Dam body deformation risk intelligent early warning method and system

Abstract

The invention discloses a dam body deformation risk intelligent early warning method and system, and the method comprises the steps: arranging a sensor in a target dam body region, carrying out the real-time monitoring of a dam body based on the sensor, and obtaining the real-time data of the dam body; constructing theoretical deformation curves of the dam body at different water levels and temperatures; performing physical-data dual-drive anomaly detection on the real-time data and the theoretical deformation curve to obtain a deformation early warning condition of the dam body; generating a rainfall deformation threshold value of the dam body based on pre-acquired rainfall data; and constructing a multi-stage early warning linkage response mechanism of the dam body based on the deformation early warning condition and the rainfall deformation threshold, and performing deformation risk intelligent monitoring on the dam body based on the multi-stage early warning linkage response mechanism. According to the invention, the dam body deformation risk intelligent early warning efficiency can be improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of dam risk early warning, and in particular relates to an intelligent early warning method and system for dam deformation risk. Background Art

[0002] Current dam deformation risk warning technologies primarily rely on single-sensor data or simple threshold judgments, making them ineffective in addressing the multi-factor coupling in complex environments. Traditional methods fail to fully integrate multi-dimensional data such as water level, temperature, and rainfall. This results in warning models failing to accurately reflect the actual deformation patterns of the dam, leading to high false alarm rates and a significant risk of missed detections. Furthermore, when the dam is subject to periodic temperature changes or seasonal water level fluctuations, the deviation between its theoretical deformation trend and real-time monitoring data is difficult to effectively identify using traditional single models, thus reducing the accuracy of the warning system.

[0003] Furthermore, existing early warning mechanisms generally lack dynamic adaptability and real-time response capabilities. Traditional methods typically use fixed thresholds for risk assessment, making it impossible to adjust early warning strategies based on dynamic parameters such as rainfall intensity and water level rise rate. In extreme weather or unexpected operating conditions, delays in data collection and transmission networks further exacerbate the problem of insufficient early warning timeliness, preventing critical risk information from triggering an emergency response immediately and delaying the handling of dangerous situations. This static early warning model is unable to meet the high-precision, real-time safety monitoring requirements of modern water conservancy projects, and there is an urgent need to improve early warning efficiency through intelligent technologies. Summary of the Invention

[0004] The purpose of the present invention is to provide an intelligent early warning method and system for dam deformation risk to solve the above-mentioned technical problems in the prior art.

[0005] In order to solve the above problems, the present invention adopts the following solutions:

[0006] An intelligent early warning method for dam deformation risk includes the following steps:

[0007] S1. Deploy sensors in the target dam area, monitor the dam in real time based on the sensors, and collect real-time data of the dam;

[0008] S2. Construct theoretical deformation curves of the dam body at different water levels and temperatures;

[0009] S3, performing physical-data dual-driven anomaly detection on the real-time data collected in step S1 and the theoretical deformation curve constructed in step S2 to obtain deformation warning information of the dam body;

[0010] S4. Calculating a rainfall deformation threshold of the dam body based on pre-acquired rainfall data;

[0011] S5. A multi-level early warning linkage response mechanism for the dam body is established based on the deformation early warning situation and rainfall deformation threshold, and intelligent deformation risk monitoring of the dam body is carried out based on the multi-level early warning linkage response mechanism.

[0012] Further optimization, the step S1 specifically includes:

[0013] S1.1. Construct a sensor deployment plan according to the target dam area, and deploy sensors on the dam body based on the deployment plan to obtain a morphological sensing array of the dam body.

[0014] S1.2. Construct a data collection network for the dam body based on the morphological sensing array, and use the data collection network to collect real-time data on the dam body to obtain real-time data of the dam body.

[0015] 3. The intelligent early warning method for dam deformation risk according to claim 2 is characterized in that in step S1.1, sensors are embedded in geomagnetic beacons and magnetometers to obtain self-calibration sensors; multiple self-calibration sensors are arranged on the dam body according to a layout plan to obtain a dam body morphology sensing array.

[0016] Further optimization, in step 1.2, the data collection network of the dam body is constructed based on the morphological sensing array, including:

[0017] S1.2.1. Divide the morphological sensing array into grid nodes. Specifically, based on the constructed dam morphological sensing array, and in accordance with the geometric shape, size, and distribution of sensors of the dam, adopt reasonable division rules to divide the dam into several grid areas. The center or representative position of each grid area is set as a grid node. Then, the entire morphological sensing array is orderly divided into a series of grid nodes with clear position identification; each node corresponds to a unique number.

[0018] S1.2.2. Collect relative differences of the dam body based on network nodes to obtain relative difference data of the dam body; specifically, for each divided grid node, use the sensors deployed around the node to collect real-time data, take the monitoring data at a certain initial moment as the benchmark, and in the subsequent monitoring process, calculate the difference between the measurement values ​​of different sensors at the same moment and the corresponding benchmark value, and record it as the relative difference.

[0019] By performing such relative difference calculations on the sensor data around each grid node, we can comprehensively collect the changes in the dam body relative to its initial state at different locations and times, and obtain relative difference data reflecting the dynamic changes in the dam body state. These data can more intuitively reflect the changing trends and degree differences in various parts of the dam body.

[0020] S1.2.3. Construct a data collection network for the dam body based on relative difference data and grid nodes.

[0021] Further optimization, in step S2, constructing a theoretical deformation curve of the dam body at different water levels and temperatures specifically includes:

[0022] S2.1. Construct a finite element model of the dam body.

[0023] By constructing a finite element model of the dam body, the complex dam structure can be simplified into a calculable combination of discrete units. The material properties and boundary conditions can be used to simulate the actual mechanical state, laying a solid foundation for subsequent analysis. Through fine unit division and precise parameter setting, the model's degree of restoration of the dam body's actual mechanical properties can be improved.

[0024] S2.2. Preset the temperature-water level operating conditions of the dam body based on the pre-acquired dam body operation history and design standards to obtain the preset operating conditions of the dam body. Specifically, based on the pre-acquired historical data collected during the dam body operation, including water level change records, temperature fluctuations and corresponding operating status information in different time periods over the years; and referring to the dam body design standards, clarify the maximum water level, minimum water level, normal operating water level range and expected temperature change range of the dam body specified in the design.

[0025] The temperature-water level working conditions are preset based on the dam operation history and design standards, and past actual conditions and design requirements are fully considered. A variety of combined working conditions are set to realistically simulate the dam operation environment, breaking the single working condition setting mode. Both extreme and common situations are fully considered to improve the comprehensiveness and reliability of the analysis.

[0026] S2.3. Conduct finite element deformation analysis on the dam body based on the preset working conditions to obtain theoretical deformation curves of the dam body at different water levels and temperatures;

[0027] Specifically, each temperature-water level working condition preset in the previous step is loaded one by one into the constructed finite element model of the dam body; the solver of the finite element analysis software is used to calculate the stress distribution inside the dam body and the resulting deformation under each working condition, and the theoretical deformation curve of the dam body under different water level and temperature combinations is drawn.

[0028] Finite element deformation analysis is carried out based on preset working conditions to obtain theoretical deformation curves under different conditions, which intuitively show the deformation trend of the dam body and provide a key basis for safety assessment. By integrating multiple factors such as temperature and water level, the mechanical response is accurately simulated, providing more scientific and comprehensive theoretical support for dam body safety monitoring.

[0029] Further optimization, the step S3 specifically includes:

[0030] S3.1. Draw a real-time deformation curve of the dam body based on real-time data. This process generates a deformation image of the real-time data and generates a real-time deformation curve. This can visually present the current deformation state of the dam body, providing a visual basis for subsequent analysis. Advanced mapping software and a scientific coordinate system are used to accurately restore the details of the dam body's real-time deformation.

[0031] S3.2. Perform point-by-point compensation calculation on the real-time deformation curve and the theoretical deformation curve to obtain the initial residual value of the real-time data;

[0032] Specifically, the obtained real-time deformation curve is matched with the theoretical deformation curve previously obtained based on finite element analysis; on the same time axis, a point-by-point subtraction operation is performed on the deformation variables corresponding to the same time point in each curve. By performing the above operation on all time points, a series of difference data are obtained. These difference data constitute the initial residual value of the real-time data.

[0033] The initial residual value is calculated by compensating the real-time and theoretical deformation curves point by point, which can clearly reflect the difference between the actual and expected deformations and quantify the degree of this difference through precise point-by-point calculations.

[0034] S3.3. Correct the data type anomaly of the initial residual value to obtain the deformation residual value of the real-time data; specifically, conduct a comprehensive screening of the initial residual value data to identify whether there are data type anomalies caused by sensor failure, data transmission errors, etc., and use data cleaning and correction algorithms for the abnormal data found, combined with the historical data of the dam body, data from surrounding monitoring points and physical common sense to make judgments and corrections.

[0035] Correcting the data type anomaly of the initial residual value to obtain the deformation residual value can ensure data reliability. Using a variety of data processing methods, combining multi-source data and physical common sense to check and correct anomalies, lay a solid data foundation for subsequent analysis.

[0036] S3.4. Compare the physical anomalies of the dam body based on real-time data to obtain abnormal deformation information of the dam body.

[0037] Obtaining deformation anomaly information by comparing physical anomalies with real-time data helps to accurately locate areas with potential safety hazards on the dam body, set normal parameter ranges based on multiple criteria, and improve the accuracy of anomaly judgment.

[0038] S3.5. Construct a deformation warning system for the dam body based on deformation residual values ​​and abnormal deformation information. This system constructs a deformation warning system based on deformation residual values ​​and abnormal deformation information, providing an intuitive and critical warning basis for dam safety monitoring. By comprehensively considering multiple factors and formulating flexible warning level standards, it comprehensively and dynamically reflects the safety status of the dam body.

[0039] Further optimization, the step S4 specifically includes:

[0040] S4.1. Generate a baseline rainfall deformation threshold for the dam body based on pre-acquired rainfall data. Specifically, first, sort out the pre-acquired historical rainfall data for the dam body and screen out various rainfall event records with different rainfall amounts, rainfall durations, and rainfall intensities.

[0041] Secondly, the deformation data of the dam body corresponding to these rainfall events were collected, and the rainfall data and deformation data were correlated and analyzed using data statistical analysis methods to determine the general law of dam body deformation under different rainfall conditions. Based on this law, the reasonable range boundary value of the dam body's deformation under normal circumstances with rainfall changes was set, that is, the benchmark rainfall deformation threshold of the dam body.

[0042] By generating a benchmark rainfall deformation threshold for the dam body based on pre-acquired rainfall data, we can deeply explore the relationship between historical rainfall and dam body deformation, sort out general rules and set reasonable boundary values, and provide a basic reference for judging whether the dam body deformation is normal during real-time rainfall. We can comprehensively analyze the rainfall amount, duration, intensity and deformation data of multiple parts of the dam body to accurately grasp the deformation characteristics of the dam body under different rainfall conditions.

[0043] S4.2. Based on the baseline rainfall deformation threshold and real-time data, calculate the rainfall deformation threshold of the dam body. The calculation formula is as follows:

[0044]

[0045] In the above formula, T is the rainfall deformation threshold of the dam body, T0 is the baseline rainfall deformation threshold, α is the influence weight of water level change, ΔH is the water level rise value in real-time data, and Δt is the time change value in real-time data. is the water level rising rate in real-time data, β is the rainfall impact weight, and R is the rainfall intensity in real-time data.

[0046] The function of this formula is to accurately and quantitatively calculate the rainfall deformation threshold of the dam body. It is based on the benchmark rainfall deformation threshold T0, and introduces the water level change influence weight α and the water level rise rate in real-time data. As well as the rainfall impact weight β and the real-time rainfall intensity R, the impact of the two key factors of water level and rainfall on the dam deformation threshold is comprehensively considered.

[0047] The rainfall deformation threshold of the dam body is calculated based on the benchmark rainfall deformation threshold and real-time data. The benchmark threshold can be flexibly adjusted in combination with the current rainfall and the real-time status of the dam body to obtain a threshold that fits the current actual situation. Its role is to timely and accurately reflect the current deformation tolerance limit of the dam body, help detect safety hazards caused by rainfall, and use special model algorithms to dynamically associate historical and real-time data to make the threshold calculation more in line with the complex and changeable actual rainfall and dam operation conditions.

[0048] This approach presents complex, practical influencing factors in a concise and scientific mathematical form, breaking away from the traditional single-factor approach and integrating multiple factors to dynamically adjust thresholds. By setting weights, the impact of water level and rainfall on thresholds can be flexibly allocated based on the actual dam conditions. This significantly improves the accuracy and adaptability of rainfall deformation threshold calculations, providing more reliable and accurate quantitative indicators for dam safety monitoring in complex hydrological environments.

[0049] Further optimization, in step S5, a multi-level early warning linkage response mechanism for the dam body is established based on the deformation early warning situation and the rainfall deformation threshold, specifically including:

[0050] S5.1. Based on the deformation warning situation obtained in step S3 and the rainfall deformation threshold obtained in step S4, the dam body is graded for danger warning to obtain a multi-level deformation warning for the dam body; specifically,

[0051] If the actual deformation of a certain area of ​​the dam body approaches or exceeds the rainfall deformation threshold, and the deformation anomaly shows a rapid development trend, combined with other relevant indicators, the area is judged to be at a high risk level;

[0052] If the deformation increases but is far below the rainfall deformation threshold, and the abnormal situation is relatively stable, it is classified as a low-risk level;

[0053] S5.2. Construct a multi-level deformation response plan for the dam body based on multi-level deformation early warning;

[0054] S5.3. Construct a multi-level early warning linkage response mechanism for the dam body based on the multi-level deformation response plan, specifically:

[0055] When it is determined to be risk-free, the dam body will be subjected to real-time deformation monitoring;

[0056] When the risk is determined to be low, the dam body will be inspected for maintenance;

[0057] When it is determined to be a medium risk, the dam structure will be reinforced;

[0058] When it is determined to be a high risk, emergency rescue and disaster relief preparations will be carried out on the dam.

[0059] An intelligent early warning system for dam deformation risk, comprising:

[0060] The real-time monitoring module is used to monitor the dam body in real time based on the sensors deployed in the target dam area to obtain real-time data of the dam body;

[0061] Deformation curve generation module, used to construct theoretical deformation curves of the dam body at different water levels and temperatures;

[0062] The deformation warning module is used to perform physical-data dual-driven anomaly detection on real-time data and theoretical deformation curves to obtain deformation warning information of the dam body;

[0063] A rainfall deformation threshold calculation module is used to generate a rainfall deformation threshold of the dam body based on pre-acquired rainfall data;

[0064] The intelligent deformation warning module is used to build a multi-level early warning linkage response mechanism for the dam body based on the deformation warning situation and rainfall deformation threshold, and to perform intelligent deformation risk monitoring of the dam body based on the multi-level early warning linkage response mechanism.

[0065] Compared with the prior art, the present invention has the following beneficial effects:

[0066] 1. This invention significantly improves early warning accuracy through a dual-driven physics-data anomaly detection technology. First, theoretical deformation curves of the dam body under different water levels and temperatures are constructed. This is then combined with a finite element model to analyze its mechanical response, forming a precise deformation benchmark. This allows rapid identification of actual deformation risks caused by structural damage or abnormal loads, effectively eliminating false alarms caused by interfering factors such as temperature fluctuations and instrument errors.

[0067] 2. Utilizing pre-acquired rainfall data to generate rainfall deformation thresholds and constructing a multi-level early warning linkage response mechanism, the system can rapidly respond to varying risk levels. When abnormal dam deformation trends occur, the corresponding level of early warning can be quickly activated based on real-time data and pre-set mechanisms. This allows for timely information acquisition and action, significantly shortening the time from risk discovery to early warning activation. This ensures a prompt response to dam deformation risks, effectively improving the timeliness of early warnings. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 Schematic diagram of the process of the intelligent early warning method for dam deformation risk according to the present invention;

[0069] Figure 2 This is a functional module diagram of the intelligent early warning system for dam deformation risk described in the present invention. DETAILED DESCRIPTION

[0070] The realization of the purpose, functional features and advantages of the present invention will be further described in conjunction with embodiments and with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only used to illustrate the present invention and are not intended to limit the present invention.

[0071] Example 1:

[0072] Reference Figure 1 As shown, an intelligent early warning method for dam deformation risk includes:

[0073] S1. Deploy sensors in the target dam area, monitor the dam in real time based on the sensors, and obtain real-time data of the dam. Specifically including:

[0074] S1.1. Construct a sensor deployment plan according to the target dam area, and deploy sensors on the dam body based on the deployment plan to obtain a morphological sensing array of the dam body.

[0075] In this embodiment, sensors are embedded in geomagnetic beacons and magnetometers to obtain self-calibration sensors; multiple self-calibration sensors are arranged on the dam body according to a layout plan to obtain a morphological sensing array of the dam body.

[0076] The specific sensor layout plan is to comprehensively and meticulously construct a dam layout plan, from the selection of dam type (such as gravity dam, arch dam, earth-rock dam, etc.), the determination of the dam axis, to the precise planning of the dimensions of each part of the dam body, such as dam height, dam top width, dam bottom width, etc., to ensure that the plan not only meets engineering needs but also adapts to the natural conditions of the area and complies with relevant safety standards.

[0077] Specifically, based on the established dam layout plan, key monitoring locations of the dam body are determined, such as the dam foundation, dam abutments, different elevations inside the dam body, and the dam top, which are prone to deformation, stress concentration, or abnormal seepage.

[0078] S1.2. Constructing a data collection network for the dam body based on the morphological sensor array, and using the data collection network to collect real-time data on the dam body to obtain real-time data of the dam body. Constructing a data collection network for the dam body based on the morphological sensor array includes:

[0079] S1.2.1. Divide the morphological sensor array into grid nodes. Specifically, based on the constructed dam morphological sensor array, a reasonable division rule is adopted according to the geometric shape and size of the dam and the distribution of sensors.

[0080] For example, for a relatively regular rectangular dam, the morphological sensing array is divided horizontally and vertically into several uniformly sized grid areas at equal intervals. For irregularly shaped dams, the boundaries and density of the grid divisions must be flexibly determined, taking into account factors such as the dam's structural characteristics and the distribution of key monitoring areas, to ensure that each grid area contains at least one or more sensors that can effectively monitor the dam's condition.

[0081] The center or representative position of each grid area is set as a grid node, thereby dividing the entire morphological sensing array into a series of grid nodes with clear position identification in an orderly manner, providing a clear structural framework for subsequent data acquisition and processing.

[0082] S1.2.2. Collect relative difference data of the dam body based on network nodes to obtain relative difference data of the dam body.

[0083] Specifically, for each divided grid node, the real-time data collected by the sensors around the node is used.

[0084] For example, for displacement monitoring, the displacement value measured by each sensor at a certain initial moment is used as a benchmark. In the subsequent monitoring process, the difference between the displacement values ​​measured by different sensors at the same moment and the benchmark value is calculated. This difference is the relative difference.

[0085] For monitoring of other physical quantities such as stress and seepage, a similar method is also used. The initial measurement value is used as a reference to obtain the relative difference between the measurement value of each sensor and the initial value at different times.

[0086] By performing such relative difference calculations on the sensor data around each grid node, we can comprehensively collect the changes in the dam body relative to its initial state at different locations and times, and obtain relative difference data reflecting the dynamic changes in the dam body state. These data can more intuitively reflect the changing trends and degree differences in various parts of the dam body.

[0087] S1.2.3. Construct a data collection network for the dam body based on relative difference data and grid nodes.

[0088] Specifically, a database system is established to store relative difference data and various information related to grid nodes, such as node location coordinates, grid area number, associated sensor type and number, etc.

[0089] Using wireless communication technology, the relative difference data from each grid node is packaged into specially formatted data packets and sent out at set intervals. Each grid node is connected to a data aggregation center via wired or wireless communication links. The data aggregation center collects data packets from different grid nodes, performs preliminary data verification and collation, and then transmits the collated data to a database according to the established data transmission protocol.

[0090] In this way, with relative difference data as the core content and grid nodes as the basic units of data collection and transmission, a data collection network is constructed that can efficiently and accurately collect dam status change data, providing strong data support for subsequent dam status analysis and evaluation.

[0091] S2. Construct theoretical deformation curves of the dam body at different water levels and temperatures, including:

[0092] S2.1. Construct a finite element model of the dam body.

[0093] Specifically, comprehensive data on the dam body was collected, including precise geometric dimensions, detailed material properties (such as elastic modulus and Poisson's ratio), and structural design drawings. Using professional finite element modeling software, the dam body was discretized into a model composed of numerous finite elements by creating nodes and dividing elements based on its actual geometry.

[0094] S2.2. Preset the temperature-water level working condition of the dam body based on the pre-acquired dam body operation history and design standards to obtain the preset working condition of the dam body.

[0095] Specifically, the previously acquired historical data on dam operation is collated and analyzed, including records of water level changes, temperature fluctuations, and corresponding operating status information in different time periods over the years.

[0096] Refer to the design standards of the dam body and clarify key parameters such as the maximum water level, minimum water level, normal operating water level range and expected temperature change range specified in the design.

[0097] For example, consider the extreme working conditions of extreme high temperature and maximum water level occurring at the same time, as well as the common working conditions corresponding to different water levels within the normal operating temperature range. These preset working conditions are used to simulate the situations that the dam may face under different actual operating conditions, providing diverse input conditions for subsequent finite element analysis.

[0098] S2.3. Conduct finite element deformation analysis on the dam body based on the preset working conditions to obtain the theoretical deformation curves of the dam body at different water levels and temperatures.

[0099] Specifically, each temperature-water level working condition preset in the previous step is loaded one by one into the constructed finite element model of the dam body. Using the solver of the finite element analysis software, according to the principles of mechanics and heat conduction theory, the stress distribution inside the dam body and the resulting deformation under each working condition are calculated. For the calculation results of each working condition, the displacement data of the key parts of the dam body (such as the dam top, dam abutment, dam foundation, etc.) at different times are extracted. By sorting and analyzing these displacement data, the theoretical deformation curves of the dam body under different water level and temperature combinations are drawn. These curves intuitively show the deformation trend and degree of the dam body under various preset working conditions, providing an important theoretical basis for evaluating the safety and stability of the dam body.

[0100] S3: Perform physical-data dual-driven anomaly detection on the real-time data collected in step S1 and the theoretical deformation curve constructed in step S2 to obtain the deformation warning of the dam body. Specifically including:

[0101] S3.1. Draw the real-time deformation curve of the dam body based on the real-time data.

[0102] Specifically, real-time data is obtained from the data collection network of the dam body. These data contain information related to deformation, such as displacement and stress of different parts of the dam body.

[0103] Using professional data analysis and mapping software, a virtual model of the dam body is constructed based on the locations of each monitoring point and the corresponding deformation data recorded in the data. Then, by setting an appropriate coordinate system and scale, the real-time deformation displacement of each monitoring point is connected with lines to form a visual deformation image. From this image, a data series of the deformation changes over time at key locations (such as the dam crest, abutment, and foundation) is extracted. With time as the horizontal axis and the deformation as the vertical axis, a real-time deformation curve is drawn that intuitively reflects the real-time deformation of the dam body.

[0104] S3.2. Perform point-by-point compensation calculation on the real-time deformation curve and the theoretical deformation curve to obtain the initial residual value of the real-time data.

[0105] Specifically, the real-time deformation curve is compared with the theoretical deformation curve previously derived from finite element analysis. On the same time axis, the deformation corresponding to the same time point in each curve is subtracted point by point.

[0106] For example, at a certain moment, the deformation variable corresponding to the real-time deformation curve is X1, and the deformation variable corresponding to the theoretical deformation curve is X2. The difference between the two is calculated (X1-X2).

[0107] By performing such calculations at all time points, a series of difference data are obtained. These difference data constitute the initial residual value of the real-time data, which reflects the degree of difference between the actual real-time deformation of the dam body and the theoretical expected deformation.

[0108] S3.3. Correct the data type anomaly of the initial residual value to obtain the deformation residual value of the real-time data;

[0109] Specifically, the initial residual value data is thoroughly reviewed to check for data type anomalies caused by sensor failure, data transmission errors, etc. For example, there may be extremely large or small values ​​that are clearly outside the reasonable range, or the data type may not match the expected value (such as character data when it should be numeric).

[0110] For any abnormal data discovered, a data cleaning and correction algorithm is used, combining historical data from the dam body, data from surrounding monitoring points, and common sense physics to determine and correct it. If the initial residual value of a monitoring point is significantly larger than that of other adjacent monitoring points and exceeds the historical fluctuation range, a reasonable value can be estimated by weighted averaging the data from adjacent monitoring points and used as a replacement. After correcting for data type anomalies, more accurate and reliable real-time deformation residual values ​​are obtained, providing effective data support for subsequent analysis.

[0111] S3.4. Compare the physical anomalies of the dam body based on real-time data to obtain the deformation anomaly information of the dam body.

[0112] Specifically, the real-time data is reviewed again, and the current physical parameters of the dam body (such as displacement, stress, seepage, etc.) are compared with the pre-set normal physical parameter range.

[0113] The normal physical parameter range is determined based on the design standards of the dam body, historical operation data and similar engineering experience.

[0114] For example, if the real-time displacement of a part of the dam exceeds the maximum displacement allowed by the design, or the stress value exceeds the normal operating stress range, a physical anomaly is determined to have occurred in that part. By summarizing and organizing all information such as the location of all physical anomalies, the type of anomaly (displacement anomaly, stress anomaly, etc.), the degree of anomaly (specific values ​​exceeding the normal range), and the time of occurrence, the deformation anomaly information of the dam body is obtained. This information helps to accurately locate areas of the dam body that may pose safety risks.

[0115] S3.5. Construct deformation warning information of the dam body based on deformation residual value and deformation anomaly information.

[0116] Specifically, corresponding warning level standards are formulated based on the size and changing trend of the deformation residual value, as well as the severity and scope of the anomaly in the deformation anomaly information.

[0117] For example, if the deformation residual value is small and only slight physical abnormalities occur in individual parts, it can be set as a low-level warning; if the deformation residual value is large and serious physical abnormalities occur in multiple key parts, reaching or exceeding the set danger threshold, it will be set as a high-level warning.

[0118] With the warning level as the core, combined with specific abnormal locations, abnormal types and other information, a detailed dam deformation warning report is generated. The report clearly shows the current deformation warning status of the dam, providing a key basis for relevant personnel to take timely response measures to ensure the safe and stable operation of the dam.

[0119] S4. Calculate the rainfall deformation threshold of the dam body based on the pre-acquired rainfall data, specifically including:

[0120] S4.1. Generate a baseline rainfall deformation threshold for the dam body based on pre-acquired rainfall data.

[0121] Specifically, the pre-acquired historical rainfall data of the dam body should be comprehensively sorted out. These data should cover various rainfall event records with different rainfall amounts, rainfall durations and rainfall intensities.

[0122] The dam deformation data corresponding to these rainfall events were collected, including information on displacement and stress changes in different parts of the dam (such as the crest, abutment, and foundation). Statistical analysis methods were used to correlate rainfall and deformation data. For example, for different rainfall levels, the maximum and average deformation characteristics of each dam part were calculated.

[0123] Through in-depth analysis of extensive historical data, combined with the structural characteristics and mechanical properties of dams, we determined the general patterns of dam deformation under varying rainfall conditions. Based on these patterns, we established a reasonable range of dam deformation under normal rainfall conditions. This threshold, known as the baseline rainfall deformation threshold, provides a fundamental reference for determining whether dam deformation during real-time rainfall is abnormal.

[0124] S4.2. Based on the baseline rainfall deformation threshold and real-time data, calculate the rainfall deformation threshold of the dam body. The calculation formula is as follows:

[0125]

[0126] In the above formula, T is the rainfall deformation threshold of the dam body, T0 is the baseline rainfall deformation threshold, α is the influence weight of water level change, ΔH is the water level rise value in real-time data, and Δt is the time change value in real-time data. is the water level rising rate in real-time data, β is the rainfall impact weight, and R is the rainfall intensity in real-time data.

[0127] Specifically, first, real-time rainfall data is obtained from the dam's data collection network, including information such as the current rainfall amount, rainfall intensity, and rainfall duration. Real-time deformation data of the dam during the current rainfall process is also obtained, namely the displacement and stress changes at each monitoring point on the dam.

[0128] Then, the real-time rainfall data is compared and analyzed with the historical rainfall data characteristics based on which the benchmark rainfall deformation threshold is generated to determine the magnitude and type of the current rainfall situation.

[0129] For example, if the current rainfall intensity and accumulated rainfall are similar to a certain type of rainfall event in history, the benchmark rainfall deformation threshold range corresponding to this type of rainfall event will be used as a reference.

[0130] Combined with the real-time deformation data of the dam body, a special calculation model or algorithm is used to adjust the benchmark rainfall deformation threshold.

[0131] For example, if real-time monitoring shows that the deformation trend of a certain part of the dam body deviates significantly from the normal range based on historical data, and other monitoring points around this part also show similar abnormal trends, the benchmark rainfall deformation threshold will be corrected according to a certain weight ratio based on the degree of abnormality.

[0132] The α and β values ​​in the above formula were determined as follows: a Delphi expert panel consisting of senior experts in water conservancy engineering, structural safety monitoring, and data modeling was assembled to ensure comprehensive professional coverage. A structured questionnaire was designed, explicitly requiring experts to rate the rationality of the initial theoretical values ​​of α and β, the scientific nature of the logical relationship between the parameters and dam deformation, and the matching of the parameter weightings with actual engineering data (e.g., 1-5 points). Open-ended questions were also provided to collect information on adjustments.

[0133] Specifically, an anonymous questionnaire was distributed to the expert panel, requiring them to independently evaluate the initial values ​​of α and β and provide revision recommendations. After the questionnaires were collected, descriptive statistics (mean, standard deviation) and text analysis were used to identify key points of contention regarding parameter adjustment, generating a preliminary summary report. This preliminary summary report (including the distribution of anonymous expert opinions and key points of contention) was then provided to the expert panel for a second, anonymous feedback session. The panel was asked to reassess their opinions based on the panel's feedback and provide additional technical justification (e.g., citing clauses from the Technical Specifications for Safety Monitoring of Earth-Rockfill Dams to support weight revisions).

[0134] Then, repeat 2-3 rounds until the standard deviation of expert ratings is ≤ 0.1 (strong consensus) or more than 80% of experts accept the revised range for α and β. This culminates in a final iteration. Based on the final convergence results, the median or weighted mean of the expert consensus interval is used to determine the final values ​​for α and β, with the reasons for the minority's dissent noted. This final value is then substituted into historical disaster data (e.g., the "1998 flood monitoring record") for backtesting.

[0135] After comprehensively considering the benchmark threshold, real-time rainfall data and real-time deformation data, we finally obtained the rainfall deformation threshold applicable to the dam body under the current rainfall conditions. This threshold can more accurately reflect the deformation tolerance limit of the dam body under the current actual rainfall conditions, and provide a key indicator for timely detection of safety problems that may be caused by rainfall on the dam body.

[0136] S5. Based on the deformation warning situation and rainfall deformation threshold, a multi-level early warning linkage response mechanism is established for the dam body, and the deformation risk of the dam body is intelligently monitored based on the multi-level early warning linkage response mechanism. Specifically, it includes:

[0137] S5.1. Based on the deformation warning status obtained in step S3 and the rainfall deformation threshold obtained in step S4, the dam body is classified into a risk warning level to obtain a multi-level deformation warning of the dam body;

[0138] Specifically, a comprehensive assessment of dam deformation warnings was conducted, analyzing various information contained therein, such as the location of abnormal deformation, the degree of abnormality, development trends, and deviations from historical data. The rainfall deformation threshold was carefully examined to determine the deformation tolerance of the dam under current rainfall conditions. Based on pre-defined hazard classification criteria, a comprehensive comparison and analysis of the deformation warnings was conducted against the rainfall deformation threshold.

[0139] For example, if the actual deformation of a certain area of ​​the dam approaches or exceeds the rainfall deformation threshold, and the deformation anomaly shows a rapid development trend, combined with other relevant indicators, the area is judged to be at a high risk level. If the deformation increases but is far below the rainfall deformation threshold, and the anomaly is relatively stable, it is classified as a low risk level. Through such detailed assessment and judgment, different parts of the dam are divided into multiple levels according to the degree of risk, and a multi-level deformation warning covering all levels of risk is generated, clearly and intuitively presenting the risk distribution of the dam in its current state.

[0140] S5.2. Construct a multi-level deformation response plan for the dam body based on multi-level deformation early warning.

[0141] Specifically, the data collection network that has been built on the dam body is used to ensure the normal operation of various sensors in the morphological sensing array (such as displacement sensors, stress and strain sensors, etc.). According to the established monitoring frequency, real-time data from each monitoring point on the dam body is collected regularly, covering physical quantity information such as displacement, stress, and seepage. After data collection, it is quickly transmitted to the data processing center through the communication network. In the data processing center, professional data processing software is used to analyze and process the collected data in real time, draw the real-time deformation curve of the dam body, compare historical data with the normal operating parameter range, and continuously monitor the operating status of the dam body to ensure that any possible abnormal signs are discovered in time.

[0142] S5.3. Construct a multi-level early warning linkage response mechanism for the dam body based on the multi-level deformation response plan, specifically:

[0143] When it is determined to be risk-free, the dam body will be subjected to real-time deformation monitoring;

[0144] When the risk is judged to be low, the dam body will be inspected for maintenance.

[0145] Specifically, a professional maintenance and inspection team was organized to rush to the dam site with necessary testing tools and equipment.

[0146] Conduct a comprehensive inspection of the dam's exterior to check for abnormalities such as cracks, peeling, and signs of leakage, and record the location and condition of any problems found in detail. Inspect the dam's ancillary facilities, such as drainage systems and observation facilities, to check whether the drainage pipes are unobstructed and whether the observation instruments are functioning properly. Target the interior of the dam structure using non-destructive testing techniques, such as ultrasonic testing and geological radar testing, to inspect key areas (such as the dam foundation and abutments) to assess whether there are potential defects within the structure. Based on the inspection results, formulate a corresponding maintenance plan and promptly repair any minor problems found, such as filling tiny cracks and clearing debris from drainage pipes, to maintain the normal operation of the dam and prevent further escalation of risks.

[0147] When it is judged to be a medium risk, the dam structure will be reinforced.

[0148] Professional structural engineers will develop a detailed dam reinforcement plan based on the specific conditions of the dam and the risk assessment report. Specifically, they will determine the required materials and equipment, such as steel bars, concrete, and support structures, and organize a construction team to start construction.

[0149] For weak areas of the dam, such as large cracks or areas of stress concentration, reinforced concrete reinforcement is used. Cracks are first cleaned and pre-treated, followed by the insertion of rebar and the pouring of concrete to enhance the structure's bearing capacity. For areas significantly affected by lateral forces, such as the abutments, anti-slide piles or retaining walls can be added to improve the dam's stability. During construction, strict adherence to design plans and construction specifications is ensured, with enhanced quality supervision and safety management. The impact of reinforcement work on the dam structure is monitored in real time to ensure that the reinforcement effectively improves dam safety and reduces risk.

[0150] When a high-risk situation is identified, emergency rescue and disaster relief preparations will be carried out on the dam. The emergency response mechanism will be immediately activated, and an emergency rescue and disaster relief command center will be established to coordinate resources and actions from all parties.

[0151] A professional rescue team, including water conservancy engineering experts, engineering technicians, and fire rescue personnel, was quickly organized to rush to the dam site. A large amount of emergency relief supplies, such as sandbags, life-saving equipment, emergency lighting equipment, and engineering machinery, were deployed and stored on site. Personnel in potentially affected areas around the dam were urgently evacuated, and safety warning areas were established to ensure their safety. Close communication was maintained with meteorological authorities to maintain real-time monitoring of weather changes, particularly rainfall, flooding, and other meteorological information that could affect the safety of the dam. A detailed emergency rescue and disaster relief plan was developed, clarifying the division of labor and operational procedures for each rescue team. The team was prepared to respond to any deterioration of the dam's dangerous situation at any time, fully safeguarding the safety of people's lives and property and the overall safety of the dam.

[0152] Example 2:

[0153] like Figure 2 As shown, the intelligent early warning system 100 for dam deformation risk includes a real-time monitoring module 101, a deformation curve generation module 102, a deformation early warning module 103, a rainfall deformation threshold calculation module 104, and an intelligent deformation early warning module 105. A module in the present invention, also referred to as a unit, is a series of computer program segments that can be executed by an electronic device processor and perform fixed functions, and is stored in the electronic device's memory.

[0154] In this embodiment, the functions of each module / unit are as follows:

[0155] The real-time monitoring module 101 is used to monitor the dam body in real time based on sensors deployed in the target dam body area to obtain real-time data of the dam body.

[0156] The deformation curve generating module 102 is used to construct theoretical deformation curves of the dam body at different water levels and temperatures.

[0157] The deformation warning module 103 is used to perform physical-data dual-driven anomaly detection on the real-time data and the theoretical deformation curve to obtain the deformation warning status of the dam body.

[0158] The rainfall deformation threshold calculation module 104 is used to generate a rainfall deformation threshold of the dam body based on pre-acquired rainfall data.

[0159] The intelligent deformation warning module 105 is used to build a multi-level warning linkage response mechanism for the dam body based on the deformation warning situation and the rainfall deformation threshold, and to perform intelligent deformation risk monitoring of the dam body based on the multi-level warning linkage response mechanism.

[0160] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An intelligent early warning method for dam deformation risk, characterized in that: The steps include: S1. Deploy sensors in the target dam area, monitor the dam in real time based on the sensors, and collect real-time data of the dam; S2. Construct theoretical deformation curves of the dam body at different water levels and temperatures; S3, performing physical-data dual-driven anomaly detection on the real-time data collected in step S1 and the theoretical deformation curve constructed in step S2 to obtain deformation warning information of the dam body; S4. Calculating a rainfall deformation threshold of the dam body based on pre-acquired rainfall data; S5. A multi-level early warning linkage response mechanism for the dam body is established based on the deformation early warning situation and rainfall deformation threshold, and intelligent deformation risk monitoring of the dam body is carried out based on the multi-level early warning linkage response mechanism.

2. The intelligent early warning method for dam deformation risk according to claim 1, characterized in that: The step S1 specifically includes: S1.

1. Construct a sensor deployment plan based on the target dam area, and deploy sensors on the dam body based on the deployment plan to obtain a morphological sensor array for the dam body; S1.

2. Construct a data collection network for the dam body based on the morphological sensing array, and use the data collection network to collect real-time data on the dam body to obtain real-time data of the dam body.

3. The intelligent early warning method for dam deformation risk according to claim 2, characterized in that: In the step S1.1, a sensor is embedded in the geomagnetic beacon and the magnetometer to obtain a self-calibration sensor; and a plurality of self-calibration sensors are arranged on the dam body according to the arrangement plan to obtain a morphological sensing array of the dam body.

4. The intelligent early warning method for dam deformation risk according to claim 3, characterized in that: In step 1.2, a data collection network for the dam body is constructed based on the morphological sensing array, including: S1.2.

1. Divide the morphology sensing array into grid nodes. Specifically, based on the constructed dam morphology sensing array, divide the dam into several grid regions according to the dam's geometry, size, and sensor distribution. The center or representative location of each grid region is set as a grid node. The entire morphology sensing array is then orderly divided into a series of grid nodes with clear location identifiers. Each node corresponds to a unique number. S1.2.

2. Collect relative difference values ​​of the dam body based on network nodes to obtain relative difference data of the dam body. Specifically, for each divided grid node, use sensors deployed around the node to collect real-time data. Use the monitoring data at a certain initial moment as a benchmark. During subsequent monitoring, calculate the difference between the measurement values ​​of different sensors at the same moment and the corresponding benchmark value, and record it as the relative difference. S1.2.

3. Construct a data collection network for the dam body based on relative difference data and grid nodes.

5. The intelligent early warning method for dam deformation risk according to claim 1, characterized in that: In step S2, constructing a theoretical deformation curve of the dam body at different water levels and temperatures specifically includes: S2.

1. Construct a finite element model of the dam body; S2.

2. Preset the temperature-water level operating conditions of the dam body based on the pre-acquired dam body operation history and design standards to obtain the preset dam body operating conditions. Specifically, based on the pre-acquired historical data collected during the dam body operation, including water level change records, temperature fluctuations, and corresponding operating status information over different time periods over the years, and referring to the dam body design standards, clarify the maximum water level, minimum water level, normal operating water level range, and expected temperature change range of the dam body specified in the design. S2.

3. Perform finite element deformation analysis on the dam body based on the preset working conditions to obtain the theoretical deformation curves of the dam body at different water levels and temperatures. Specifically, load each temperature-water level working condition preset in the previous step into the constructed finite element model of the dam body one by one. Use the solver of the finite element analysis software to calculate the stress distribution inside the dam body and the resulting deformation under each working condition, and draw the theoretical deformation curves of the dam body under different water level and temperature combinations.

6. The intelligent early warning method for dam deformation risk according to claim 1, characterized in that: The step S3 specifically includes: S3.

1. Draw the real-time deformation curve of the dam body based on the real-time data; S3.

2. Perform point-by-point compensation calculation on the real-time deformation curve and the theoretical deformation curve to obtain the initial residual value of the real-time data; Specifically, the real-time deformation curve is compared with the theoretical deformation curve obtained based on finite element analysis. On the same time axis, the deformation corresponding to the same time point in each curve is subtracted point by point. By performing the above operation on all time points, a series of difference data are obtained. These difference data constitute the initial residual value of the real-time data. S3.

3. Correct the data type anomalies of the initial residual values ​​to obtain the deformation residual values ​​of the real-time data. Specifically, the initial residual value data is comprehensively screened to identify any data type anomalies caused by sensor failure, data transmission errors, etc. The data cleaning and correction algorithms are then applied to the discovered abnormal data, combined with historical data of the dam body, data from surrounding monitoring points, and common sense physics to make judgments and corrections. S3.

4. Compare the physical anomalies of the dam body based on real-time data to obtain abnormal deformation information of the dam body; S3.

5. Construct deformation warning information of the dam body based on deformation residual value and deformation anomaly information.

7. The intelligent early warning method for dam deformation risk according to claim 1, characterized in that: The step S4 specifically includes: S4.

1. Generate a baseline rainfall deformation threshold for the dam body based on pre-acquired rainfall data. Specifically, first, sort through the pre-acquired historical rainfall data for the dam body and select various rainfall event records with different rainfall amounts, rainfall durations, and rainfall intensities; Secondly, the corresponding deformation data of the dam body during these rainfall events were collected. Using statistical analysis methods, the rainfall data and deformation data were correlated and analyzed to determine the general pattern of dam body deformation under different rainfall conditions. Based on this pattern, the reasonable range boundary value of the dam body deformation caused by rainfall changes under normal conditions was set, namely the baseline rainfall deformation threshold of the dam body. S4.

2. Based on the baseline rainfall deformation threshold and real-time data, calculate the rainfall deformation threshold of the dam body. The calculation formula is as follows: In the above formula, T is the rainfall deformation threshold of the dam body, T0 is the baseline rainfall deformation threshold, α is the influence weight of water level change, ΔH is the water level rise value in real-time data, and Δt is the time change value in real-time data. is the water level rising rate in real-time data, β is the rainfall impact weight, and R is the rainfall intensity in real-time data.

8. The intelligent early warning method for dam deformation risk according to claim 1, characterized in that: In step S5, a multi-level early warning linkage response mechanism for the dam body is established based on the deformation early warning situation and the rainfall deformation threshold, specifically including: S5.

1. Based on the deformation warning situation obtained in step S3 and the rainfall deformation threshold obtained in step S4, the dam body is graded for danger warning to obtain a multi-level deformation warning for the dam body; specifically, If the actual deformation of a certain area of ​​the dam body approaches or exceeds the rainfall deformation threshold, and the deformation anomaly shows a rapid development trend, combined with other relevant indicators, the area is judged to be at a high risk level; If the deformation increases but is far below the rainfall deformation threshold, and the abnormal situation is relatively stable, it is classified as a low-risk level; S5.

2. Construct a multi-level deformation response plan for the dam body based on multi-level deformation early warning; S5.

3. Construct a multi-level early warning linkage response mechanism for the dam body based on the multi-level deformation response plan, specifically: When it is determined to be risk-free, the dam body will be subjected to real-time deformation monitoring; When the risk is determined to be low, the dam body will be inspected for maintenance; When it is determined to be a medium risk, the dam structure will be reinforced; When it is determined to be a high risk, emergency rescue and disaster relief preparations will be carried out on the dam.

9. An intelligent early warning system for dam deformation risk, characterized in that: include: The real-time monitoring module is used to monitor the dam body in real time based on the sensors deployed in the target dam area to obtain real-time data of the dam body; Deformation curve generation module, used to construct theoretical deformation curves of the dam body at different water levels and temperatures; The deformation warning module is used to perform physical-data dual-driven anomaly detection on real-time data and theoretical deformation curves to obtain deformation warning information of the dam body; A rainfall deformation threshold calculation module is used to generate a rainfall deformation threshold of the dam body based on pre-acquired rainfall data; The intelligent deformation warning module is used to build a multi-level early warning linkage response mechanism for the dam body based on the deformation warning situation and rainfall deformation threshold, and to perform intelligent deformation risk monitoring of the dam body based on the multi-level early warning linkage response mechanism.

Images