Data-driven dam safety state analysis method and system

Abstract

The invention belongs to the technical field of intelligent hydropower, and discloses a data-driven dam safety state analysis method and system, and the method comprises the steps: obtaining dam operation data, and carrying out the preprocessing of the dam operation data; verifying the preprocessed dam operation data, and filtering out invalid or abnormal data; obtaining a real-time index, an accuracy index and an integrity index according to the verified dam operation data, and carrying out weighted calculation on the real-time index, the accuracy index and the integrity index to obtain a comprehensive safety index; according to the comprehensive safety index, an Echarts engine is adopted for real-time rendering to obtain a dam safety state thermodynamic diagram, visualization and intelligent warning of the dam safety state are achieved, accurate and efficient decision support is provided for dam safety management, and the problems that in the prior art, a dam safety management evaluation method is low in efficiency, high in subjectivity and high in reliability are solved. And dynamic behaviors and long-term performance changes of the dam are difficult to comprehensively reflect.

Description

Technical Field

[0001] This invention belongs to the field of intelligent hydropower technology and relates to a data-driven method and system for analyzing the safety status of dams. Background Technology

[0002] As the core hub of water conservancy projects, dams play an irreplaceable role in flood control and disaster reduction, water resource allocation, power generation, and irrigation. Their safety management level is directly related to the implementation of the national water security strategy and the protection of people's well-being. With the acceleration of the modernization of water conservancy projects, the limitations of the traditional dam safety management evaluation system have become increasingly prominent, becoming a significant bottleneck restricting the upgrading of governance capabilities and making it difficult to adapt to the needs of refined management under complex operating conditions.

[0003] The lag of existing evaluation methods is primarily reflected in their over-reliance on human experience. In the traditional model, safety assessments largely depend on on-site inspections and subjective judgments by technical personnel. This not only consumes a significant amount of manpower and time but is also prone to biased evaluation results due to differences in individual experience. When faced with hidden problems such as dam crack development and abnormal seepage, manual inspections struggle to detect subtle changes, often missing the optimal intervention window.

[0004] The complexity of the evaluation process further exacerbates the management dilemma. The current system involves dozens of parameters, including dam material properties, structural mechanical performance, and hydrogeological conditions, requiring cross-validation through multiple models. The calculation process is cumbersome and lacks sufficient correlation analysis. Grassroots units often struggle to efficiently process massive amounts of data due to limited technical reserves, resulting in excessively long evaluation cycles and an inability to provide timely support for decision-making.

[0005] The contradiction between static evaluation logic and the dynamic characteristics of dams is particularly prominent. Existing methods are mostly based on the assumption of fixed operating conditions, treating the dam body as a static structure for analysis, ignoring the long-term cumulative effects of dynamic factors such as water level fluctuations, temperature stress, and material aging. For example, the carbonation process of concrete dams progresses slowly over time, and traditional evaluation models struggle to quantify its gradual impact on structural strength, easily leading to misjudgments of safety margins.

[0006] Fragmented standards and a lack of tools hinder the effectiveness of evaluation. Different regions and types of dams employ varying evaluation indicators and methods, lacking a unified technical framework, resulting in poor comparability of safety conditions across projects. Furthermore, the scarcity of systematic analysis tools makes it difficult to deeply integrate multi-source data, failing to meet the safety evaluation needs of modern water conservancy projects. Therefore, an intelligent evaluation system adapted to the requirements of the new era is urgently needed. Summary of the Invention

[0007] The purpose of this invention is to address the problems in existing dam safety management evaluation methods, which rely heavily on manual inspection and experience-based judgment, resulting in low efficiency and strong subjectivity; and that existing methods are mostly based on static or simplified assumptions, making it difficult to comprehensively reflect the dynamic behavior and long-term performance changes of dams. This invention provides a data-driven method and system for dam safety status analysis.

[0008] To achieve the above objectives, the present invention employs the following technical solution:

[0009] This invention provides a data-driven method for dam safety status analysis, comprising:

[0010] Acquire dam operation data and preprocess the dam operation data;

[0011] Verify the preprocessed dam operation data and filter out invalid or abnormal data;

[0012] Based on the verified dam operation data, real-time indicators, accuracy indicators, and integrity indicators are obtained, and the real-time indicators, accuracy indicators, and integrity indicators are weighted and calculated to obtain a comprehensive safety index.

[0013] Based on the comprehensive safety index, a heat map of the dam's safety status was generated in real time using the Echarts engine.

[0014] Optionally, the dam operation data is acquired through a sensor network, edge computing nodes, and industrial IoT gateways deployed at the dam site;

[0015] The preprocessing of dam operation data includes real-time filtering / denoising, cleaning, and aggregation of the dam operation data through edge computing nodes.

[0016] Optionally, the verification of the preprocessed dam operation data includes verifying whether the preprocessed dam operation data is an empty data packet, whether the CRC check of the data passes, and whether the timestamp of the data exceeds the current processing window range, thereby filtering out invalid or abnormal data.

[0017] Optionally, the real-time performance index is calculated as follows:

[0018]

[0019] The total number of time periods is the total amount of data that should theoretically be collected within a fixed time window, the number of missing time periods is the amount of data that was not successfully collected / transmitted within the same time window, λ is the time decay factor, and Δt is the difference between the current time and the data collection time.

[0020] Optionally, the method for calculating the accuracy index includes:

[0021] A sensor cross-validation matrix is ​​established, and an LSTM-AE is used to construct a data reconstruction model for anomaly detection. The anomaly degree Ac is defined and calculated as follows:

[0022]

[0023] Among them, X 异常 This is abnormal data, X 重构 To reconstruct the data, σ represents the standard deviation.

[0024] Optionally, the method for calculating the integrity index includes:

[0025] Multidimensional completeness testing is performed across spatial and temporal dimensions, missing data is compensated for, and information entropy is used for evaluation. The calculation method is as follows:

[0026] H(x) = -∑p(x) i logp(x) i )

[0027]

[0028] Where H(x) is the information entropy of a single set of data integrity analysis, and data x i This refers to the valid data collected from various devices within the dam system; p(x i H represents the percentage of valid data collected from various devices in the dam system. max The maximum information entropy;

[0029] Optionally, the method for weighting the real-time performance index, accuracy index, and integrity index to obtain the comprehensive security index employs an improved AHP-entropy weighting method, as follows:

[0030] W total =α×W AHP +(1-α)×W Entropy

[0031] Among them, W total For the comprehensive safety index, W AHP W represents the weights of the AHP (Analytic Hierarchy Process). Entropy The entropy weight is the weight of the entropy weight method, and α is the expert experience coefficient, which is dynamically adjusted from 0.4 to 0.6.

[0032] The evaluation result R is:

[0033] R = [N,C,V] = [dam system, {Rt,Ac,Cp}, [v1,v2,v3]]

[0034] Where N represents the dam system, C represents the real-time performance index, accuracy index, and integrity index, V represents the normalized value of the index, and v1, v2, v3 are the normalized values ​​of Rt, Ac, and Cp, respectively.

[0035] Substituting the normalized values ​​into the membership function yields the membership degree, and combining this with the weights of each indicator yields the safety level index. The method is as follows:

[0036] The security level is:

[0037]

[0038] Among them, w i As the weight of each indicator, u i (v i ) represents the membership degree corresponding to the normalized values ​​of each indicator;

[0039] W Entropy The calculation method is as follows:

[0040]

[0041] Among them, E i E is the information entropy of a certain indicator at multiple time points, used to assess the uncertainty of that indicator over a time series. i (Rt) is the information entropy of Rt, E i (Ac) is the information entropy of Ac, E i (Cp) is the information entropy of Cp; E i The calculation method is as follows:

[0042]

[0043] Where, p ij For a given test point, the data from several days of data records represents the standardized value of the i-th data point and the j-th sample. k is the normalization coefficient used to calculate the information entropy, and its calculation method is as follows:

[0044]

[0045] Where m is the total number of data samples collected during that period, and 1 Hz is the frequency of 86,400 samples collected per day.

[0046] A data-driven dam safety status analysis system includes:

[0047] The data acquisition layer is used to acquire dam operation data and preprocess the dam operation data;

[0048] The stream processing layer is used to validate the pre-processed dam operation data and filter out invalid or abnormal data.

[0049] The stream processing layer is used to obtain real-time indicators, accuracy indicators, and integrity indicators based on the verified dam operation data, and to perform weighted calculations on the real-time indicators, accuracy indicators, and integrity indicators to obtain a comprehensive safety index.

[0050] The visualization and alarm layer is used to render a heat map of the dam's safety status in real time using the Echarts engine based on the comprehensive safety index.

[0051] A terminal device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the steps of the method described above.

[0052] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method described above.

[0053] Compared with the prior art, the present invention has the following beneficial effects:

[0054] This invention provides a data-driven method for dam safety status analysis. It acquires dam operation data and preprocesses it. The preprocessed data is then validated, filtering out invalid or abnormal data. Based on the validated data, real-time indicators, accuracy indicators, and completeness indicators are obtained. These indicators are then weighted and calculated to obtain a comprehensive safety index, enabling dynamic and objective safety assessment. Finally, a heatmap is rendered in real-time using Echarts to visualize the safety status and provide intelligent alarms. This improves data processing efficiency, evaluation accuracy, and completeness, providing precise and efficient decision support for dam safety management.

[0055] This invention also provides a data-driven dam safety status analysis system. This system achieves the acquisition of dam operation data and preprocesses the data through a high degree of integration of a data acquisition layer, a stream processing layer, and a visualization and alarm layer. The system verifies the preprocessed dam operation data, filtering out invalid or abnormal data. Based on the verified dam operation data, it acquires real-time indicators, accuracy indicators, and completeness indicators, and performs weighted calculations on these indicators to obtain a comprehensive safety index and a final real-time rendered heatmap. The system ensures the reliability of the input data through a real-time verification mechanism in stream processing. The combination of three-dimensional indicator calculation and a fuzzy evaluation model enables a multi-dimensional and dynamic assessment of dam safety status, overcoming the limitations of single indicators. The heatmap visualization and hierarchical response mechanism transform complex data into intuitive decision-making basis, shortening the risk handling cycle. It possesses both versatility and adaptability, and can be flexibly adjusted according to different dam characteristics, providing an efficient and intelligent implementation path for dam safety management and promoting the transformation of evaluation work from experience-driven to data-driven. Attached Figure Description

[0056] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 This is a schematic diagram of a data-driven dam safety status analysis method according to the present invention.

[0058] Figure 2 This is a schematic diagram of the structure of a data-driven dam safety status analysis system according to the present invention;

[0059] Figure 3 This is a logical schematic diagram of a data-driven dam safety status analysis system according to the present invention. Detailed Implementation

[0060] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0061] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0062] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0063] In the description of the embodiments of the present invention, it should be noted that if terms such as "upper," "lower," "horizontal," or "inner" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product of the invention is in use, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. Furthermore, terms such as "first" and "second" are only used to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0064] Furthermore, the use of the term "horizontal" does not imply that the component must be absolutely horizontal, but rather that it can be slightly tilted. For example, "horizontal" simply means that its direction is more horizontal than "vertical," and does not mean that the structure must be completely horizontal, but can be slightly tilted.

[0065] In the description of the embodiments of the present invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in the present invention according to the specific circumstances.

[0066] The present invention will now be described in further detail with reference to the accompanying drawings:

[0067] See Figure 1 This invention discloses a data-driven method for dam safety status analysis, comprising:

[0068] S1: Acquire dam operation data and preprocess the dam operation data, specifically:

[0069] The dam operation data is acquired through a sensor network, edge computing nodes, and industrial IoT gateways deployed at the dam site.

[0070] The preprocessing of dam operation data includes real-time filtering / denoising, cleaning, and aggregation of the dam operation data through edge computing nodes.

[0071] S2: Verify the preprocessed dam operation data, filtering out invalid or abnormal data. Specifically:

[0072] The verification of the preprocessed dam operation data includes verifying whether the preprocessed dam operation data is an empty data packet, whether the CRC check of the data passes, and whether the timestamp of the data exceeds the current processing window range, and filtering out invalid or abnormal data.

[0073] S3: Based on the verified dam operation data, obtain the real-time performance index, accuracy index, and integrity index, and perform a weighted calculation on the real-time performance index, accuracy index, and integrity index to obtain the comprehensive safety index, specifically:

[0074] The calculation method for the real-time performance index is as follows:

[0075]

[0076] The total number of time periods is the total amount of data that should theoretically be collected within a fixed time window, the number of missing time periods is the amount of data that was not successfully collected / transmitted within the same time window, λ is the time decay factor, and Δt is the difference between the current time and the data collection time.

[0077] The calculation method for the accuracy index includes:

[0078] A sensor cross-validation matrix is ​​established, and an LSTM-AE is used to construct a data reconstruction model for anomaly detection. The anomaly degree Ac is defined and calculated as follows:

[0079]

[0080] Among them, X 异常 This is abnormal data, X 重构 To reconstruct the data, σ represents the standard deviation.

[0081] The calculation method for the integrity index includes:

[0082] Multidimensional completeness testing is performed across spatial and temporal dimensions, missing data is compensated for, and information entropy is used for evaluation. The calculation method is as follows:

[0083] H(x) = -∑p(x) i logp(x) i )

[0084]

[0085] Where H(x) is the information entropy of a single set of data integrity analysis, and data x i This refers to the valid data collected from various devices within the dam system; p(x i H represents the percentage of valid data collected from various devices in the dam system. max The maximum information entropy;

[0086] S4: Based on the comprehensive safety index, a heat map of the dam's safety status is obtained in real time using the Echarts engine. The method is as follows:

[0087] W total =α×W AHP +(1-α)×W entropy

[0088] Among them, W total For the comprehensive safety index, W AHP W represents the weights of the AHP (Analytic Hierarchy Process). Entropy The entropy weight is the weight of the entropy weight method, and α is the expert experience coefficient, which is dynamically adjusted from 0.4 to 0.6.

[0089] The evaluation result R is:

[0090] R = [N,C,V] = [dam system, {Rt,Ac,Cp}, [v1,v2,v3]]

[0091] Where N represents the dam system, C represents the real-time performance index, accuracy index, and integrity index, V represents the normalized value of the index, and v1, v2, v3 are the normalized values ​​of Rt, Ac, and Cp, respectively.

[0092] Substituting the normalized values ​​into the membership function yields the membership degree, and combining this with the weights of each indicator yields the safety level index. The method is as follows:

[0093] The security level is:

[0094] μ=∑w i ×u i (v i )

[0095] Among them, w i As the weight of each indicator, u i (v i ) represents the membership degree corresponding to the normalized values ​​of each indicator;

[0096] W Entropy The calculation method is as follows:

[0097]

[0098] Among them, E iE is the information entropy of a certain indicator at multiple time points, used to assess the uncertainty of that indicator over a time series. i (Rt) is the information entropy of Rt, E i (Ac) is the information entropy of Ac, E i (Cp) is the information entropy of Cp; E i The calculation method is as follows:

[0099] E i =-k∑p ij ×lnp ij

[0100] Where, p ij For a given test point, the data from several days of data records represents the standardized value of the i-th data point and the j-th sample. k is the normalization coefficient used to calculate the information entropy, and its calculation method is as follows:

[0101]

[0102] Where m is the total number of data samples collected during that period, and 1 Hz is the frequency of 86,400 samples collected per day.

[0103] See Figure 2 and Figure 3 The present invention also provides a data-driven dam safety status analysis system, comprising:

[0104] The data acquisition layer is used to acquire dam operation data and preprocess the dam operation data;

[0105] The stream processing layer is used to validate the pre-processed dam operation data and filter out invalid or abnormal data.

[0106] The stream processing layer is used to obtain real-time indicators, accuracy indicators, and integrity indicators based on the verified dam operation data, and to perform weighted calculations on the real-time indicators, accuracy indicators, and integrity indicators to obtain a comprehensive safety index.

[0107] The visualization and alarm layer is used to render a heat map of the dam's safety status in real time using the Echarts engine based on the comprehensive safety index.

[0108] See Figure 3 The data acquisition layer includes a sensor network, edge computing nodes, and an industrial IoT gateway deployed at the dam site; the sensor network performs real-time filtering / denoising processing on the collected dam operation data through the edge computing nodes, and transmits it to the stream processing layer through the industrial IoT gateway;

[0109] The stream processing layer includes an Apache Kafka message queue, a Flink stream processing engine, and a data verification module. It receives the data stream transmitted from the data acquisition layer, partitions it using the Apache Kafka message queue, and then inputs it into the Flink stream processing engine for real-time data cleaning and aggregation. Simultaneously, the data verification module verifies whether the data is an empty packet, whether the CRC check passes, and whether the timestamp exceeds the current processing window range, filtering out invalid or abnormal data. The output is then transmitted to the analysis layer through the data verification module.

[0110] The analysis layer receives data transmitted from the stream processing layer through the API gateway, calculates the comprehensive security index, stores the calculation results in the time series database, and pushes the security index to the visualization and alarm layer.

[0111] The visualization and alarm layer calls the calculation results of the analysis layer through the RESTful API and uses the Echarts engine to render the heat map of the dam's safety status in real time.

[0112] The sensor network includes displacement sensors, seepage sensors, stress sensors, and environmental quantity sensors.

[0113] The analysis layer includes: a real-time calculation module, an accuracy assessment module, an integrity compensation module, and a comprehensive evaluation engine;

[0114] The real-time calculation module, accuracy assessment module, and integrity compensation module respectively calculate the real-time index, accuracy index, and integrity index of the received data;

[0115] The comprehensive security index is obtained by weighting the real-time performance index, accuracy index, and integrity index based on the comprehensive evaluation engine.

[0116] The real-time calculation module is a sliding window processor, the accuracy evaluation module is an LSTM-AE anomaly detector, the integrity compensation module is an STL / GAN data generator, and the comprehensive evaluation engine uses AHP-entropy weighting method combined with a fuzzy evaluation model.

[0117] The calculation method for the real-time performance index is as follows:

[0118]

[0119] The total number of time periods is the total amount of data that should theoretically be collected within a fixed time window, the number of missing time periods is the amount of data that was not successfully collected / transmitted within the same time window, λ is the time decay factor, and Δt is the difference between the current time and the data collection time.

[0120] The calculation of the accuracy index includes: establishing a sensor cross-validation matrix, using LSTM-AE to construct a data reconstruction model for anomaly detection, defining the anomaly degree Ac, and calculating it as follows:

[0121]

[0122] Among them, X 异常 This is abnormal data, X 重构 To reconstruct the data, σ represents the standard deviation;

[0123] Reconstructed data refers to data that is reconstructed from the input sensor data and whose characteristics match those of the original data, after the data reconstruction model built by LSTM-AE processes the input sensor data.

[0124] The calculation of the integrity index includes: performing multi-dimensional completeness detection in spatial and temporal dimensions, compensating for missing data, and using information entropy evaluation. The calculation method is as follows:

[0125] H(x) = -∑p(x) i logp(x) i )

[0126]

[0127] Where H(x) is the information entropy of a single set of data integrity analysis, used to calculate a certain data integrity index, data x i This refers to the valid data collected from various devices within the dam system; p(x i H represents the percentage of valid data collected from various devices in the dam system. max The maximum information entropy;

[0128] The weighting calculation employs a modified AHP-entropy weighting method, with the following formula:

[0129] W total =α×W AHP +(1-α)×W Entropy

[0130] Among them, W total W is the comprehensive weight vector. AHP The weights for the AHP (Analytic Hierarchy Process) are calculated by constructing a judgment matrix through expert scoring and then calculating the weights from the eigenvectors, reflecting the domain expert experience; W Entropy The entropy weight is the weight, and α is the expert experience coefficient, which is dynamically adjusted from 0.4 to 0.6. When α is close to 1, it indicates reliance on expert experience; when α is close to 0, it indicates reliance on the statistical characteristics of the data.

[0131] The comprehensive evaluation engine uses AHP-entropy weighting method combined with a fuzzy evaluation model, and the evaluation result R is:

[0132] R = [N,C,V] = [dam system, {Rt,Ac,Cp}, [v1,v2,v3]]

[0133] Where N represents the dam system, C represents the real-time performance index, accuracy index, and integrity index, V represents the normalized value of the index, and v1, v2, v3 are the normalized values ​​of Rt, Ac, and Cp, respectively.

[0134] v is calculated using the core algorithm of the three indicators Rt, Ac, and Cp. The v value is then substituted into the membership function (such as a trapezoidal function) of the corresponding indicator to obtain its membership degree u to each security level (e.g., red / orange / yellow / blue). i (v i Then combine the weight w of each indicator. i This yields the safety level indicators for dam operation.

[0135] Substituting the normalized values ​​into the membership function yields the membership degree, and combining this with the weights of each indicator yields the safety level index. The method is as follows:

[0136] The security level is:

[0137] μ=∑w i ×u i (v i )

[0138] Among them, w i As the weight of each indicator, u i (v i ) represents the membership degree corresponding to the normalized values ​​of each indicator.

[0139] The W Entropy The calculation method is as follows:

[0140]

[0141] Among them, E i E is the information entropy of a certain indicator at multiple time points, used to assess the uncertainty of that indicator over a time series. i (Rt) is the information entropy of Rt, E i (Ac) is the information entropy of AC, E i (Cp) is the information entropy of Cp; E i The calculation method is as follows:

[0142] E i =-k∑p ij ×lnp ij

[0143] Where, p ijFor a given test point, the data from several days of data records represents the standardized value of the i-th data point and the j-th sample. k is the normalization coefficient used to calculate the information entropy, and its calculation method is as follows:

[0144]

[0145] Where m is the total number of data samples collected during that period, and 1 Hz is the frequency of 86,400 samples collected per day.

[0146] H(x) is the information entropy of a single set of data integrity analysis, used to calculate a certain data integrity index.

[0147] E i It is the information entropy of a certain indicator at multiple time points, and it assesses the uncertainty of a certain indicator over a time series.

[0148] Both are essentially the same, but they describe the information entropy of the data layer and the indicator layer respectively, belonging to different levels. For example, measuring height 100 times a day, the amount of valid data from these 100 measurements falls within the scope of H(x) evaluation. However, the real-time performance and accuracy of height data over 10 consecutive days, etc., fall within the scope of E. i The scope of the assessment.

[0149] For example, in a data analysis, based on past experience and expert evaluation, Rt has a weight of 0.5 with respect to Ac, meaning it is slightly less important; Rt has a weight of 3 with respect to Cp, meaning it is relatively important; and Ac has a weight of 4 with respect to Cp, meaning it is very important. Therefore, the judgment matrix can be listed as follows:

[0150]

[0151] The i-th row represents the weight of the i-th parameter with respect to Rt, Ac, and Cp. The diagonal lines indicate that each parameter has a weight of 1 for itself. Calculating the eigenvectors of the matrix yields W. AHP = [0.297, 0.540, 0.163]. This refers to the weights of the three parameters used in the comprehensive judgment.

[0152] Then calculate the entropy weight W of the data. Entropy First, input the 30-day data record for this measurement point, labeling each data point as p. ijThis refers to the standardized value of the j-th sample from the i-th data point. Standardization transforms data of varying sizes and units into values ​​between 0 and 1, eliminating the influence of units on weights. For example, in real-time analysis, if the standard time delay is between 0.2s and 5s, for a time duration t, the standardization method is (t-t_min) / (t_max-t_min). In this case, it's (t-0.2) / (5-0.2). Thus, a data point with a delay of 0.8s, after standardization, has a value of (0.8-0.2) / (5-0.2) = 0.125.

[0153] Let Rt be the first data point, Ac be the second data point, and Cp be the third data point. If the Rt value of the node data on the fourth day is 0.5, then it is represented as p. 14 It is 0.5.

[0154] The time-series database is InfluxDB.

[0155] The evaluation criteria for the real-time performance indicators are as follows:

[0156] When Rt≥0.95, the evaluation index is excellent, the data transmission delay is less than 1 second, and the sampling frequency is not less than 1 Hz;

[0157] When 0.8≤Rt<0.95, the evaluation index is qualified, the data transmission delay is less than 5 seconds, and the sampling frequency is not less than 0.2 Hz;

[0158] When Rt < 0.8, the evaluation metric is risk, triggering the data link self-check procedure;

[0159] The evaluation criteria for the accuracy index are as follows:

[0160] High accuracy (Ac≥0.99): Data residuals <3σ, requiring passing the KS test;

[0161] Medium accuracy (0.95≤Ac<0.99): Data residuals <5σ, manual verification required;

[0162] Low accuracy (Ac < 0.95): Initiate sensor calibration procedure;

[0163] The evaluation criteria for the integrity index are as follows:

[0164] High integrity (Cp≥0.9): missing rate <5%, spatial coverage ≥95%;

[0165] Medium completeness (0.7≤Cp<0.9): Missing rate 5-15%, initiate supplemental monitoring;

[0166] Low integrity (Cp<0.7): Trigger emergency monitoring plan.

[0167] This invention provides a terminal device comprising: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps in the various method embodiments described above. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the various device embodiments described above.

[0168] The computer program can be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention.

[0169] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0170] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0171] The memory can be used to store the computer program and / or module. The processor implements various functions of the terminal device by running or executing the computer program and / or module stored in the memory and calling the data stored in the memory.

[0172] If the modules / units integrated into the terminal device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0173] This method ensures the reliability of input data through a real-time verification mechanism in stream processing; the combination of three-dimensional index calculation and fuzzy evaluation model enables a multi-dimensional and dynamic assessment of dam safety status, overcoming the limitations of single indicators; heat map visualization and a graded response mechanism transform complex data into intuitive decision-making basis, shortening the risk management cycle. This invention combines versatility and adaptability, and can be flexibly adjusted according to different dam characteristics, providing an efficient and intelligent implementation path for dam safety management and promoting the transformation of evaluation work from experience-driven to data-driven.

[0174] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A data-driven method for dam safety status analysis, characterized in that, include: Acquire dam operation data and preprocess the dam operation data; Verify the preprocessed dam operation data and filter out invalid or abnormal data; Based on the verified dam operation data, real-time indicators, accuracy indicators, and integrity indicators are obtained, and the real-time indicators, accuracy indicators, and integrity indicators are weighted and calculated to obtain a comprehensive safety index. Based on the comprehensive safety index, a heat map of the dam's safety status was generated in real time using the Echarts engine.

2. The data-driven dam safety status analysis method according to claim 1, characterized in that, The dam operation data is acquired through a sensor network, edge computing nodes, and industrial IoT gateways deployed at the dam site. The preprocessing of dam operation data includes real-time filtering / denoising, cleaning, and aggregation of the dam operation data through edge computing nodes.

3. The data-driven dam safety status analysis method according to claim 1, characterized in that, The verification of the preprocessed dam operation data includes verifying whether the preprocessed dam operation data is an empty data packet, whether the CRC check of the data passes, and whether the timestamp of the data exceeds the current processing window range, and filtering out invalid or abnormal data.

4. The data-driven dam safety status analysis method according to claim 1, characterized in that, The calculation method for the real-time performance index is as follows: The total number of time periods is the total amount of data that should theoretically be collected within a fixed time window, the number of missing time periods is the amount of data that was not successfully collected / transmitted within the same time window, λ is the time decay factor, and Δt is the difference between the current time and the data collection time.

5. The data-driven dam safety status analysis method according to claim 1, characterized in that, The calculation method for the accuracy index includes: A sensor cross-validation matrix is ​​established, and an LSTM-AE is used to construct a data reconstruction model for anomaly detection. The anomaly degree Ac is defined and calculated as follows: Among them, X 异常 This is abnormal data, X 重构 To reconstruct the data, σ represents the standard deviation.

6. The data-driven dam safety status analysis method according to claim 1, characterized in that, The calculation method for the integrity index includes: Multidimensional completeness testing is performed across spatial and temporal dimensions, missing data is compensated for, and information entropy is used for evaluation. The calculation method is as follows: H(x)=-∑p(x i )logp(x i ) Where H(x) is the information entropy of a single set of data integrity analysis, and data x i This refers to the valid data collected from various devices within the dam system; p(x) i H represents the percentage of valid data collected from various devices in the dam system. max This represents the maximum information entropy.

7. The data-driven dam safety status analysis method according to claim 1, characterized in that, The method for weighting real-time performance, accuracy, and integrity indicators to obtain the comprehensive security index employs an improved AHP-entropy weighting method. The method is as follows: W total =α×W AHP +(1-a)×W Entropy Among them, W total For the comprehensive safety index, W AHP W represents the weights of the AHP (Analytic Hierarchy Process). Entropy The entropy weight is the weight of the entropy weight method, and α is the expert experience coefficient, which is dynamically adjusted from 0.4 to 0.

6. The evaluation result R is: R = [N,C,V] = [dam system, {Rt,Ac,Cp}, [v1,v2,v3]] Where N represents the dam system, C represents the real-time performance index, accuracy index, and integrity index, V represents the normalized value of the index, and v1, v2, v3 are the normalized values ​​of Rt, Ac, and Cp, respectively. Substituting the normalized values ​​into the membership function yields the membership degree, and combining this with the weights of each indicator yields the safety level index. The method is as follows: The security level is: μ=Σw i ×u i (v. i ) Among them, w i As the weight of each indicator, u i (v i ) represents the membership degree corresponding to the normalized values ​​of each indicator; W Entropy The calculation method is as follows: Among them, E i E is the information entropy of a certain indicator at multiple time points, used to assess the uncertainty of that indicator over a time series. i (Rt) is the information entropy of Rt, E i (Ac) is the information entropy of AC, E i (Cp) is the information entropy of Cp; E i The calculation method is as follows: AND i =-k∑p ij ×lnp ij Where, p ij For a given test point, the data from several days of data records represents the standardized value of the i-th data point and the j-th sample. k is the normalization coefficient used to calculate the information entropy, and its calculation method is as follows: Where m is the total number of data samples collected during that period, and 1 Hz is the frequency of 86,400 samples collected per day.

8. A data-driven dam safety status analysis system, characterized in that, include: The data acquisition layer is used to acquire dam operation data and preprocess the dam operation data; The stream processing layer is used to validate the pre-processed dam operation data and filter out invalid or abnormal data. The stream processing layer is used to obtain real-time indicators, accuracy indicators, and integrity indicators based on the verified dam operation data, and to perform weighted calculations on the real-time indicators, accuracy indicators, and integrity indicators to obtain a comprehensive safety index. The visualization and alarm layer is used to render a heat map of the dam's safety status in real time using the Echarts engine based on the comprehensive safety index.

9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1-7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1-7.

Images