Dam displacement early warning method based on multi-dimensional kernel density estimation

Abstract

The invention provides a dam displacement early warning method based on multi-dimensional kernel density estimation, and relates to the technical field of dam monitoring, and the method comprises the steps: obtaining the historical data and measurement displacement of a dam, and carrying out the preprocessing; establishing a water season time model and an expected displacement model according to the preprocessed data; according to the historical data and the measured displacement of the dam, using an expected displacement model to obtain expected displacement, and subtracting the measured displacement from an expected displacement value to obtain a displacement residual error; and according to the displacement residual error, calculating the displacement residual error by using a multi-dimensional kernel density estimation method so as to obtain cumulative distribution functions, dividing the state of the dam into different early warning levels, setting an early warning threshold for each early warning level, and comparing the cumulative distribution functions so as to obtain the displacement state of the dam. According to the method, the influence of non-normality of displacement residual distribution and high dimension of a confidence region on the accuracy and availability of a traditional early warning method is solved.

Description

Technical Field

[0001] The invention relates to the technical field of dam monitoring, and in particular to an early warning method for dam displacement based on multidimensional kernel density estimation. Background Art

[0002] The main purpose of dam monitoring is to track the changes in the dam's shape in real time to ensure the stability of its structure. Through monitoring data, potential risks can be warned, dam anomalies can be discovered in time, warnings can be issued in advance, measures can be taken to prevent accidents and guide maintenance and reinforcement. Based on the monitoring data, scientific maintenance and reinforcement plans can be formulated to ensure the long-term safety of the dam and verify the design and construction quality.

[0003] Displacement is a key parameter that characterizes the behavior of a dam. The dam monitoring system uses displacement variables as warning objects and compares them with warning indexes. When the warning object exceeds the corresponding warning index, an alarm is triggered to prompt dam managers to investigate the cause and take necessary reinforcement measures to ensure the safety of the dam. The behavior of warning displacement can be studied using monitoring models, limit displacement method, and structural calculation method.

[0004] Traditional dam displacement early warning methods usually assume that the displacement residuals follow a normal distribution. This assumption deviates from the actual situation and affects the reliability of the early warning results. Traditional dam displacement early warning methods usually use the confidence region of the displacement residuals as the early warning index. Since the dimension of the confidence region is the same as the number of monitoring points, high-dimensional confidence regions are less convenient in practical applications. Summary of the invention

[0005] The present invention provides an early warning method for dam displacement based on multidimensional kernel density estimation, which solves the influence of non-normality of displacement residual distribution and high dimensionality of confidence region on the accuracy and availability of traditional warning methods.

[0006] In order to solve the above technical problems, the technical solution of the present invention is as follows: In a first aspect, a dam displacement early warning method based on multidimensional kernel density estimation is provided, the method comprising: Acquire historical data of the dam and measured displacement, and perform preprocessing to obtain preprocessed data, wherein the historical data of the dam includes a water level factor, a temperature factor, and an aging factor; According to the preprocessed data, a water season time model and an expected displacement model are established, wherein the water season time model includes a water level factor displacement model, a temperature factor displacement model, and an aging factor displacement model; According to the historical data of the dam and the measured displacement, the expected displacement is obtained using the expected displacement model, and the measured displacement is subtracted from the expected displacement value to obtain the displacement residual; Based on the displacement residuals, the displacement residuals are calculated using the multidimensional kernel density estimation method to obtain the cumulative distribution function; According to the cumulative distribution function, the state of the dam is divided into different warning levels, and a warning threshold is set for each warning level, wherein the warning levels include normal, slightly abnormal, and abnormal; According to the warning threshold and warning level, the cumulative distribution function is compared to obtain the status of dam displacement.

[0007] Further, the dam historical data and measured displacement are obtained and preprocessed to obtain preprocessed data, wherein the dam historical data includes water level factor, temperature factor, aging factor, including: According to the actual situation of the dam and monitoring needs, the dam is divided into n monitoring points; Assign a unique number to each monitoring point and store the number in the monitoring point set; According to the monitoring point number, the water level factor, temperature factor, aging factor, and measured displacement are extracted from the corresponding monitoring point historical records to obtain the dam historical data and measured displacement; Preprocessing the dam historical data and measured displacement to obtain preprocessed data; The preprocessed data is stored in the database.

[0008] Furthermore, based on the preprocessed data, a water season time model and an expected displacement model are established. The water season time model includes a water level factor displacement model, a temperature factor displacement model, and an aging factor displacement model, including: According to the preprocessed data, the water level factor is selected as the independent variable to establish the water level factor displacement model. ,in, is the water level, for , is the regression coefficient of water level factor displacement, is the error term of water level factor displacement; From the preprocessed data, the observed average temperature of different time periods is selected as the independent variable to establish the temperature factor displacement model. ,in, represents the temperature factor displacement, For the The observed average temperature for the time period, is the regression coefficient of temperature factor displacement, is the regression coefficient of the temperature hysteresis effect, Lag Temperature of the time period, is the error term of temperature factor displacement; From the preprocessed data, the initial date and monitoring date were selected as independent variables to establish the aging factor displacement model. ,in, is the aging factor displacement, is the time difference between the initial date and the monitoring date, is the regression coefficient of the aging factor displacement, is the error term of the aging factor displacement; The water level factor displacement model, temperature factor displacement model and aging factor displacement model are integrated to obtain the expected displacement model ,in, is the expected displacement, is the water level factor displacement, Temperature factor displacement, is the aging factor displacement, is the error term of the overall model.

[0009] Furthermore, based on the historical data and displacement of the dam, the expected displacement is obtained using the expected displacement model, and the measured displacement is subtracted from the expected displacement value to obtain the displacement residual, including: According to the historical data of the dam and the measured displacement, the expected displacement value is obtained using the expected displacement model; According to the expected displacement value, use The difference between the measured displacement and the expected displacement is calculated to obtain the displacement residual, where represents the displacement residual, represents the expected displacement, Represents the measured displacement.

[0010] Furthermore, based on the displacement residuals, the displacement residuals are calculated using a multidimensional kernel density estimation method to obtain a cumulative distribution function, including: Obtain the displacement residuals and organize the displacement residual data of all monitoring points into a set, recorded as ,in, represents the displacement residual, n is the number of monitoring points, i is the index; According to the displacement residual data set of all monitoring points, use The displacement residuals of all monitoring points are estimated to obtain the density function, where Represents the density function estimate of the displacement residuals of all monitoring points, is the kernel function, m is the sample size, for The jth sample of , h is the window width, and n is the number of monitoring points; According to the density function, determine the optimal window width of the product kernel with Gaussian kernel as the unit kernel function , where m is the sample size and n is the number of monitoring points; According to the density function, use The cumulative distribution function of the displacement residuals of all monitoring points is calculated, where represents the cumulative distribution function of the displacement residuals of all monitoring points, is the displacement residual, and n is the number of monitoring points.

[0011] Furthermore, according to the cumulative distribution function, the state of the dam is divided into different warning levels, and a warning threshold is set for each warning level, wherein the warning levels include normal, slightly abnormal, and abnormal, including: The status of the dam is divided into different warning levels, including normal, slightly abnormal, and abnormal; According to the cumulative distribution function, the distribution of dam state parameters is obtained; According to the distribution of dam status parameters, the corresponding cumulative distribution function value range is established for each warning level to obtain the warning index; According to the early warning index, establish an early warning index table.

[0012] Furthermore, the warning threshold and warning level are compared with the cumulative distribution function to obtain the state of dam displacement, including: Obtain the real-time displacement data of the monitoring points and calculate the cumulative distribution function; Compare the cumulative distribution function with the warning threshold to determine the warning level; The displacement status of the monitoring point is determined according to the warning level.

[0013] In a second aspect, an electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, an early warning method for dam displacement based on multidimensional kernel density estimation as described above is implemented.

[0014] In a third aspect, a non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, an early warning method for dam displacement based on multidimensional kernel density estimation as described above is implemented.

[0015] In a fourth aspect, a computer program product includes a computer program, characterized in that when the computer program is executed by a processor, it implements any of the above-mentioned methods for early warning of dam displacement based on multidimensional kernel density estimation.

[0016] The above solution of the present invention includes at least the following beneficial effects: The above scheme of the present invention adopts a multidimensional kernel density estimation method to calculate the cumulative distribution function of the displacement residual, which can more accurately describe the true distribution characteristics of the displacement residual, thereby improving the accuracy of the early warning; the present invention utilizes a multidimensional kernel density estimation method, which can effectively deal with this high dimensionality, making the early warning method more comprehensive and reliable; by using the cumulative distribution function as an early warning index, the present invention provides an intuitive and easy-to-understand early warning method, and the early warning index can directly reflect the degree to which the dam displacement behavior deviates from the normal state, which is convenient for decision makers to take measures in time, thereby enhancing the availability and practicality of the early warning method; the present invention can detect the abnormality of the dam displacement behavior in advance through an accurate and reliable early warning method, provide strong support for the safety management of the dam, help reduce the occurrence of dam safety accidents, protect people’s lives and property safety, and improve the overall safety management level of the dam. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flow chart of a dam displacement early warning method based on multidimensional kernel density estimation provided by an embodiment of the present invention.

[0018] Figure 2 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0020] like Figure 1 As shown, an embodiment of the present invention proposes an early warning method for dam displacement based on multidimensional kernel density estimation, the method comprising: Step 11, obtaining historical data of the dam and measured displacement, and performing preprocessing to obtain preprocessed data, wherein the historical data of the dam includes a water level factor, a temperature factor, and an aging factor; Step 12, establishing a water season time model and an expected displacement model based on the preprocessed data, wherein the water season time model includes a water level factor displacement model, a temperature factor displacement model, and an aging factor displacement model; Step 13, according to the historical data of the dam and the measured displacement, the expected displacement is obtained using the expected displacement model, and the measured displacement is subtracted from the expected displacement value to obtain the displacement residual; Step 14, calculating the displacement residuals using a multidimensional kernel density estimation method according to the displacement residuals to obtain a cumulative distribution function; Step 15, according to the cumulative distribution function, the state of the dam is divided into different warning levels, and a warning threshold is set for each warning level, wherein the warning levels include normal, slightly abnormal, and abnormal; Step 16, comparing the cumulative distribution function according to the warning threshold and the warning level to obtain the state of the dam displacement.

[0021] In an embodiment of the present invention, the dam displacement early warning method based on multidimensional kernel density estimation establishes an accurate expected displacement model by comprehensively considering multiple influencing factors such as water level factors, temperature factors and aging factors. By calculating the displacement residual and using the multidimensional kernel density estimation method to obtain the cumulative distribution function, it is possible to more accurately evaluate the degree to which the dam displacement behavior deviates from the normal state, thereby achieving early warning of the dam displacement, which not only improves the accuracy of the warning, but also enhances the reliability and practicality of the warning, and is helpful to timely discover the safety hazards of the dam and ensure the safe operation of the dam. In step 11, historical data of the dam, including water level factors, temperature factors, aging factors, etc., and measured displacement data are collected, and preprocessing operations such as cleaning, denoising, and normalization are performed on the collected data to ensure the accuracy and consistency of the data; in step 12, according to the preprocessed data, a water level factor displacement model, a temperature factor displacement model, and an aging factor displacement model are established respectively, and the relationship between each factor and the displacement is described. The three factor displacement models are combined to establish a water season time model and an expected displacement model. , which is used to predict the expected displacement of the dam under specific conditions; in step 13, the expected displacement model is used to calculate the expected displacement value according to the historical data and current conditions of the dam, and the actual measured displacement is compared with the expected displacement value to calculate the difference, that is, the displacement residual; in step 14, the displacement residual data is subjected to multidimensional kernel density estimation to obtain the probability density function of the displacement residual, and the cumulative distribution function of the displacement residual is obtained by integrating the probability density function, which describes the probability that the displacement residual is less than a certain value; in step 15, according to the cumulative distribution function and actual needs, the state of the dam is divided into different warning levels, such as normal, slightly abnormal, abnormal, etc., and corresponding warning thresholds are set for each warning level. These thresholds are used to determine whether the displacement state of the dam reaches or exceeds a specific warning level; in step 16, the cumulative distribution function value of the current displacement residual is compared with the warning threshold, and the state of the dam displacement is determined according to the comparison result, and a corresponding warning signal is issued. If the displacement state reaches or exceeds a specific warning level, the warning mechanism is triggered to remind relevant personnel to take timely measures.

[0022] like Figure 1 As shown, step 11, obtaining the dam historical data and measured displacement, and preprocessing to obtain preprocessed data, the dam historical data includes water level factor, temperature factor, aging factor, including: Step 111, dividing the dam into n monitoring points according to the actual situation of the dam and monitoring requirements; Step 112, assigning a unique number to each monitoring point and storing the number in the monitoring point set; Step 113, extracting the water level factor, temperature factor, aging factor, and measured displacement from the corresponding monitoring point historical records according to the monitoring point number, so as to obtain the dam historical data and measured displacement; Step 114, preprocessing the dam historical data and the measured displacement to obtain preprocessed data; Step 115, storing the preprocessed data in a database.

[0023] In the embodiment of the present invention, accurate, complete and properly preprocessed historical data and measured displacement of the dam are obtained to provide reliable data support for the subsequent establishment of models, calculation of expected displacements and displacement residuals. By dividing the monitoring points in detail, extracting key factors and displacement data, and performing preprocessing, the availability and accuracy of the data can be significantly improved, thereby providing a solid foundation for subsequent analysis and early warning. In step 111, the dam is divided into n monitoring points according to the structural characteristics and monitoring requirements of the dam, which should be able to fully reflect the displacement of the dam and facilitate subsequent data collection and processing. The division of monitoring points should take into account different parts of the dam, such as the dam top, dam foundation, dam slope, and areas that may be affected by different environmental factors, such as water level and temperature. In step 112, a unique number is assigned to each monitoring point for subsequent data processing. Each monitoring point can be accurately identified in the analysis, and these numbers are stored in the monitoring point set for easy management and query; in step 113, according to the number of the monitoring point, the water level factor, temperature factor, aging factor and measured displacement data are extracted from the corresponding monitoring point historical records to ensure that the extracted data corresponds to the actual situation of the monitoring point, and the time range and sampling frequency of the data meet the needs of subsequent analysis; in step 114, the extracted dam historical data and measured displacement are preprocessed to improve the accuracy and consistency of the data, providing a reliable data basis for subsequent modeling and calculation; in step 115, the preprocessed data is stored in the database for subsequent analysis and query. The design of the database should take into account factors such as the structural characteristics of the data, query requirements and data security to ensure efficient and reliable data storage and management.

[0024] like Figure 1 As shown, step 12, based on the pre-processed data, a water season time model and an expected displacement model are established, wherein the water season time model includes a water level factor displacement model, a temperature factor displacement model, and an aging factor displacement model, including: Step 121, based on the preprocessed data, select the water level factor as the independent variable and establish the water level factor displacement model ,in, is the water level, for , is the regression coefficient of water level factor displacement, is the error term of water level factor displacement; Step 122, from the preprocessed data, select the observed average temperature of different time periods as the independent variable, and establish the temperature factor displacement model ,in, represents the temperature factor displacement, For the The observed average temperature for the time period, is the regression coefficient of temperature factor displacement, is the regression coefficient of the temperature hysteresis effect, Lag Temperature of the time period, is the error term of temperature factor displacement; Step 123, from the preprocessed data, select the initial date and the monitoring date as independent variables to establish an aging factor displacement model ,in, is the aging factor displacement, is the time difference between the initial date and the monitoring date, is the regression coefficient of the aging factor displacement, is the error term of the aging factor displacement; Step 124: Integrate the water level factor displacement model, the temperature factor displacement model and the aging factor displacement model to obtain an expected displacement model. ,in, is the expected displacement, is the water level factor displacement, Temperature factor displacement, is the aging factor displacement, is the error term of the overall model.

[0025] In the embodiment of the present invention, a water season time model and an expected displacement model are established, and a water level factor displacement model, a temperature factor displacement model and an aging factor displacement model are established, so that the specific influence of each factor on the dam displacement can be deeply understood. These models are integrated into an expected displacement model, which can more comprehensively predict the displacement of the dam under different conditions, not only improving the accuracy of displacement prediction, but also providing a reliable basis for subsequent calculation of displacement residuals and early warning. In step 121, a water level factor is selected as an independent variable from the preprocessed data, and the dependent variable is determined to be the displacement caused by the corresponding water level factor. A water level factor displacement model is fitted according to the data using an appropriate regression method, and the specific influence of water level changes on the dam displacement can be analyzed through the model. In step 122, the observed average temperature of different time periods is selected as an independent variable from the preprocessed data, and the dependent variable is determined to be the displacement caused by the corresponding temperature factor, taking into account the possible lag effect of temperature. In response, the temperature and its regression coefficient of the lag time period are introduced; a temperature factor displacement model is fitted according to the data, and the specific influence of temperature change and its lag effect on the dam displacement can be analyzed through the model; in step 123, the time difference between the initial date and the monitoring date is selected as the independent variable from the preprocessed data, and the dependent variable is determined to be the displacement caused by the corresponding aging factor; using an appropriate regression method, an aging factor displacement model is fitted according to the data, and the specific influence of dam aging over time on the displacement can be analyzed through the model; in step 124, the above three factor displacement models are integrated to obtain an expected displacement model. Through the integrated model, the joint influence of water level, temperature and aging factors on dam displacement can be comprehensively considered, and the accuracy and comprehensiveness of displacement prediction can be improved. Using the expected displacement model, the displacement of the dam can be predicted based on the real-time monitored water level, temperature and aging information, providing a reliable basis for subsequent calculation of displacement residuals and early warning.

[0026] like Figure 1 As shown, step 13, according to the historical data and displacement of the dam measurement, the expected displacement is obtained using the expected displacement model, and the measured displacement is subtracted from the expected displacement value to obtain the displacement residual, including: Step 131, using an expected displacement model to obtain an expected displacement value according to the historical data of the dam and the measured displacement; Step 132, based on the expected displacement value, use The difference between the measured displacement and the expected displacement is calculated to obtain the displacement residual, where represents the displacement residual, represents the expected displacement, Represents the measured displacement.

[0027] In the embodiment of the present invention, the displacement residual is calculated by comparing the actual measured displacement with the displacement predicted by the expected displacement model, which is the key to evaluating the difference between the actual displacement behavior of the dam and the expected behavior, and is crucial for the subsequent multidimensional kernel density estimation and early warning level division. By calculating the displacement residual, the abnormal components in the dam displacement can be identified. These abnormalities may be caused by potential safety problems or unknown factors, which not only improves the depth of understanding of the dam displacement behavior, but also provides an important basis for timely discovery and handling of dam safety hazards. In step 131, according to the historical data and current conditions of the dam, the expected displacement value of each monitoring point is calculated using the expected displacement model. The expected displacement value is obtained by substituting the independent variables in the model into the model and calculating the corresponding displacement value, ensuring that the data used when calculating the expected displacement is accurate and up-to-date to ensure the reliability of the expected displacement. In step 132, for each monitoring point, the calculated expected displacement value is compared with the actual measured displacement value. The displacement residual is obtained by subtracting the expected displacement from the measured displacement. The displacement residual of each monitoring point is recorded for subsequent multidimensional kernel density estimation and early warning analysis.

[0028] like Figure 1 As shown, step 14, based on the displacement residuals, the displacement residuals are calculated using a multidimensional kernel density estimation method to obtain a cumulative distribution function, including: Step 141, obtain the displacement residual, and organize the displacement residual data of all monitoring points into a set, recorded as ,in, represents the displacement residual, n is the number of monitoring points, i is the index; Step 142: Based on the displacement residual data set of all monitoring points, use The displacement residuals of all monitoring points are estimated to obtain the density function, where Represents the density function estimate of the displacement residuals of all monitoring points, is the kernel function, m is the sample size, for The jth sample of , h is the window width, and n is the number of monitoring points; Step 143, based on the density function, determine the optimal window width of the product kernel of the Gaussian kernel as the unit kernel function , where m is the sample size and n is the number of monitoring points; Step 144, according to the density function, use The cumulative distribution function of the displacement residuals of all monitoring points is calculated, where represents the cumulative distribution function of the displacement residuals of all monitoring points, is the displacement residual, and n is the number of monitoring points.

[0029] In an embodiment of the present invention, a multidimensional kernel density estimation method is used to perform in-depth statistical analysis on the displacement residuals, thereby obtaining a cumulative distribution function of the displacement residuals. Through the multidimensional kernel density estimation, the distribution details of the displacement residual data can be captured more accurately, thereby providing a solid statistical basis for formulating a scientific and reasonable early warning mechanism; in step 141, the displacement residual data of all monitoring points are obtained, and these data are organized into a set. In step 142, the displacement residuals of all monitoring points are estimated using the multidimensional kernel density estimation method; in step 143, when performing multidimensional kernel density estimation, the selection of the window width h is crucial, and cross-validation or other optimization methods are used to determine the optimal window width of the product kernel of the unit kernel function with a Gaussian kernel. The selection of the optimal window width should balance the accuracy and smoothness of the estimation to avoid overfitting or underfitting; in step 144, based on the obtained density function estimate, the cumulative distribution function of the displacement residuals of all monitoring points is calculated, and the cumulative distribution function of the displacement residuals is obtained, which not only provides a comprehensive description of the distribution of the displacement residuals, but also provides an important statistical basis for the subsequent setting of early warning thresholds and risk assessment.

[0030] like Figure 1 As shown, step 15, according to the cumulative distribution function, the state of the dam is divided into different warning levels, and a warning threshold is set for each warning level, wherein the warning levels include normal, slightly abnormal, and abnormal, including: Step 151, classifying the status of the dam into different warning levels, wherein the warning levels include normal, slightly abnormal, and abnormal; Step 152, obtaining the distribution of the dam state parameters according to the cumulative distribution function; Step 153, according to the distribution of the dam state parameters, establish a corresponding cumulative distribution function value range for each warning level to obtain a warning index; Step 154, establishing a warning index table according to the warning index.

[0031] In the embodiment of the present invention, the state warning level of the dam is scientifically divided based on the cumulative distribution function, and a clear warning threshold is set for each warning level, which can help managers to timely and accurately identify the safety status of the dam, so as to take appropriate countermeasures to prevent potential safety risks. By reasonably setting the warning level and its threshold, it can be ensured that a warning can be issued in time when an abnormal situation occurs in the dam, providing a strong guarantee for the safe operation of the dam; in step 151, according to the actual needs of dam safety management, the state of the dam is clearly divided into three warning levels: normal, slightly abnormal, and abnormal; step 152: using the cumulative distribution function, in-depth analysis of the distribution of the displacement residuals of the dam, including understanding the central trend, dispersion and possible abnormal values ​​of the displacement residuals; in step 153, based on the distribution of the dam state parameters, the corresponding cumulative distribution function value range is established for each warning level, that is, the value range of the cumulative distribution function corresponding to "normal" is , the value range of the cumulative distribution function corresponding to "slightly abnormal" is , the value range of the cumulative distribution function corresponding to "abnormal" is In step 154, according to the warning index range, a warning index table is obtained as follows:

[0032] like Figure 1 As shown, step 16, according to the early warning threshold and the early warning level, is compared with the cumulative distribution function to obtain the state of the dam displacement, including: Step 161, obtaining real-time displacement data of the monitoring point and calculating the cumulative distribution function; Step 162, comparing the cumulative distribution function with the warning threshold to determine the warning level; Step 163: determine the displacement state of the monitoring point according to the warning level.

[0033] In the embodiment of the present invention, the displacement data monitored in real time is compared with the preset warning threshold and cumulative distribution function to determine the displacement state of each monitoring point of the dam, so as to timely identify the possible safety risks. Through real-time comparison and analysis, the abnormal changes of the dam displacement can be quickly discovered, which provides an important basis for taking countermeasures in time and ensuring the safety of the dam. In step 161, the displacement data of each monitoring point is obtained in real time through the dam safety monitoring system, which can timely reflect the actual displacement of the dam. The cumulative distribution function is recalculated based on the displacement data obtained in real time by using the multidimensional kernel density estimation method established in step 14 to ensure that the cumulative distribution function can reflect the current displacement state of the dam, rather than the historical state. In step 162, the cumulative distribution function value calculated in real time is compared with the warning threshold set in step 15. The values ​​are compared for each monitoring point respectively; the corresponding warning level is determined according to the warning threshold range of the cumulative distribution function value; if the cumulative distribution function value is within the warning index range of the "normal" state, the monitoring point is determined to be in a normal state; if it is within the warning index range of the "slightly abnormal" or "abnormal" state, it is determined to be slightly abnormal or abnormal respectively; in step 163, according to the warning level determined in step 162, the displacement state of each monitoring point is clarified, and the displacement state information is reported to the dam management department in a timely manner so that corresponding response measures can be taken according to the specific circumstances. If it is determined to be an abnormal state, the emergency plan should be immediately activated, and further inspection and analysis should be carried out to ensure the safe operation of the dam, which improves the real-time and accuracy of dam safety management and provides a strong guarantee for timely response to potential safety risks.

[0034] Figure 2 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention.

[0035] like Figure 2 As shown, the electronic device may include: a processor 610, a communication interface 620, a memory 630 and a communication bus 640, wherein the processor 610, the communication interface 620 and the memory 630 communicate with each other through the communication bus 640. The processor 610 may call the logic instructions in the memory 630 to execute the dam displacement early warning method based on multi-dimensional kernel density estimation.

[0036] In addition, the logic instructions in the above-mentioned memory 630 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0037] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the dam displacement early warning method based on multidimensional kernel density estimation provided by the above methods.

[0038] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the dam displacement early warning method based on multidimensional kernel density estimation provided by the above methods.

[0039] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0040] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0041] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A dam displacement early warning method based on multidimensional kernel density estimation, characterized in that: The method comprises: Acquire historical data of the dam and measured displacement, and perform preprocessing to obtain preprocessed data, wherein the historical data of the dam includes a water level factor, a temperature factor, and an aging factor; According to the preprocessed data, a water season time model and an expected displacement model are established, wherein the water season time model includes a water level factor displacement model, a temperature factor displacement model, and an aging factor displacement model; According to the historical data of the dam and the measured displacement, the expected displacement is obtained using the expected displacement model, and the measured displacement is subtracted from the expected displacement value to obtain the displacement residual; Based on the displacement residuals, the displacement residuals are calculated using the multidimensional kernel density estimation method to obtain the cumulative distribution function; According to the cumulative distribution function, the state of the dam is divided into different warning levels, and a warning threshold is set for each warning level, wherein the warning levels include normal, slightly abnormal, and abnormal; According to the warning threshold and warning level, the cumulative distribution function is compared to obtain the status of dam displacement.

2. The dam displacement early warning method based on multidimensional kernel density estimation according to claim 1 is characterized in that: Acquire the dam historical data and measured displacement, and perform preprocessing to obtain preprocessed data, wherein the dam historical data includes water level factor, temperature factor, aging factor, including: According to the actual situation of the dam and monitoring needs, the dam is divided into n monitoring points; Assign a unique number to each monitoring point and store the number in the monitoring point set; According to the monitoring point number, the water level factor, temperature factor, aging factor, and measured displacement are extracted from the corresponding monitoring point historical records to obtain the dam historical data and measured displacement; Preprocessing the dam historical data and measured displacement to obtain preprocessed data; The preprocessed data is stored in the database.

3. The dam displacement early warning method based on multidimensional kernel density estimation according to claim 1 is characterized in that: According to the preprocessed data, a water season time model and an expected displacement model are established. The water season time model includes a water level factor displacement model, a temperature factor displacement model, and an aging factor displacement model, including: According to the preprocessed data, the water level factor is selected as the independent variable to establish the water level factor displacement model. ,in, is the water level, for , is the regression coefficient of water level factor displacement, is the error term of water level factor displacement; From the preprocessed data, the observed average temperature of different time periods is selected as the independent variable to establish the temperature factor displacement model. ,in, represents the temperature factor displacement, For the The observed average temperature for the time period, is the regression coefficient of temperature factor displacement, is the regression coefficient of the temperature hysteresis effect, Lag Temperature of the time period, is the error term of temperature factor displacement; From the preprocessed data, the initial date and monitoring date were selected as independent variables to establish the aging factor displacement model. ,in, is the aging factor displacement, is the time difference between the initial date and the monitoring date, is the regression coefficient of the aging factor displacement, is the error term of the aging factor displacement; The water level factor displacement model, temperature factor displacement model and aging factor displacement model are integrated to obtain the expected displacement model ,in, is the expected displacement, is the water level factor displacement, Temperature factor displacement, is the aging factor displacement, is the error term of the overall model.

4. The dam displacement early warning method based on multidimensional kernel density estimation according to claim 1 is characterized in that: According to the historical data of the dam and the measured displacement, the expected displacement is obtained using the expected displacement model, and the measured displacement is subtracted from the expected displacement value to obtain the displacement residual, including: According to the historical data of the dam and the measured displacement, the expected displacement value is obtained using the expected displacement model; According to the expected displacement value, use The difference between the measured displacement and the expected displacement is calculated to obtain the displacement residual, where represents the displacement residual, represents the expected displacement, Represents the measured displacement.

5. The dam displacement early warning method based on multidimensional kernel density estimation according to claim 1 is characterized in that: Based on the displacement residuals, the displacement residuals are calculated using the multidimensional kernel density estimation method to obtain the cumulative distribution function, including: Obtain the displacement residuals and organize the displacement residual data of all monitoring points into a set, recorded as ,in, represents the displacement residual, n is the number of monitoring points, i is the index; According to the displacement residual data set of all monitoring points, use The displacement residuals of all monitoring points are estimated to obtain the density function, where Represents the density function estimate of the displacement residuals of all monitoring points, is the kernel function, m is the sample size, for The jth sample of , h is the window width, and n is the number of monitoring points; According to the density function, determine the optimal window width of the product kernel with Gaussian kernel as the unit kernel function , where m is the sample size and n is the number of monitoring points; According to the density function, use The cumulative distribution function of the displacement residuals of all monitoring points is calculated, where represents the cumulative distribution function of the displacement residuals of all monitoring points, is the displacement residual, and n is the number of monitoring points.

6. The dam displacement early warning method based on multidimensional kernel density estimation according to claim 1 is characterized in that: According to the cumulative distribution function, the status of the dam is divided into different warning levels, and a warning threshold is set for each warning level. The warning levels include normal, slightly abnormal, and abnormal, including: The status of the dam is divided into different warning levels, including normal, slightly abnormal, and abnormal; According to the cumulative distribution function, the distribution of dam state parameters is obtained; According to the distribution of dam status parameters, the corresponding cumulative distribution function value range is established for each warning level to obtain the warning index; According to the early warning index, establish an early warning index table.

7. The dam displacement early warning method based on multidimensional kernel density estimation according to claim 1 is characterized in that: According to the warning threshold and warning level, the cumulative distribution function is compared to obtain the state of dam displacement, including: Obtain the real-time displacement data of the monitoring points and calculate the cumulative distribution function; Compare the cumulative distribution function with the warning threshold to determine the warning level; The displacement status of the monitoring point is determined according to the warning level.

8. An electronic 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 program, the dam displacement early warning method based on multidimensional kernel density estimation as described in any one of claims 1 to 7 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the dam displacement early warning method based on multidimensional kernel density estimation as claimed in any one of claims 1 to 7 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the dam displacement early warning method based on multidimensional kernel density estimation as claimed in any one of claims 1 to 7 is implemented.

Images