Dam slope drilling inclinometry displacement data assessment method based on environmental factors

Abstract

The invention provides a dam slope drilling inclinometry displacement data evaluation method and system based on environmental factors. The method comprises the following steps that equipment state information and environmental change information of equipment when the dam slope drilling inclinometry displacement data are obtained are collected in real time; calculating an abnormal value hidden danger evaluation coefficient according to the collected equipment state information and environment change information; comparing the abnormal value hidden danger evaluation coefficient with a first threshold value to generate a high hidden danger signal or a low hidden danger signal; when the high hidden danger signal is generated, obtaining an abnormal value hidden danger evaluation coefficient sequence corresponding to a plurality of continuous high hidden danger signal sequences containing the currently generated high hidden danger signal; and determining an evaluation result of the data exception risk level according to the exception value hidden danger evaluation coefficient sequence. The abnormal condition of the displacement data can be effectively predicted and prompted, so that the monitoring precision of the monitoring equipment is improved.

Description

Technical Field

[0001] The invention belongs to the technical field of dam slope drilling inclination displacement monitoring, and in particular relates to a dam slope drilling inclination displacement data evaluation method based on environmental factors. Background Art

[0002] Knowledge graph is a structured knowledge representation method that represents the relationship between entities in the form of a graph structure. This structured knowledge representation method makes data easy to understand, scalable, semantically rich, and supports reasoning.

[0003] With the increasing importance of dam slope engineering safety monitoring, the processing and analysis of borehole inclination displacement data, as a key indicator reflecting the stability of dam slopes, has become particularly important. Traditional data processing methods are often limited to a single statistical analysis, making it difficult to deeply mine the correlation and implicit knowledge between data. Therefore, constructing a knowledge graph of dam slope borehole inclination displacement data can achieve deep mining and effective use of these data, which is of great significance for improving the safety monitoring level of dam slope engineering.

[0004] In the process of building a knowledge graph, the key steps include knowledge extraction, knowledge fusion, and knowledge processing. Knowledge extraction aims to extract entities, attributes, and relationships related to borehole inclination displacement from various data sources. Knowledge fusion integrates the extracted knowledge, eliminates contradictions and ambiguities, and forms a unified knowledge system. Finally, the knowledge processing stage will evaluate and optimize the quality of the fused knowledge to ensure the accuracy and effectiveness of the knowledge graph.

[0005] By constructing a knowledge graph of the dam slope borehole inclination displacement data, the data can be visualized, queried and analyzed, which helps monitoring personnel to more intuitively understand the deformation of the dam slope and promptly discover potential safety hazards. In addition, based on the reasoning and analysis of the knowledge graph, the correlation and regularity between the data can also be mined, providing strong support for the safety monitoring and early warning of the dam slope project.

[0006] The existing technology has the following deficiencies: in the process of constructing the knowledge graph of dam slope incline displacement, instrument and equipment failures and environmental changes often occur, which will lead to outliers in the monitoring data, that is, data points that deviate greatly from normal data, which will affect the accuracy and completeness of the geological information and monitoring records in the knowledge graph, and then affect the prediction and reasoning results of slope stability, resulting in deviations in the prediction and reasoning results of slope stability, affecting the accuracy and reliability of engineering decisions. The existing technology cannot predict the occurrence of outlier data in a timely manner, which will have a negative impact on the data quality of the knowledge graph and reduce the credibility of the overall data. Summary of the invention

[0007] The purpose of the present invention is to address the deficiencies of the above-mentioned background technology and to provide a method for evaluating dam slope drilling incline displacement data based on environmental factors, which can effectively predict and prompt abnormal situations of displacement data, thereby improving the monitoring accuracy of monitoring equipment.

[0008] The technical solution adopted by the present invention is: a method for evaluating dam slope drilling inclination displacement data based on environmental factors, comprising the following steps:

[0009] Real-time acquisition of equipment status information and environmental change information when the equipment is acquiring dam slope drilling and displacement data;

[0010] The abnormal value hidden danger assessment coefficient is calculated based on the collected equipment status information and environmental change information;

[0011] Comparing the abnormal value hidden danger assessment coefficient with the first threshold value to generate a high hidden danger signal or a low hidden danger signal;

[0012] When a low risk signal is generated, it is determined that there is no data anomaly risk;

[0013] When a high hidden danger signal is generated, an abnormal value hidden danger assessment coefficient sequence corresponding to a plurality of continuous high hidden danger signal sequences including the currently generated high hidden danger signal is obtained; and an assessment result of the data abnormality risk level is determined according to the abnormal value hidden danger assessment coefficient sequence.

[0014] In the above technical solution, the device status information includes: sensor signal strength fluctuation coefficient and transmission rate variation coefficient.

[0015] In the above technical solution, the environmental change information includes the probability of temperature anomaly.

[0016] In the above technical solution, the following formula is used to calculate the abnormal value hidden danger assessment coefficient Q yczyh :

[0017]

[0018] Where r1, r2 and r3 are the sensor signal strength fluctuation coefficients S respectively. bd , Transmission rate variation coefficient TR vc and the temperature anomaly probability W ycp The preset proportional coefficient, and r1, r2 and r3 are all greater than 0.

[0019] In the above technical solution, if the abnormal value hidden danger assessment coefficient is greater than the first threshold, a high hidden danger signal is generated; if the abnormal value hidden danger assessment coefficient is less than or equal to the first threshold, a low hidden danger signal is generated.

[0020] In the above technical solution, the process of determining the evaluation result of the data anomaly risk level according to the outlier hidden danger evaluation coefficient sequence includes:

[0021] Calculate the mean and standard deviation of the outlier hidden danger assessment coefficient series;

[0022] If the average value of the abnormal value hidden danger assessment coefficient sequence is greater than or equal to the second threshold, a high risk signal is generated;

[0023] If the average value of the abnormal value hidden danger assessment coefficient sequence is less than the second threshold value, and the standard deviation of the abnormal value hidden danger assessment coefficient sequence is greater than or equal to the third threshold value, a medium risk signal is generated;

[0024] If the average value of the outlier hidden danger assessment coefficient sequence is less than the second threshold value, and the standard deviation of the outlier hidden danger assessment coefficient sequence is less than the third threshold value, a low risk signal is generated.

[0025] In the above technical solution, the process of obtaining the sensor signal strength fluctuation coefficient includes:

[0026] Collect and calibrate the sensor signal strength at different times within the set time period corresponding to the current time;

[0027] Calibrate the average value of the sensor signal strength at each moment in the sliding window;

[0028] The sensor signal strength fluctuation coefficient is calculated based on the difference between the sensor signal strength at different times and the average value of the sensor signal strength at each time in the sliding window.

[0029] In the above technical solution, the process of obtaining the transmission rate variation coefficient includes:

[0030] Collect and calibrate the sensor transmission rate at different times within the set time period corresponding to the current time;

[0031] The transmission rate variation coefficient is calculated based on the average value of the variation rate of the sensor transmission rate at every two adjacent moments.

[0032] In the above technical solution, the process of obtaining the temperature anomaly probability includes:

[0033] Collect and calibrate the temperature at different times within the set time period corresponding to the current time and perform normalization processing to form a temperature set;

[0034] Use the temperature set to construct an isolation forest, which consists of several isolated trees; and obtain the path length of each temperature in each isolated tree;

[0035] The anomaly score at the corresponding moment is calculated based on the path length of each temperature in each isolated tree;

[0036] Use the sigmoid function to map the anomaly score to the range of and get the temperature anomaly probability.

[0037] The beneficial effects of the present invention are as follows: the present invention uses the abnormal value hidden danger assessment coefficient generated by the instrument when obtaining the dam slope incline displacement data, combined with the preset mean value and standard deviation, to analyze and predict the possibility of abnormal values ​​appearing in the subsequent monitoring data. By analyzing the generated high, medium and low risk signals, the possibility of abnormal values ​​appearing in the subsequent monitoring data can be quickly judged, providing an important reference for engineering decision-making. Specifically, when the instrument generates a high-risk signal, it means that the possibility of abnormal values ​​in the subsequent monitoring data is very high, requiring special attention and timely processing; while the generation of a medium-risk signal indicates that the possibility of abnormal values ​​is relatively high and needs to be handled with caution; the generation of a low-risk signal means that the possibility of abnormal values ​​is low, but still requires attention. This analysis method can help engineering personnel to promptly discover and deal with potential abnormal situations, improve the monitoring and early warning capabilities of the dam slope stability, thereby reducing the risk of accidents and ensuring the safe operation of the project. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a schematic diagram of the method of the present invention. DETAILED DESCRIPTION

[0039] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments to facilitate a clear understanding of the present invention, but they do not constitute a limitation on the present invention.

[0040] The technical solution adopted by the present invention is: a method for evaluating dam slope drilling inclination displacement data based on environmental factors, comprising the following steps:

[0041] Real-time acquisition of equipment status information and environmental change information when the equipment is acquiring dam slope drilling and displacement data;

[0042] The abnormal value hidden danger assessment coefficient is calculated based on the collected equipment status information and environmental change information;

[0043] Comparing the abnormal value hidden danger assessment coefficient with the first threshold value to generate a high hidden danger signal or a low hidden danger signal;

[0044] When a low risk signal is generated, it is determined that there is no data anomaly risk;

[0045] When a high hidden danger signal is generated, an abnormal value hidden danger assessment coefficient sequence corresponding to a plurality of continuous high hidden danger signal sequences including the currently generated high hidden danger signal is obtained; and an assessment result of the data abnormality risk level is determined according to the abnormal value hidden danger assessment coefficient sequence.

[0046] The principle of the present invention is further explained below with reference to specific embodiments.

[0047] like Figure 1 As shown, this specific embodiment 1 provides a method for evaluating dam slope drilling inclination displacement data based on environmental factors, which specifically includes the following steps:

[0048] S1. Collect multiple pieces of information when the equipment acquires the dam slope drilling displacement data, including equipment status information and environmental change information. After collection, the equipment status information and environmental change information are processed separately;

[0049] It should be noted that the equipment obtains the dam slope drilling displacement data mainly including the dam slope position, curve type, drilling position, monitoring time, monitoring displacement data, monitoring records, environmental impact factors and other data. The specific data included and the specific acquisition method of these data are not specifically limited here, and can be selected according to actual needs;

[0050] S2. Establish a data analysis model using the processed equipment status information and environmental change information when the equipment obtains the dam slope drilling displacement data, and generate an abnormal value hidden danger assessment coefficient;

[0051] S3, comparing and analyzing the abnormal value hidden danger assessment coefficient threshold (i.e., the first threshold) pre-set by the equipment when acquiring the dam slope drilling displacement data, and generating a high hidden danger signal and a low hidden danger signal;

[0052] S4. Obtain several abnormal value hidden danger assessment coefficients generated by the equipment when acquiring the dam slope drilling and displacement data and generating a high hidden danger signal, conduct a comprehensive analysis on the several abnormal value hidden danger assessment coefficients obtained, determine the risk level when the equipment generates a high hidden danger signal when acquiring the dam slope drilling and displacement data, generate a corresponding risk level signal, and issue a corresponding early warning prompt.

[0053] In step S1, the device status information of the device when acquiring the dam slope drilling displacement data includes the sensor signal strength fluctuation coefficient and the transmission rate change coefficient. After acquisition, the sensor signal strength fluctuation coefficient and the transmission rate change coefficient are calibrated as S and S, respectively. bd and TR vc When the equipment acquires the displacement data of the dam slope drilling, the environmental change information includes the temperature anomaly probability. After acquisition, the temperature anomaly probability is calibrated as W ycp .

[0054] The logic for obtaining the sensor signal strength fluctuation coefficient is as follows:

[0055] S111, obtaining the sensor signal strength at different times within the time T (i.e., the set time period corresponding to the current time), and calibrating the sensor signal strength at different times within the time T as S m , m represents the number of the sensor signal strength at different times within the time T, m = 1, 2, 3, ..., k, k is a positive integer.

[0056] It should be noted that the method for acquiring the sensor signal strength at different times within the time T is as follows: first, it is necessary to set multiple time points or time intervals within the time range T so as to acquire the sensor signal strength data at different time points or time periods. This can be achieved through the data recording function provided by the sensor device itself or the interface for acquiring sensor data in real time. At each time point or at the end of the time period, the signal strength value of the sensor is recorded. For example, data can be read from the sensor device once at fixed time intervals, or fixed time points can be set for data collection. In this way, the sensor signal strength data at multiple different times within the time T can be obtained, or it can also be acquired by other methods. The method for acquiring the sensor signal strength at different times within the time T is not specifically limited here and can be selected according to actual needs.

[0057] S112, calibrate the sliding window size to ω, and calibrate the average value of the sensor signal strength at each moment in the sliding window ω to

[0058] It should be noted that the calculation of the sensor signal strength fluctuation coefficient by sliding window has many advantages: first, through the real-time update of the sliding window, the fluctuation of the sensor signal strength can be quickly and dynamically evaluated at different time points, and this process is continuous and can be analyzed immediately when the data source is generated. Secondly, the application of sliding window makes the calculation results smoother, because it can effectively reduce the impact of noise, so that users can pay more attention to the real fluctuation pattern and not be misled by temporary fluctuations. In addition, the size of the sliding window can be adjusted according to needs, and can be flexibly adapted to different time scales, so as to better capture the fluctuation characteristics of the data. Most importantly, by monitoring the changes in the sensor signal strength fluctuation coefficient, it is possible to timely detect abnormal conditions in the data, such as sudden fluctuations or abnormal changes, which may mean system failures, abnormal events or environmental changes, so that timely measures can be taken to solve the problem and ensure the stability and reliability of the system. In summary, the calculation of the sensor signal strength fluctuation coefficient by sliding window is an effective method, which can achieve real-time monitoring, smooth processing, strong adaptability and abnormal detection, and help improve the quality and reliability of monitoring data.

[0059] It should be further explained that the sliding window size refers to the number of data points used when calculating the sliding window standard deviation. It determines the range of historical data considered when calculating the standard deviation at each time point in the time series. Specifically, the sliding window size ω represents the number of data points considered forward when calculating the standard deviation at a certain time point. For example, if the sliding window size is selected as 10, 10 data points from k-9 to k will be considered when calculating the standard deviation at time point k. The choice of sliding window size is usually determined by the characteristics of the data and the needs of analysis. A smaller sliding window size can more sensitively capture the short-term volatility of the data, but may be affected by noise; a larger sliding window size can smooth the data and reduce noise, but may cause a lag in response to long-term trends. Therefore, when selecting the sliding window size, it is necessary to consider the volatility of the data, the sampling frequency, and the purpose of the analysis to ensure that appropriate results are obtained. The sliding window size is not specifically limited here and can be selected according to actual needs;

[0060] S113. The calculation formula of the sensor signal strength fluctuation coefficient is as follows:

[0061]

[0062] Among them, S bd is the sensor signal intensity fluctuation coefficient, is the average value of the sensor signal strength at each moment in the sliding window ω, and the calculation formula is:

[0063] It should be noted that evaluating the possibility of outliers in monitoring data is crucial to ensuring the accuracy and reliability of the data. By evaluating the sensor signal strength fluctuation coefficient, it is possible to capture abnormal changes in the data more sensitively, so as to detect possible abnormalities in a timely manner. These anomalies may be caused by equipment failure, environmental changes or other unknown factors, and timely detection and handling of these anomalies is crucial to maintaining the stability and reliability of the system. By evaluating the possibility of outliers, you can be prepared before the anomaly occurs and take appropriate measures to solve the problem and avoid potential losses and risks. In addition, timely handling of outliers can also help improve the quality and credibility of the data, thereby better supporting subsequent data analysis and decision-making. Therefore, evaluating the possibility of outliers in monitoring data is an important task that can ensure the normal operation of the monitoring system and improve the reliability and practicality of the data.

[0064] The logic for obtaining the transmission rate change coefficient is as follows:

[0065] S121, obtain the sensor transmission rate at different times within the time T, and calibrate the sensor transmission rate at different times within the time T as TR n , n represents the number of the sensor transmission rate at different times in time T, n = 1, 2, 3, ..., j, j is a positive integer;

[0066] It should be noted that the method for acquiring the sensor transmission rate at different times within the time T can be: first, a sensor data acquisition module needs to be set up in the monitoring system, and the module is responsible for recording the sensor transmission rate data in real time. Then, the real-time data of the sensor transmission rate is regularly acquired through the interface that communicates with the sensor. This can be done through a network interface, serial port communication, Bluetooth connection, etc. The acquired data can be stored in chronological order to form a time series. Within a period of time T, the transmission rate data is continuously acquired, and the transmission rate value at each moment is recorded to construct a time series data of the transmission rate. In this way, the sensor transmission rate at different times within the time T can be acquired. The method for acquiring the sensor transmission rate at different times within the time T is not specifically limited here and can be selected according to actual needs.

[0067] S122. Calculate the transmission rate variation coefficient. The specific calculation formula is as follows:

[0068]

[0069] In the formula, TR vc is the transmission rate variation coefficient;

[0070] It should be noted that the transmission rate variation coefficient is an indicator to measure the transmission rate variation trend, which reflects the degree of change of the transmission rate between adjacent time points. When the transmission rate variation coefficient is large, it means that the transmission rate has experienced a large fluctuation or change in a short period of time, which may mean that there is an abnormality in the sensor transmission. Therefore, by monitoring the size of the transmission rate variation coefficient, abnormal changes in the transmission process can be detected more sensitively, and potential outliers can be identified in a timely manner. By evaluating the transmission rate variation coefficient, it can help determine the possibility of outliers in the monitoring data, and then take appropriate measures to process and repair them to ensure the normal operation of the monitoring system and the reliability of the data.

[0071] The logic for obtaining the temperature anomaly probability is as follows:

[0072] S131, obtain the temperature at different times within the time T, and calibrate the temperature at different times within the time T as W x, x represents the number of the temperature at different times within the time T, x = 1, 2, 3, ..., b, b is a positive integer, and the temperature at different times within the time T is used to establish a temperature set, and the temperature set is calibrated as G, then G = {W1, W2, W3, W4, ..., W b};

[0073] S132, normalize the temperature at different times within the time T so that it is within the range of [0, 1], expressed as G'={W'1, W'2, W'3, W'4, ..., W' b}, the specific steps for normalizing the temperature at different times within the time T are as follows:

[0074] S1321, set the temperature set G = {W1, W2, W3, W4, ..., W b The maximum value in} is marked as W max , the minimum value is marked as W min ;

[0075] S1322, use the following formula to convert the original temperature W x Converted to normalized temperature W' x , the specific calculation formula is:

[0076]

[0077] S1323, apply each data in the temperature set G to the above formula to obtain a normalized temperature set G'={W'1, W'2, W'3, W'4, ..., W' b};

[0078] It should be noted that, in this way, the temperature range of the normalization processing is limited to [0, 1]. This processing method helps to eliminate the influence of the absolute value of the data on model training and prediction, while ensuring that all features have similar numerical ranges, thereby improving the stability and convergence speed of the model;

[0079] S133. Use the normalized temperature set G' to construct an isolation forest. The isolation forest consists of z trees, representing F = {T1, T2, T3, T4, ..., T z};

[0080] S134, for each moment temperature W' x , get its y The path length h in xy , expressed as H = {h xy}, where x = 1, 2, 3, ..., b, b is a positive integer, y = 1, 2, 3, ..., z, z is a positive integer, and the path length h xy The steps to obtain are as follows:

[0081] S1341. Starting from the root node, traverse down along the branches in the tree until reaching a leaf node;

[0082] S1342. During the traversal process, each time a non-leaf node is passed, the path length is increased by 1. If the node reached is a leaf node, the path length is added to the depth of the leaf node (i.e., the path length from the root node to the leaf node);

[0083] S1343. Finally, the path length h is obtained. xy That is, the temperature W' at each moment x In the isolated tree y The path length in ;

[0084] S135, calculate temperature W' x The specific calculation of the anomaly score at the corresponding moment is as follows:

[0085]

[0086] in, is the temperature W' at each moment x The average path length cut by the isolated tree, c is the normalization factor of the average path length of the temperature at b moments in time T.

[0087] It should be noted that the normalization factor of the average path length is obtained as follows: the average path length under a given sample size can be calculated by theoretical analysis and deduction of the structure of the isolated tree, and the value of the normalization factor can be obtained accordingly. The average path length under a given sample size can also be calculated using a computer program or algorithm, and then the value of the normalization factor can be obtained based on the calculation result. The method for obtaining the normalization factor of the average path length is not specifically limited here and can be selected according to actual needs.

[0088] S136. Use the sigmoid function to convert the abnormal score Mapped to the range of (0, 1), the temperature anomaly probability is obtained. The specific calculation formula is as follows:

[0089]

[0090] It should be noted that the sigmoid function is used to convert the abnormal score The method of mapping to the range of (0, 1) is as follows: first, the calculated anomaly score is used as the input of the sigmoid function, and then the anomaly score is mapped to the range of (0, 1) through the calculation of the sigmoid function. The mathematical expression of the sigmoid function is f(x) = 1 / (1+exp(-x)), where x is the anomaly score and f(x) is the result after mapping. This ensures that the anomaly score ranges between 0 and 1, which is convenient for quantifying and comparing the degree of anomaly.

[0091] Specifically, the T time period is set as a time sequence including the current time t. During use, the time from time tq to time t can be taken as the T time, where q is a set integer.

[0092] The sensor signal strength fluctuation coefficient calculated in step S113 is used as the sensor signal strength fluctuation coefficient corresponding to the current time t. The transmission rate variation coefficient calculated in step S122 is used as the transmission rate variation coefficient corresponding to the current time t.

[0093] The temperature anomaly probability corresponding to each moment in the time T is calculated by step S136, and then the temperature anomaly probability corresponding to the current moment t is obtained.

[0094] The data analysis model is established by processing the equipment status information and environmental change information when the equipment obtains the dam slope drilling displacement data, that is, the sensor signal intensity fluctuation coefficient S bd , Transmission rate variation coefficient TR vc and the temperature anomaly probability W ycp Establish a data analysis model, generate an outlier hidden danger assessment coefficient, and calibrate the outlier hidden danger assessment coefficient as Q yczyh , according to the formula:

[0095]

[0096] Where r1, r2 and r3 are the sensor signal strength fluctuation coefficients S respectively. bd , Transmission rate variation coefficient TR vc and the temperature anomaly probability W ycp The preset proportional coefficient, and r1, r2 and r3 are all greater than 0;

[0097] It should be noted that the sensor signal strength fluctuation coefficient S bd , Transmission rate variation coefficient TR vc and the temperature anomaly probability W ycp The preset proportional coefficients r1, r2 and r3 are designed to more flexibly adapt to different working conditions and environmental changes in actual monitoring. These proportional coefficients can be adjusted according to specific circumstances to improve the performance and applicability of the monitoring system.

[0098] The pre-set abnormal value hidden danger assessment coefficient threshold (i.e., the first threshold) is calibrated as Q yz , the abnormal value hidden danger assessment coefficient Q generated by the equipment when obtaining the dam slope drilling displacement data yczyh Compare and analyze with the pre-set abnormal value hidden danger assessment coefficient threshold to generate high hidden danger signals and low hidden danger signals. The specific comparison and analysis is as follows:

[0099] If Q yczyh ≤Q yz , generating a low-risk signal;

[0100] If Q yczyh >Q yz , generating a high risk signal.

[0101] When generating a high hidden danger signal, obtain a number of abnormal value hidden danger assessment coefficients that have been generated when the device generates a high hidden danger signal when acquiring the dam slope drilling displacement data, that is, obtain a sequence of abnormal value hidden danger assessment coefficients corresponding to a number of continuous high hidden danger signal sequences containing the currently generated high hidden danger signal; and recalibrate it as a represents the serial numbers of several abnormal value hazard assessment coefficients corresponding to the high hazard signal generated by the equipment when obtaining the dam slope drilling displacement data, including the high hazard signal generated this time and the high hazard signal generated 1 to u-1 times before, a=1, 2, 3, ..., u, where u is a positive integer.

[0102] The average value of several abnormal value hidden danger assessment coefficients is calibrated as Q - , the specific calculation formula is:

[0103] The standard deviation of several outlier hidden danger assessment coefficients is calibrated as Q σ , the specific calculation formula is:

[0104] The equipment obtains several abnormal value hidden danger assessment coefficients generated when acquiring the dam slope drilling displacement data and generating high hidden danger signals, and calibrates the preset average values ​​and preset standard deviations of the several abnormal value hidden danger assessment coefficients as and The average value Q of several abnormal value hidden danger assessment coefficients - and standard deviation Q σ The preset average values ​​of several abnormal value hidden danger assessment coefficients are and preset standard deviation A comparative analysis is conducted, and the specific analysis is as follows:

[0105] like A high-risk signal is generated and transmitted to the mobile terminal, which issues a high-risk early warning prompt. When the device generates a high-risk signal when acquiring the dam slope drilling displacement data, it indicates that there is a high possibility of abnormal values ​​appearing in the monitoring data acquired by the subsequent equipment.

[0106] like and A medium-risk signal is generated and transmitted to the mobile terminal, and a medium-risk early warning prompt is issued through the mobile terminal. When the equipment generates a high-risk signal when acquiring the dam slope drilling displacement data, and generates a medium-risk signal, it indicates that there is a high possibility of abnormal values ​​appearing in the monitoring data acquired by the subsequent equipment;

[0107] like and A low-risk signal is generated and transmitted to the mobile terminal, and a low-risk early warning prompt is issued through the mobile terminal. When the equipment generates a high-risk signal when acquiring the dam slope drilling displacement data and generates a low-risk signal, it indicates that the possibility of abnormal values ​​appearing in the monitoring data acquired by subsequent equipment is low.

[0108] The contents not described in detail in this specification belong to the prior art known to professional and technical personnel in this field.

Claims

1. A method for evaluating dam slope borehole inclination displacement data based on environmental factors, characterized by: The following steps are involved: Real-time acquisition of equipment status information and environmental change information when the equipment is acquiring dam slope drilling and displacement data; The abnormal value hidden danger assessment coefficient is calculated based on the collected equipment status information and environmental change information; Comparing the abnormal value hidden danger assessment coefficient with the first threshold value to generate a high hidden danger signal or a low hidden danger signal; When a low risk signal is generated, it is determined that there is no data anomaly risk; When a high hidden danger signal is generated, an abnormal value hidden danger assessment coefficient sequence corresponding to a plurality of continuous high hidden danger signal sequences including the currently generated high hidden danger signal is obtained; and an assessment result of the data abnormality risk level is determined according to the abnormal value hidden danger assessment coefficient sequence.

2. The method according to claim 1, characterized in that: The device status information includes: a sensor signal strength fluctuation coefficient and a transmission rate variation coefficient.

3. The method according to claim 2, characterized in that: The environmental change information includes temperature anomaly probability.

4. The method according to claim 3, characterized in that: The outlier hazard assessment coefficient Q is calculated using the following formula: yczyh : Where r1, r2 and r3 are the sensor signal strength fluctuation coefficients S respectively. bd , Transmission rate variation coefficient TR vc and the temperature anomaly probability W ycp The preset proportional coefficient, and r1, r2 and r3 are all greater than 0.

5. The method according to claim 1, characterized in that: If the abnormal value hidden danger assessment coefficient is greater than the first threshold, a high hidden danger signal is generated; if the abnormal value hidden danger assessment coefficient is less than or equal to the first threshold, a low hidden danger signal is generated.

6. The method according to claim 1, characterized in that: The process of determining the evaluation result of the data anomaly risk level according to the outlier hidden danger evaluation coefficient sequence includes: Calculate the mean and standard deviation of the outlier hidden danger assessment coefficient series; If the average value of the abnormal value hidden danger assessment coefficient sequence is greater than or equal to the second threshold, a high risk signal is generated; If the average value of the abnormal value hidden danger assessment coefficient sequence is less than the second threshold value, and the standard deviation of the abnormal value hidden danger assessment coefficient sequence is greater than or equal to the third threshold value, a medium risk signal is generated; If the average value of the outlier hidden danger assessment coefficient sequence is less than the second threshold value, and the standard deviation of the outlier hidden danger assessment coefficient sequence is less than the third threshold value, a low risk signal is generated.

7. The method according to claim 1, characterized in that: The process of obtaining the sensor signal strength fluctuation coefficient includes: Collect and calibrate the sensor signal strength at different times within the set time period corresponding to the current time; Calibrate the average value of the sensor signal strength at each moment in the sliding window; The sensor signal strength fluctuation coefficient is calculated based on the difference between the sensor signal strength at different times and the average value of the sensor signal strength at each time in the sliding window.

8. The method according to claim 1, characterized in that: The process of obtaining the transmission rate variation coefficient includes: Collect and calibrate the sensor transmission rate at different times within the set time period corresponding to the current time; The transmission rate variation coefficient is calculated based on the average value of the variation rate of the sensor transmission rate at every two adjacent moments.

9. The method according to claim 1, characterized in that: The process of obtaining the temperature anomaly probability includes: Collect and calibrate the temperature at different times within the set time period corresponding to the current time and perform normalization processing to form a temperature set; Use the temperature set to construct an isolation forest, which consists of several isolated trees; and obtain the path length of each temperature in each isolated tree; The anomaly score at the corresponding moment is calculated based on the path length of each temperature in each isolated tree; Use the sigmoid function to map the anomaly score to the range of and get the temperature anomaly probability.

Images