Earth and rockfill dam intelligent monitoring risk early warning system

Abstract

The invention belongs to the field of earth and rockfill dam monitoring risk early warning, and discloses an earth and rockfill dam intelligent monitoring risk early warning system. The system comprises an earth and rockfill dam risk evaluation system module, a monitoring data extraction and preprocessing module, a risk factor scoring algorithm module based on monitoring data, a subjective and objective combination weight module, a two-dimensional cloud model algorithm module, a two-dimensional cloud visualization module and an early warning module. The method is more real-time, objective and accurate, and facilitates timely understanding of the risk state of the dam and early warning.

Description

Technical Field

[0001] The present invention belongs to the field of earth-rock dam monitoring risk early warning, and in particular relates to an earth-rock dam intelligent monitoring risk early warning system. Background Art

[0002] Earth-rockfill dams, due to their readily available materials and adaptability to terrain, have become a widely used dam type in water conservancy projects. Their safety and stability are directly linked to the safety of people and property downstream, as well as the ecological environment. However, over the long term, earth-rockfill dams can accumulate risks due to complex factors such as material aging, seepage damage, and seismic loads, posing a threat to the stability and safety of the dam. Given the complexity and uncertainty of earth-rockfill dam operation, and the constant changes in risk status, there is an urgent need for an intelligent monitoring and dynamic risk warning system that can monitor the operating status of earth-rockfill dams in real time and issue timely warnings.

[0003] Currently, the practical application of earth-rockfill dam risk early warning technology faces several pressing challenges. First, the risk factor scoring process primarily relies on expert scoring. This approach not only results in a certain lag in risk assessments but also, due to differences in expert judgment, can introduce a degree of subjectivity in the scoring results. Second, a single weighting approach presents significant limitations. While subjective weighting can reflect expert experience and knowledge, it can easily introduce human bias. While objective weighting is based on the statistical characteristics of data, it can overlook the actual project context and specific circumstances, resulting in weighting results that fail to truly reflect the importance of risk factors. Furthermore, traditional one-dimensional cloud models, faced with complex, multi-dimensional uncertainties in dam risk assessments, often struggle to fully and accurately reflect risk characteristics and cannot meet the high-precision requirements of modern earth-rockfill dam risk early warning.

[0004] Therefore, in view of the shortcomings of the above technologies, it is urgent to propose a two-dimensional cloud model that scores risk factors based on monitoring data and assigns weights to them to realize an intelligent monitoring dynamic risk warning system for earth-rock dams. Summary of the Invention

[0005] In order to solve the problems existing in the prior art, the present invention provides an intelligent monitoring risk early warning system for earth-rock dams, which realizes dynamic risk early warning of earth-rock dam intelligent monitoring based on monitoring data, combined weighting and two-dimensional cloud model.

[0006] To achieve the above object, the present invention provides the following solutions:

[0007] An intelligent monitoring and risk early warning system for earth-rock dams, comprising: an earth-rock dam risk assessment system module, a monitoring data extraction and preprocessing module, a risk factor scoring algorithm module based on monitoring data, a subjective and objective combined weight module, a two-dimensional cloud model algorithm module, a two-dimensional cloud visualization module, and an early warning module;

[0008] The earth-rock dam risk assessment system module is used to establish a corresponding risk assessment system based on monitoring factors; wherein the earth-rock dam risk assessment system includes a target layer, a criterion layer, and a factor layer;

[0009] The monitoring data extraction and preprocessing module is used to establish an information database based on the risk assessment system, obtain monitoring data from the information database and preprocess the monitoring data;

[0010] The risk factor scoring algorithm module based on monitoring data is used to obtain the scoring results of each monitoring element in the factor layer based on the preprocessed monitoring data;

[0011] The subjective and objective combined weight module is used to select the subjective weight and objective weight algorithm according to the scoring results, calculate the combined weight to obtain the weight of each monitoring element at the factor layer and the criterion layer;

[0012] The two-dimensional cloud model algorithm module is used to establish a two-dimensional cloud model from the two dimensions of risk probability and consequence based on the weights of each monitoring element in the factor layer and the criterion layer;

[0013] The two-dimensional cloud visualization module is used to generate a risk cloud map based on the two-dimensional cloud model and the digital features of the standard cloud, the evaluation cloud, and the comprehensive cloud;

[0014] The early warning module is used to determine the risk level of the earth-rock dam based on the risk cloud map and the proximity and to make corresponding early warnings.

[0015] Preferably, the information database stores dam monitoring data collected from multiple data sources;

[0016] The monitoring data include: reservoir water level, reservoir water temperature, rainfall, horizontal displacement, vertical displacement, crack joints;

[0017] The preprocessing of the monitoring data includes: eliminating gross errors in the monitoring data and filling in missing monitoring data.

[0018] Preferably, the risk factor scoring algorithm module based on monitoring data includes: an abnormality rate unit, a time component unit, a variation range unit, a development trend unit, a unification unit, and a scoring unit;

[0019] The anomaly rate unit is used to introduce a mathematical regression model to obtain a residual sequence, and then use the DBSCAN clustering algorithm to analyze the residual sequence to determine the outliers in the dam monitoring data;

[0020] The time-sensitive component unit is used to calculate the period of the monitoring data using the FFT algorithm and pass it as an input parameter to the STL decomposition algorithm, automatically extract the time-sensitive component of the data, use different types of time-sensitive component models for regression fitting, select a model that meets the preset requirements, and finally analyze and judge the trend of the time-sensitive component based on the rate of change;

[0021] The variation range unit is used to quantify the variation range of the monitoring data through the data range;

[0022] The development trend unit is used to determine the slope of the monitoring data using a slope index method, and quantify the development trend of the monitoring data using the absolute value of the slope;

[0023] The unified unit is used to unify the dimensions of the four indicators;

[0024] The scoring unit is used to determine the proportional weights of the consequences and probabilities of risk factors with respect to the abnormality rate, timeliness component, variation range and development trend in the system, and obtain the consequence and probability scores of the risk factors.

[0025] Preferably, determining abnormal values in dam monitoring data includes:

[0026]

[0027] Where A is the abnormal rate, Y i and Z i are the number of abnormal data and total data in the monitoring data during the study period;

[0028] The range of data used to quantify the fluctuation of monitoring data includes:

[0029] R=X max -X min ;

[0030] Where, X max and X min are the maximum and minimum values of the monitoring data respectively;

[0031] Unify the dimensions of the four indicators, including:

[0032]

[0033] In the formula, F is the index standardized data, f is the index value, and f max is the maximum value among the indicators, f min is the minimum value among the indicators;

[0034] The system determines the consequences and probabilities of risk factors with respect to the proportion weights of abnormality rate, timeliness component, variation range, and development trend, and obtains the consequence and probability scores of risk factors, including:

[0035]

[0036] Where Z is the composite value, w i is the weight of index i, x i is the value of indicator i, i is the abnormal rate, timeliness component, variation range and development trend of the monitoring data.

[0037] Preferably, the subjective and objective combined weight module includes:

[0038]

[0039] Where W i is the combined weight, α i is the subjective weight, β i is the objective weight.

[0040] Preferably, in the two-dimensional cloud model algorithm module, the two-dimensional cloud model includes: a standard cloud, an evaluation cloud and a comprehensive cloud;

[0041] The standard cloud computing formula includes:

[0042]

[0043] Where, and are the expectation, standard deviation and super entropy of the risk probability standard cloud, and are the expectation, standard deviation and super entropy of the risk consequence standard cloud, Q max , Q max are the maximum and minimum values of the risk level standard interval respectively;

[0044] The cloud computing evaluation formula includes:

[0045]

[0046] Where n is the number of samples, x is m 、y m are the mth indicator values of risk probability and risk consequence, respectively; Ex, Enx, Hex are the expectation, standard deviation, and super entropy of the risk probability evaluation cloud, respectively; Ey, Eny, Hey are the expectation, standard deviation, and super entropy of the risk consequence evaluation cloud, respectively;

[0047] The comprehensive cloud computing formula includes:

[0048]

[0049] where Ex′, Enx′ and Hex′ are the expectation, entropy and super entropy of the risk probability comprehensive cloud, Ey′, Eny′ and Hey′ are the expectation, entropy and super entropy of the risk consequence comprehensive cloud, ω1, ω2, …, ω n are the weights of each indicator respectively.

[0050] Preferably, the two-dimensional cloud visualization module includes: a calculation unit, a generation unit;

[0051] The calculation unit is used to calculate the digital features (Ex, Ey, Enx, Eny, Hex, Hey) of the standard cloud, the evaluation cloud, and the comprehensive cloud in sequence according to the two-dimensional cloud model;

[0052] The generating unit is used to generate a risk cloud map based on the digital features of the standard cloud, the evaluation cloud, and the comprehensive cloud.

[0053] Preferably, in the early warning module, the closeness calculation formula includes:

[0054]

[0055] Where Ex′ and Ey′ respectively represent the risk probability and expected value of risk consequences of the integrated cloud; Standardize the risk probability and expected value of risk consequences of cloud respectively.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] 1. The proposed intelligent monitoring dynamic risk warning system for earth-rock dams can acquire and process various monitoring data of earth-rock dams in real time. By deeply mining the monitoring data, it objectively and promptly reflects the current status of the dam, avoiding the lag and subjectivity of traditional expert scoring of factor layers.

[0058] 2. The proposed intelligent monitoring dynamic risk warning system for earth-rock dams adopts a combination of subjective and objective weighting, which can more accurately determine the importance of risk factors, improve the accuracy of weighting, and provide a reliable basis for dam safety management.

[0059] 3. The proposed intelligent monitoring dynamic risk warning system for earth-rock dams adopts a two-dimensional cloud model to overcome the limitations of traditional one-dimensional cloud models in complex multi-dimensional risk assessment, and conducts a comprehensive and systematic assessment of the risk status of earth-rock dams.

[0060] 4. The proposed intelligent monitoring dynamic risk warning system for earth-rock dams facilitates timely understanding of the risk status of the dam. The system can quickly issue warnings, buy time for taking corresponding emergency measures, and effectively prevent and reduce potential safety accidents.

[0061] 5. The proposed intelligent monitoring dynamic risk warning system for earth-rock dams realizes the intelligence and automation of earth-rock dam risk warning, improves work efficiency, reduces labor costs, and reduces the possibility of human errors.

[0062] 6. The proposed intelligent monitoring dynamic risk warning system for earth-rock dams can better serve engineering practice, provide scientific technical support for the safe operation and maintenance of earth-rock dams, help ensure the safety and stability of water conservancy projects, and have high engineering application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0064] Figure 1 A structural connection diagram of the intelligent monitoring and risk warning system for earth-rock dams provided by an embodiment of the present invention;

[0065] Figure 2 A flow chart of monitoring factor scoring provided by an embodiment of the present invention;

[0066] Figure 3 Provides standard cloud and level 2 comprehensive cloud maps for embodiments of the present invention;

[0067] Figure 4 Provides standard cloud and level 1 comprehensive cloud map for embodiments of the present invention.

[0068] In the attached figure, 100 is an earth-rock dam risk assessment system module; 200 is a monitoring data extraction and preprocessing module; 300 is a risk factor scoring algorithm module based on monitoring data; 400 is a subjective and objective combined weight module; 500 is a two-dimensional cloud model algorithm module; 600 is a two-dimensional cloud visualization module; and 700 is an early warning module. DETAILED DESCRIPTION

[0069] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0070] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0071] Example 1

[0072] Currently, the scoring of risk factors mainly relies on inviting experts to score, which is lagging and subjective. The single weighting method is one-sided. When conducting dam risk assessment, the traditional one-dimensional cloud model faces complex and multi-dimensional uncertainty problems, and it is difficult to fully and accurately reflect the characteristics of risks. It cannot meet the high-precision requirements of modern earth-rock dam risk warning.

[0073] like Figure 1 As shown, the present invention addresses the existing problems of subjectivity and lag in risk factor scoring, the difficulty of single weighting in accurately reflecting the importance of risk factors, and the unsuitability of traditional one-dimensional cloud models for risk assessment of complex, multi-dimensional earth-rockfill dams. The present invention proposes an intelligent monitoring and dynamic risk warning system for earth-rockfill dams. Compared with existing technologies, the present invention is more real-time, objective, and accurate, facilitating timely understanding of dam risk status and issuing warnings. The system comprises: an earth-rockfill dam risk assessment system module 100, a monitoring data extraction and preprocessing module 200, a risk factor scoring algorithm module 300 based on monitoring data, a subjective and objective combined weighting module 400, a two-dimensional cloud model algorithm module 500, a two-dimensional cloud visualization module 600, and an early warning module 700.

[0074] The earth-rock dam risk assessment system module 100 is used to establish a corresponding risk assessment system based on monitoring factors; wherein the earth-rock dam risk assessment system includes a target layer, a criterion layer, and a factor layer;

[0075] The monitoring data extraction and preprocessing module 200 is used to establish an information database according to the risk assessment system, obtain monitoring data from the information database and preprocess the monitoring data;

[0076] The risk factor scoring algorithm module 300 based on monitoring data is used to obtain the scoring result of each monitoring element in the factor layer based on the pre-processed monitoring data;

[0077] The subjective and objective combined weight module 400 is used to select subjective weight and objective weight algorithm according to the scoring results, calculate the combined weight to obtain the weight of each monitoring element at the factor layer and the criterion layer;

[0078] The two-dimensional cloud model algorithm module 500 is used to establish a two-dimensional cloud model from the two dimensions of risk probability and consequence based on the weights of each monitoring element at the factor layer and the criterion layer;

[0079] A two-dimensional cloud visualization module 600 is used to generate a risk cloud map based on the two-dimensional cloud model and the digital features of the standard cloud, the evaluation cloud, and the comprehensive cloud;

[0080] The early warning module 700 is used to determine the risk level of the earth-rock dam based on the risk cloud map and the proximity and to issue corresponding early warnings.

[0081] In this embodiment, the information database stores dam monitoring data collected from multiple data sources;

[0082] Monitoring data include: reservoir water level, reservoir water temperature, rainfall, horizontal displacement, vertical displacement, crack joints, etc.;

[0083] The preprocessing of monitoring data includes: eliminating gross errors in monitoring data and filling in missing monitoring data.

[0084] In this embodiment, the risk factor scoring algorithm module 300 based on monitoring data includes: an abnormality rate unit, a time component unit, a change amplitude unit, a development trend unit, a unification unit, and a scoring unit;

[0085] The anomaly rate unit is used to introduce a mathematical regression model to obtain a residual sequence, and then use the DBSCAN clustering algorithm to analyze the residual sequence to determine the outliers in the dam monitoring data;

[0086] The time-sensitive component unit is used to calculate the period of the monitoring data using the FFT algorithm and pass it as an input parameter to the STL decomposition algorithm to automatically extract the time-sensitive components of the data. Different types of time-sensitive component models are used for regression fitting, and the best model is selected. Finally, the trend of the time-sensitive components is analyzed based on the rate of change.

[0087] The variation range unit is used to quantify the variation range of monitoring data through data range;

[0088] The development trend unit is used to determine the slope of the monitoring data using the slope index method and quantify the development trend of the monitoring data using the absolute value of the slope;

[0089] Unified unit, used to unify the dimensions of the four indicators;

[0090] The scoring unit is used to determine the consequences and probabilities of risk factors in the system in relation to the proportion weights of abnormality rate, timeliness component, variation range and development trend, and obtain the consequences and probability scores of risk factors.

[0091] Specifically, the monitoring factor scoring process is as follows: Figure 2 As shown in Table 1, the risk probability level and risk consequence level scores of some monitoring factors within 5 years are shown in Table 1 and Table 2.

[0092] The mathematical regression model is introduced to obtain the residual sequence, and then the DBSCAN clustering algorithm is used to analyze the residual sequence to determine the outliers in the dam monitoring data. The anomaly rate formula includes:

[0093]

[0094] Where A is the abnormal rate, Y i and Z iare the number of abnormal data and total data in the monitoring data during the study period, respectively.

[0095] The FFT algorithm is used to calculate the period of the monitoring data and pass it as an input parameter to the STL decomposition algorithm to automatically extract the time-sensitive components of the data. Different types of time-sensitive component models are used for regression fitting, and the best model is selected. Finally, the trend of the time-sensitive components is analyzed based on the rate of change, as shown in Table 3.

[0096] Among them, the time-dependent component model can be summarized into logarithmic function form, hyperbolic function form, polynomial function form and exponential function form. The time-dependent component is extracted by FFT and STL algorithm. The polynomial function form is fitted by least squares method, and the other forms are fitted by nonlinear least squares method. Finally, the complex correlation coefficient R of the fitting results is compared. 2 and root mean square error RMSE to determine the best model.

[0097] The range of data is used to quantify the fluctuation of monitoring data. The calculation formula of the range includes:

[0098] R=X max -X min

[0099] Where, X max and X min They are the maximum and minimum values of the monitoring data respectively.

[0100] The slope index method is used to determine the slope of the monitoring data, and the absolute value of the slope is used to quantify the development trend of the monitoring data.

[0101] The four indicators (abnormal rate, timeliness component, change range and development trend) are unified in dimension and the calculation formulas are standardized as follows:

[0102]

[0103] In the formula, F is the index standardized data, f is the index value, and f max is the maximum value among the indicators, f min is the minimum value among the indicators.

[0104] In the system, the consequences and probabilities of risk factors are determined in proportion to the abnormality rate, timeliness component, variation range, and development trend, and the consequence and probability scores of risk factors are obtained using the following formula.

[0105]

[0106] Where Z is the composite value, w i is the weight of index i, x i is the value of indicator i, i is the abnormal rate, timeliness component, variation range and development trend of the monitoring data.

[0107] Table 1

[0108]

[0109] Table 2

[0110]

[0111] Table 3

[0112]

[0113]

[0114] In this embodiment, the subjective and objective combined weight module 400 selects the subjective weight and objective weight algorithm to calculate the subjective weight α i , calculate the objective weight β i Among them, the subjective weight algorithm includes: hierarchical analysis method, fuzzy hierarchical analysis method, improved hierarchical analysis method (IAHP) to calculate subjective weight α i The objective weight algorithms include: CRITIC, entropy weight method, grey correlation method, and the objective weight β i The combined weight W is obtained based on the subjective weight and objective weight. i .

[0115] Subjective weight α i The calculation formula includes:

[0116] (1) Analytical Hierarchy Process (1 to 9 scale, CR test required) i The calculation formula includes:

[0117]

[0118] Where x ij Indicates the relative importance of factor i relative to factor j.

[0119] (2) Fuzzy analytic hierarchy process α i The calculation formula includes:

[0120]

[0121] Where x ij Indicates the relative importance of factor i relative to factor j.

[0122] (3) Improved analytic hierarchy process (1 to 3 scales, no need for CR test) i The calculation formula includes:

[0123]

[0124] Where x ij Indicates the relative importance of factor i relative to factor j.

[0125] Objective weight β i The calculation formula includes

[0126] (1) CRITIC method β i The calculation formula includes:

[0127]

[0128] Where, is the mean value of the i-th indicator data, r ij Represents the correlation coefficient between evaluation indicators i and j.

[0129] (2) Entropy weight method β i The calculation formula includes:

[0130]

[0131] Where, e i Represents the entropy value of the i-th indicator.

[0132] (3) Grey correlation method β i The calculation formula includes:

[0133]

[0134] Where s ij is the grey relational coefficient.

[0135] Combination weight W i The calculation formula includes:

[0136]

[0137] Where, α i is the subjective weight, β i is the objective weight.

[0138] In this embodiment, the two-dimensional cloud model algorithm module 500 establishes a two-dimensional cloud model from two dimensions: risk occurrence probability and consequence.

[0139] In the two-dimensional cloud model algorithm module, the two-dimensional cloud model includes: standard cloud, evaluation cloud and comprehensive cloud;

[0140] The standard cloud computing formula includes:

[0141]

[0142] Where, and are the expectation, standard deviation and super entropy of the risk probability standard cloud, and are the expectation, standard deviation and super entropy of the risk consequence standard cloud, Q max , Q max are the maximum and minimum values of the risk level standard interval respectively;

[0143] The cloud computing evaluation formula includes:

[0144]

[0145] Where n is the number of samples, x is m 、y m are the mth indicator values of risk probability and risk consequence, respectively; Ex, Enx, Hex are the expectation, standard deviation, and super entropy of the risk probability evaluation cloud, respectively; Ey, Eny, Hey are the expectation, standard deviation, and super entropy of the risk consequence evaluation cloud, respectively;

[0146] The comprehensive cloud computing formula includes:

[0147]

[0148] where Ex′, Enx′ and Hex′ are the expectation, entropy and super entropy of the risk probability comprehensive cloud, Ey′, Eny′ and Hey′ are the expectation, entropy and super entropy of the risk consequence comprehensive cloud, ω1, ω2, …, ω n are the weights of each indicator respectively.

[0149] The digital characteristics of the standard cloud for the risk probability and consequence level of earth-rockfill dams are shown in Table 4. The combined weight is determined by calculating the IAHP-CRITIC combined weight using the geometric mean method. The cloud model digital characteristics of the evaluation cloud, level 2 comprehensive cloud, and level 1 comprehensive cloud are then calculated using the evaluation cloud formula and the comprehensive cloud formula. The digital characteristics of the risk probability level and risk consequence level clouds of some monitoring elements are shown in Tables 5 and 6.

[0150] CRITIC (Criteria Importance Through Intercriteria Correlation) is generally referred to as the CRITIC method in the literature. IAHP, also known as the Improved Analytic Hierarchy Process (AHP), is a combination of the aforementioned weighting methods, with the subjective weightings being IAHP and the objective weightings being CRITIC.

[0151] Table 4

[0152]

[0153] Table 5

[0154]

[0155] Table 6

[0156]

[0157]

[0158] In this embodiment, the two-dimensional cloud visualization module 600 includes: a calculation unit, a generation unit;

[0159] A calculation unit, used to calculate the digital features (Ex, Ey, Enx, Eny, Hex, Hey) of the standard cloud, the evaluation cloud, and the comprehensive cloud in sequence according to the two-dimensional cloud model;

[0160] The generation unit is used to generate a risk cloud map based on the digital features of the standard cloud, evaluation cloud, and comprehensive cloud. Specifically, the risk cloud map is constructed based on the digital features of the standard cloud, evaluation cloud, and comprehensive cloud (the risk cloud map can be formed by inputting the digital features (Ex, Ey, Enx, Eny, Hex, Hey) and the number of cloud droplets) to achieve two-dimensional cloud visualization. The comparison results of the comprehensive cloud and the standard cloud of each criterion layer indicator are as follows: Figure 3 As shown in the figure, the comparison results of the target layer index comprehensive cloud and standard cloud are as follows: Figure 4 As shown in Figure 1, A represents the risk assessment of earth-rock dams, and B1, B2, B3, and B4 represent the environmental quantity, deformation, seepage, and stress-strain monitoring elements of earth-rock dams, respectively.

[0161] In this embodiment, the early warning module 700 determines the dam risk level including "extremely high risk", "high risk", "medium risk" and "low risk" according to the risk consequences and risk probability in accordance with SL829-2024 "Guidelines for the Classification and Assessment of Reservoir Dam Risk Levels", which are expressed as Level I, Level II, Level III and Level IV, corresponding to red warning, orange warning, yellow warning and blue warning, as shown in Table 7.

[0162] Table 7

[0163]

[0164] The calculation formula of the proximity algorithm in the early warning module includes:

[0165]

[0166] Where Ex′ and Ey′ respectively represent the risk probability and expected value of risk consequences of the integrated cloud; Standardize the risk probability and expected value of risk consequences of cloud respectively.

[0167] The results of the closeness calculation are shown in Table 8.

[0168] Table 8

[0169]

[0170] This invention improves the objectivity and real-time nature of risk factor scoring by quantifying risk factors based on monitoring data. It also employs a combined weighting approach to comprehensively consider both subjective and objective factors, enhancing the scientificity and rationality of weight assignment. Furthermore, it introduces a two-dimensional cloud model, enhancing the ability of traditional models to handle complex, multi-dimensional uncertainty. This enables real-time, dynamic monitoring of the operating status of earth-rockfill dams and risk warnings, providing a more reliable technical guarantee for their safe operation.

[0171] The embodiments described above are merely descriptions of preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by persons skilled in the art should fall within the scope of protection defined by the claims of the present invention.

Claims

1. An intelligent monitoring and risk warning system for earth-rock dams, characterized by: The system includes: an earth-rock dam risk assessment system module, a monitoring data extraction and preprocessing module, a risk factor scoring algorithm module based on monitoring data, a subjective and objective combined weight module, a two-dimensional cloud model algorithm module, a two-dimensional cloud visualization module, and an early warning module; The earth-rock dam risk assessment system module is used to establish a corresponding risk assessment system based on monitoring factors; wherein the earth-rock dam risk assessment system includes a target layer, a criterion layer, and a factor layer; The monitoring data extraction and preprocessing module is used to establish an information database based on the risk assessment system, obtain monitoring data from the information database and preprocess the monitoring data; The risk factor scoring algorithm module based on monitoring data is used to obtain the scoring results of each monitoring element in the factor layer based on the preprocessed monitoring data; The subjective and objective combined weight module is used to select the subjective weight and objective weight algorithm according to the scoring results, calculate the combined weight to obtain the weight of each monitoring element at the factor layer and the criterion layer; The two-dimensional cloud model algorithm module is used to establish a two-dimensional cloud model from the two dimensions of risk probability and consequence based on the weights of each monitoring element in the factor layer and the criterion layer; The two-dimensional cloud visualization module is used to generate a risk cloud map based on the two-dimensional cloud model and the digital features of the standard cloud, the evaluation cloud, and the comprehensive cloud; The early warning module is used to determine the risk level of the earth-rock dam based on the risk cloud map and the proximity and to make corresponding early warnings.

2. The system according to claim 1, wherein: The information database is used to store dam monitoring data collected from multiple data sources; The monitoring data include: reservoir water level, reservoir water temperature, rainfall, horizontal displacement, vertical displacement, crack joints; The preprocessing of the monitoring data includes: eliminating gross errors in the monitoring data and filling in missing monitoring data.

3. The system according to claim 1, wherein: The risk factor scoring algorithm module based on monitoring data includes: an abnormality rate unit, a time component unit, a change amplitude unit, a development trend unit, a unification unit, and a scoring unit; The anomaly rate unit is used to introduce a mathematical regression model to obtain a residual sequence, and then use the DBSCAN clustering algorithm to analyze the residual sequence to determine the outliers in the dam monitoring data; The time-sensitive component unit is used to calculate the period of the monitoring data using the FFT algorithm and pass it as an input parameter to the STL decomposition algorithm, automatically extract the time-sensitive component of the data, use different types of time-sensitive component models for regression fitting, select a model that meets the preset requirements, and finally analyze and judge the trend of the time-sensitive component based on the rate of change; The variation range unit is used to quantify the variation range of the monitoring data through the data range; The development trend unit is used to determine the slope of the monitoring data using a slope index method, and quantify the development trend of the monitoring data using the absolute value of the slope; The unified unit is used to unify the dimensions of the four indicators; The scoring unit is used to determine the proportional weights of the consequences and probabilities of risk factors with respect to the abnormality rate, timeliness component, variation range and development trend in the system, and obtain the consequence and probability scores of the risk factors.

4. The system according to claim 3, characterized in that Identify outliers in dam monitoring data, including: Where A is the abnormal rate, Y i and Z i are the number of abnormal data and total data in the monitoring data during the study period; The range of data used to quantify the fluctuation of monitoring data includes: R=X max -X min ; Where, X max and X min are the maximum and minimum values of the monitoring data respectively; Unify the dimensions of the four indicators, including: In the formula, F is the index standardized data, f is the index value, and f max is the maximum value among the indicators, f min is the minimum value among the indicators; The system determines the consequences and probabilities of risk factors with respect to the proportion weights of abnormality rate, timeliness component, variation range, and development trend, and obtains the consequence and probability scores of risk factors, including: Where Z is the composite value, w i is the weight of index i, x i is the value of the standardized index i, and i is the abnormal rate, timeliness component, variation range and development trend of the monitoring data.

5. The system according to claim 1, wherein: The subjective and objective combined weight module includes: Where W i is the combined weight, α i is the subjective weight, β i is the objective weight.

6. The system according to claim 1, wherein: In the two-dimensional cloud model algorithm module, the two-dimensional cloud model includes: standard cloud, evaluation cloud and comprehensive cloud; The standard cloud computing formula includes: Where, and are the expectation, standard deviation and super entropy of the risk probability standard cloud, and are the expectation, standard deviation and super entropy of the risk consequence standard cloud, Q max , Q max are the maximum and minimum values of the risk level standard interval respectively; The cloud computing evaluation formula includes: Where n is the number of samples, x m 、y m are the mth indicator values of risk probability and risk consequence, respectively; Ex, Enx, Hex are the expectation, standard deviation, and super entropy of the risk probability evaluation cloud, respectively; Ey, Eny, Hey are the expectation, standard deviation, and super entropy of the risk consequence evaluation cloud, respectively; The comprehensive cloud computing formula includes: where Ex′, Enx′, and Hex′ are the expectation, entropy, and super entropy of the risk probability comprehensive cloud, respectively; Ey′, Eny′, and Hey′ are the expectation, entropy, and super entropy of the risk consequence comprehensive cloud, respectively. n are the weights of each indicator respectively.

7. The system according to claim 6, characterized in that The two-dimensional cloud visualization module includes: a calculation unit and a generation unit; The calculation unit is used to calculate the digital features (Ex, Ey, Enx, Eny, Hex, Hey) of the standard cloud, the evaluation cloud, and the comprehensive cloud in sequence according to the two-dimensional cloud model; The generating unit is used to generate a risk cloud map based on the digital features of the standard cloud, the evaluation cloud, and the comprehensive cloud.

8. The system according to claim 1, wherein: In the early warning module, the closeness calculation formula includes: Where Ex′ and Fy′ respectively represent the risk probability and expected value of risk consequences of the integrated cloud; Standardize the risk probability and expected value of risk consequences of cloud respectively.

Images