scores, and early warning information is issued in a timely manner to improve safety.
[0205] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.
[0206] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.
Claims
1. A method for formulating a spatiotemporal joint early warning index for concrete dam deformation, characterized in that: The following steps are involved: Step S1: collecting dam body engineering data of the concrete dam to generate dam body engineering data; constructing a 3D digital terrain model based on the dam body engineering data, optimizing the dynamic model, and generating a hydrological-dam body deformation dynamic model; performing hydrological-dynamic numerical simulation through the hydrological-dam body deformation dynamic model to generate hydrological-dynamic action point data; Step S2: Identify the dam body boundary contour according to the dam body engineering data to generate dam body boundary contour data; process the dam body boundary contour data for potential deformation hotspots through the hydrodynamic action point data to generate potential deformation hotspot data; construct a dam body observation network according to the potential deformation hotspot data, and perform real-time dam body deformation monitoring to generate real-time dam body deformation monitoring data; Step S3: extracting trend components according to the real-time dam deformation monitoring data to generate real-time dam trend component data; The real-time dam body deformation mode is judged through the real-time dam body trend component data to generate the real-time dam body deformation mode data; Perform abnormal detection event processing based on real-time dam body deformation pattern data to generate abnormal detection event data; Step S4: screening dam body deformation influencing events according to the abnormal detection event data to obtain dam body deformation influencing event data; The data of dam deformation influencing events are processed by geographic space fitting to generate geographic fitting data of abnormal events; spatiotemporal clustering is performed based on the geographic fitting data of abnormal events to generate regional data of abnormal events; key impact indicators are processed based on the regional data of abnormal events to generate spatiotemporal joint early warning indicator data; Step S5: performing finite element mechanical response solution on the hydrological-dam deformation dynamic model through the abnormal event area data to generate dam body response force data; The deformation stress of the concrete dam is iteratively processed through the dam body response force data to generate the predicted dam body deformation trend data; Step S6: Perform risk weight factor analysis on the spatiotemporal joint early warning index data to generate index weight factor data; perform risk factor scoring processing on the predicted dam deformation trend data using the index weight factor data to generate risk factor scoring data; perform early warning level classification based on the risk factor scoring data to generate dam early warning data.
2. The method for formulating the spatiotemporal joint early warning index of concrete dam deformation according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: collecting dam body engineering data of the concrete dam, and performing data cleaning processing to generate dam body engineering data, wherein the dam body engineering data includes dam body surface graphic data, dam body historical deformation data, dam body historical hydrological data and dam body geological structure data; Step S12: constructing a 3D digital terrain model based on the dam body surface graphic data, and performing dam body gridding processing to generate a dam body surface terrain model; Step S13: Analyze the hydrological deformation characteristics according to the historical hydrological data and historical deformation data of the dam body to generate hydrological-deformation relationship data; Step S14: optimizing the hydrological-deformation dynamic model of the dam surface terrain model using the hydrological-deformation relationship data to generate a hydrological-dam deformation dynamic model; Step S15: Perform hydrological dynamic numerical simulation through the hydrological-dam deformation dynamic model, and process the water body action point to generate hydrological dynamic action point data.
3. The method for formulating the spatiotemporal joint early warning index of concrete dam deformation according to claim 2 is characterized in that: Step S13 includes the following steps: Step S131: performing time series decomposition processing on the historical hydrological data of the dam body and the historical deformation data of the dam body to obtain sequence dam hydrological data and sequence dam deformation data respectively; Step S132: extracting nonlinear features from the sequence dam body hydrological data to generate hydrological dynamic characteristic data; Step S133: performing spatial entropy correlation analysis based on the hydrological dynamic characteristic data to obtain hydrological spatial entropy correlation data; Step S134: calculating the deformation rate of the sequence dam body deformation data to generate deformation rate data; Step S135: performing deformation pattern clustering processing on the sequence dam body deformation data through the deformation rate data to generate deformation pattern clustering data; Step S136: using the hydrological spatial entropy related data to perform multivariate time series fitting processing on the deformation pattern clustering data to obtain hydrological-deformation fitting data; Step S137: Perform principal component analysis based on the hydrological-deformation fitting data, and perform coupled regression processing to generate hydrological-deformation spatial relationship data.
4. The method for formulating the spatiotemporal joint early warning index of concrete dam deformation according to claim 2 is characterized in that: Step S2 includes the following steps: Step S21: identifying the boundary contour of the dam body according to the dam body surface terrain model, and generating dam body boundary contour data; Step S22: extracting key geological features according to the dam body geological structure data to generate key geological feature data; Step S23: using the dam surface terrain model to perform geological structure analysis processing on the key geological feature data to generate geological structure analysis array data; Step S24: processing the dam body boundary contour data for potential deformation hotspots of the dam body through the geological structure analysis array data and the hydrological dynamic action point data to generate potential deformation hotspot data; Step S25: Optimize the observation points according to the potential deformation hotspot data and deploy adaptive observation equipment to build a dam body observation network; Step S26: Using the dam body observation network to perform real-time dam body deformation monitoring on the concrete dam, and generate real-time dam body deformation monitoring data.
5. The method for formulating the spatiotemporal joint early warning index of concrete dam deformation according to claim 4 is characterized in that: Step S24 includes the following steps: Step S241: Calculate the surface curvature according to the dam body boundary contour data to generate dam body surface curve data; Step S242: performing dam body slope analysis according to the dam body surface curve data to generate dam body slope area data; performing flat area analysis according to the dam body surface curve data to generate dam body flat area data; Step S243: marking the dam body corners of the dam body flat area data by using the dam body slope area data, thereby obtaining the dam body corner data; marking the dam body corner data as easily deformable area data; Step S244: performing geological weak point processing on the dam body boundary contour data through the geological structure analysis array data to generate geological weak point data; Step S245: performing surface action processing on the easily deformable area data through the geological weak point data to generate high deformation hot spot area data; Step S246: supplementing the high deformation hotspot area data with potential deformation hotspots through the hydrodynamic action point data to obtain potential deformation hotspot data.
6. The method for formulating the spatiotemporal joint early warning index of concrete dam deformation according to claim 1 is characterized in that: Step S3 includes the following steps: Step S31: performing multi-scale data preprocessing according to the real-time dam deformation monitoring data to generate multi-scale dam monitoring data, wherein the multi-scale dam monitoring data includes real-time water pressure data, dam slope inclinometer readings, thermocouple temperature readings and precipitation data; Step S32: performing data cutting processing on the multi-scale dam body monitoring data through a preset time window, and performing wavelet analysis processing to generate dam body monitoring time series data; Step S33: extracting trend components according to the dam body monitoring time series data to generate real-time dam body trend component data; Step S34: performing seasonal component processing on the dam body monitoring time series data based on the real-time dam body trend component data to generate real-time dam body seasonal component data; Step S35: using the dam body monitoring time series data to calculate residual data items for the real-time dam body trend component data and the real-time dam body seasonal component data, to generate real-time dam body residual data; Step S36: using the preset standard dam body monitoring threshold data to perform abnormal component detection on the real-time dam body seasonal component data and the real-time dam body trend component data, and perform periodic pattern judgment to generate real-time dam body deformation pattern data, wherein the real-time dam body deformation pattern data includes periodic dam body deformation data and non-periodic dam body deformation data; Step S37: when the real-time dam body deformation mode data is periodic dam body deformation data, performing detrending detection processing on the periodic dam body deformation data to generate detrending detection data; Step S38: when the real-time dam body deformation mode data is non-periodic dam body deformation data, abnormal data points are marked on the non-periodic dam body deformation data to generate abnormal data point data; Step S39: Perform abnormal detection event processing on the real-time dam residual data by removing the trend detection data and the abnormal data point data to generate abnormal detection event data.
7. The method for formulating the spatiotemporal joint early warning index of concrete dam deformation according to claim 1 is characterized in that: Step S4 includes the following steps: Step S41: performing abnormal event impact assessment processing according to the abnormal detection event data to generate abnormal event assessment data; Step S42: screening dam body deformation influencing events according to the abnormal event assessment data to obtain dam body deformation influencing event data; Step S43: performing geographic orientation fitting processing on the concrete dam through the dam body deformation impact event data to generate abnormal event geographic fitting data; Step S44: performing space-time distance calculation on the geographic fitting data of the abnormal event to generate space-time distance data; Step S45: using a preset spatiotemporal clustering algorithm and spatiotemporal distance data to perform spatiotemporal clustering processing on the geographic fitting data of abnormal events, and perform abnormal event area identification to generate abnormal event area data; Step S46: Process the key impact indexes of the potential abnormal event area data through the dam deformation impact event data to generate spatiotemporal joint early warning index data.
8. The method for formulating the spatiotemporal joint early warning index of concrete dam deformation according to claim 1 is characterized in that: Step S5 includes the following steps: Step S51: constructing a finite element relationship model according to the abnormal event area data to obtain a finite element relationship model of the dam body; Step S52: solving the mechanical response of the finite element relationship model of the dam body to generate dam body response force data; Step S53: using the dam body response force data to perform spatial correlation processing on the hydrological-dam body deformation dynamic model to generate spatial deformation stress data; Step S54: constructing a chaotic attractor according to the spatial deformation stress data, and defining the attractor characteristics to generate attractor characteristic data; Step S55: Perform iterative calculation of deformation stress of the concrete dam through attractor characteristic data, and perform chaos simulation prediction to generate predicted dam deformation trend data.
9. The method for formulating the spatiotemporal joint early warning index of concrete dam deformation according to claim 1 is characterized in that: Step S6 includes the following steps: Step S61: Perform risk weight factor analysis on the spatiotemporal joint warning indicator data to generate indicator weight factor data; Step S62: Formulate deformation-scoring fuzzy rules according to the indicator weight factor data to generate deformation-scoring fuzzy rule data; Step S63: constructing a rule reasoning model through deformation-scoring fuzzy rule data; Step S64: using the rule inference model to process the predicted dam deformation trend data with risk factor scores to generate risk factor score data; Step S65: Calculate the warning value according to the risk factor score data, classify the warning level, and generate dam body warning data.
10. A system for formulating joint spatiotemporal early warning indicators for concrete dam deformation, characterized in that: Used to execute the method for formulating the spatiotemporal joint early warning index of concrete dam deformation as claimed in claim 1, the spatiotemporal joint early warning index formulating system for concrete dam deformation comprises: The dam hydrological simulation module is used to collect dam engineering data for concrete dams and generate dam engineering data; construct a 3D digital terrain model based on the dam engineering data, optimize the dynamic model, and generate a hydrological-dam deformation dynamic model; perform hydrological dynamic numerical simulation through the hydrological-dam deformation dynamic model to generate hydrological dynamic action point data; The dam deformation monitoring module is used to identify the dam boundary contour according to the dam engineering data and generate the dam boundary contour data; to process the dam boundary contour data for potential deformation hotspots through the hydrological dynamic action point data and generate potential deformation hotspot data; to construct a dam observation network according to the potential deformation hotspot data and perform real-time dam deformation monitoring to generate real-time dam deformation monitoring data; The dam body anomaly detection module is used to extract trend components based on real-time dam body deformation monitoring data to generate real-time dam body trend component data; to judge the real-time dam body deformation mode based on the real-time dam body trend component data to generate real-time dam body deformation mode data; to process anomaly detection events based on the real-time dam body deformation mode data to generate anomaly detection event data; The early warning indicator formulation module is used to screen dam deformation impact events based on abnormal detection event data to obtain dam deformation impact event data; perform geographic space fitting processing on dam deformation impact event data to generate abnormal event geographic fitting data; perform spatiotemporal clustering processing based on abnormal event geographic fitting data to generate abnormal event area data; perform key impact indicator processing based on abnormal event area data to generate spatiotemporal joint early warning indicator data; The deformation trend prediction module is used to solve the finite element mechanical response of the hydrological-dam deformation dynamic model through the abnormal event area data to generate the dam body response force data; the deformation stress of the concrete dam is iteratively processed through the dam body response force data to generate the predicted dam body deformation trend data; The dam body early warning assessment module is used to perform risk weight factor analysis on the spatiotemporal joint early warning index data to generate index weight factor data; use the index weight factor data to perform risk factor scoring on the predicted dam body deformation trend data to generate risk factor scoring data; classify the early warning level according to the risk factor scoring data to generate dam body early warning data.
Citation Information
Cited By
Multi-source sensor dynamic fusion dam body deformation real-time monitoring method and system
CN121007529A
