Water conservancy network safety monitoring system based on big data
By using a big data-based water conservancy network safety monitoring system, intelligent analysis modules and platform evaluation modules are employed to construct behavioral baselines and flow thresholds, thus solving the problem of errors in manual monitoring data and improving the safety and stability of the water conservancy system as well as optimizing resource allocation.
Patent Information
- Application Number
- CN202410996676.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-24
- Publication Date
- 2024-11-12
AI Technical Summary
Existing technologies rely on regular manual monitoring and human-based data analysis, without considering the potential for errors in manually collected data, which could lead to inaccurate data.
A big data-based water conservancy network safety monitoring system is adopted, including an intelligent analysis module, a data acquisition module, and a platform evaluation module. Through the analysis of real-time and historical data, behavioral baselines and flow thresholds are constructed, abnormal situations are identified, and the safety coefficient of the monitoring platform is calculated and adjusted.
This improved the overall quality of the dataset, ensured the normal operation of the water conservancy system, enabled the timely detection and handling of anomalies, enhanced the system's security and stability, optimized resource allocation, and improved the efficiency of safety monitoring.
Smart Images

Figure CN118940134A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of water conservancy information, and relates to a water conservancy network safety monitoring technology, and specifically is a water conservancy network safety monitoring system based on big data. Background Art
[0002] Establishing a water conservancy network flow monitoring system based on big data, changing from the original rule-based security analysis and disposal method to characterizing security data with multiple network elements, characterizing security data with multiple network elements and combining big data technology for analysis and disposal can significantly improve the security of the water conservancy network; changing from the traditional passive monitoring with sample library as the core to the security AI model analysis model as the benchmark, supplemented by real-time flow baseline model analysis, to discover threats and abnormal flow, improve in-depth protection capabilities, and warn of security risks in advance; the system can intelligently dispatch resources for response and processing according to the actual situation of network flow and the severity of the threat, which helps to optimize resource allocation and improve resource utilization efficiency; at the same time, with the continuous evolution and change of network threats, traditional sample library methods are often difficult to keep up with such changes, while AI models can continuously adapt to new threat types and attack methods through continuous learning and optimization; this method can achieve early detection, early warning and early disposal of potential threats, thereby reducing security risks and ensuring the stable operation of the water conservancy network, improving risk identification capabilities, restoring the complete attack process, and timely responding to closed-loop decision-making capabilities.
[0003] The prior art (invention patent application with publication number CN110160506A) discloses a water conservancy unitized supervision system and method based on big data; it includes a data acquisition end, a data storage and processing end, a monitoring application end and a system management end; the data acquisition end is used to collect data from all monitoring equipment in the water conservancy unitized management network and data from investigations, research, measurements and tests by various relevant management agencies; the data storage and processing end is used to store and process big data collected by the data acquisition end; the prior art uses manual regular monitoring of data and human investigation and analysis of data, without considering that manual collection of data will cause errors, which may lead to inaccurate data.
[0004] The present invention provides a water conservancy network safety monitoring system based on big data to solve the above technical problems. Summary of the invention
[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a water conservancy network safety monitoring system based on big data, which is used to solve the technical problem that the prior art uses manual regular monitoring of data and human investigation and analysis of data without considering that manual data collection will cause errors and may lead to inaccurate data.
[0006] To achieve the above-mentioned object, the first aspect of the present invention provides a water conservancy flow monitoring system based on big data, comprising: an intelligent analysis module, and a data acquisition module and a platform evaluation module connected thereto;
[0007] Data collection module: obtain real-time data of the water conservancy network; obtain historical data of the water conservancy network from the database; real-time data includes real-time behavior data, real-time flow data and real-time log data; the data types of historical data and real-time data are the same;
[0008] Intelligent analysis module: build a behavior baseline based on historical behavior data in historical data; determine whether the real-time behavior data is abnormal based on the behavior baseline; if yes, issue an early warning signal to predict the trend of the behavior baseline; if no, continue to analyze the real-time behavior data; and,
[0009] Retrieve historical flow data and determine whether the real-time flow is abnormal based on the historical flow data; if so, issue a notification signal; if not, continue to monitor the real-time flow data; analyze whether the water conservancy network has a fault based on the real-time log data and historical log data; if so, generate a fault alarm signal; if not, continue to analyze the real-time log data;
[0010] Platform evaluation module: detect anomalies and faults based on processing information to obtain detection data, calculate the safety factor of the monitoring platform based on the detection data; adjust the monitoring platform based on the safety factor; the detection data includes accuracy, false alarm rate and missed alarm rate.
[0011] Preferably, constructing a behavior baseline based on historical behavior data in the historical data includes:
[0012] Retrieve historical behavior data from historical data to determine whether the historical behavior data is normal; if so, integrate the corresponding historical behavior data to construct a threshold sequence; if not, determine it as abnormal historical behavior data and remove the corresponding historical behavior data; use the threshold sequence as the behavior baseline.
[0013] The present invention effectively improves the overall quality of the data set by eliminating abnormal historical behavior data; and helps prevent potential losses by constructing a threshold sequence as a benchmark for a real-time monitoring and early warning system.
[0014] Preferably, judging whether the real-time behavior data is abnormal according to the behavior baseline includes:
[0015] Retrieve the real-time behavior data of the real-time data of the water conservancy network, compare the real-time behavior data with the behavior baseline, and determine whether the real-time behavior data deviates from the behavior baseline; if so, determine that the real-time behavior data has an abnormal situation; if not, determine that the real-time behavior data has no abnormal situation.
[0016] The present invention helps to ensure the normal operation of the water conservancy system through real-time behavior data monitoring and abnormality determination; by timely discovering and handling abnormal situations, the system can be prevented from suffering serious damage due to failures, thereby improving the safety and stability of the entire system.
[0017] Preferably, the trend of the predicted behavioral baseline includes:
[0018] Retrieve a number of real-time behavior data, sort the real-time behavior data in chronological order, mark the n-1th real-time behavior data value as Y(n-1), and mark the n-1th exponential smoothing value as S(n-1);
[0019] The nth exponential smoothing value is calculated by the formula S(n)=a×Y(n-1)+(1-a)×S(n-1); where a is a smoothing constant greater than 0 and less than 1; n is a numerical value greater than 1; a series scatter plot is constructed based on the exponential smoothing value; and the prediction results are analyzed based on the trend presented in the series scatter plot.
[0020] It should be noted that the value of the smoothing coefficient α has an important impact on the prediction results. When the data fluctuates greatly, the α value is larger to increase the impact of recent data on the prediction results; when the data fluctuates steadily, the α value is smaller; the actual value of α can be determined by empirical judgment.
[0021] The present invention uses the exponential smoothing method to smooth random fluctuations in time series data by giving higher weights to newer data points and gradually reducing the weights of older data points. The recent change trend of the observed value can be more quickly reflected in the exponential moving average, revealing the potential trend of the data.
[0022] Preferably, analyzing the prediction results according to the trend presented by the series scatter plot includes:
[0023] Retrieve a series scatter plot to determine whether the points on the scatter plot are distributed in a straight line. If so, the change trend of the real-time behavior data is linear; if not, the change trend of the real-time data is nonlinear;
[0024] Determine whether the points on the scatter plot are randomly distributed. If yes, it is determined that the real-time behavior data has a large randomness and the stability of the prediction result is poor. If not, calculate the slope of the curve formed by the points of the scatter plot and predict the strength of the trend based on the slope.
[0025] It should be noted that the larger the slope of the scatter plot, the more obvious the trend is and the more reliable the prediction results are.
[0026] By judging whether the points on the scatter plot are distributed in a straight line, the present invention can intuitively understand whether the real-time behavior data is showing a linear change; by judging whether the points on the scatter plot are randomly distributed and calculating the slope of the curve formed by the scatter plot points, the intensity of the change trend of the real-time behavior data can be quantified.
[0027] Preferably, judging whether the real-time traffic is abnormal based on the historical traffic data includes:
[0028] Retrieve real-time traffic data and historical traffic data; mark the real-time traffic data as Ai, calculate the mean value Bi of the historical traffic data; mark the number of historical traffic data as n;
[0029] By formula Calculate the standard deviation of historical flow data; where i = 1, 2, 3, ..., n; n is a positive integer greater than 0;
[0030] The absolute value of the flow data difference is calculated by the formula SCi=|BZ-Ai|;
[0031] Obtain the abnormal traffic threshold from the database and mark it as T;
[0032] When SCi ≥ T, the real-time traffic is abnormal and a warning signal is issued;
[0033] When SCi<T, there is no abnormality in the real-time traffic.
[0034] The present invention analyzes the standard deviation of historical flow data, calculates the absolute value of the flow data difference, and then determines the relationship between the absolute value of the flow data difference and the abnormal flow threshold, so that the system can send a warning signal when an abnormal situation occurs, allowing managers and operators to take quick measures to prevent or mitigate potential negative impacts.
[0035] Preferably, the analyzing whether a fault occurs in the water conservancy network according to the real-time log data and the historical log data includes:
[0036] Retrieving real-time log data and historical log data, classifying the historical log data to obtain the type of the historical log data; matching the real-time log data with the historical log data to obtain the type corresponding to the real-time log data; wherein the type of the historical log data includes: normal and abnormal;
[0037] Determine whether the type of real-time log data is abnormal; if yes, determine that the water conservancy network has a fault; if no, determine that the water conservancy network has no fault.
[0038] The present invention enables the system to quickly identify whether the current log is of an abnormal type by matching real-time log data with historical log data. This immediacy enables managers to obtain the problem at the early stage of the failure, which is conducive to taking intervention measures before the problem expands, thereby minimizing losses and impacts.
[0039] Preferably, the calculating the safety factor of the monitoring platform according to the detection data includes:
[0040] Retrieve the detection data and mark the accuracy rate, false alarm rate, and missed alarm rate in the detection data as ZL, ZB, and LB respectively;
[0041] The safety factor of the monitoring platform is calculated by the formula AX = θ × [α × (ZL) / (β × ZB + ɡ × LB)]; where α, β, ɡ, and θ are all proportional coefficients greater than 0;
[0042] Obtain a grade mapping table, and match the safety factor with the grade mapping table to obtain the grade of the monitoring platform; wherein the grades of the monitoring platform include excellent, good, qualified, and unqualified.
[0043] It should be noted that the level mapping table is set manually based on experience.
[0044] The present invention quantifies the performance of the monitoring platform through accuracy, false alarm rate and missed alarm rate, making the evaluation results more objective and accurate. This quantitative evaluation method helps to eliminate the bias caused by subjective judgment and improve the fairness and credibility of the evaluation.
[0045] Preferably, the adjusting the monitoring platform based on the safety factor includes:
[0046] When the safety factor is excellent, continuous monitoring and regular system maintenance and performance checks on the monitoring platform are performed;
[0047] When the safety factor is good, analyze the key factors that cause the safety factor to be less than excellent; optimize the key factors;
[0048] When the safety factor is qualified, the monitoring platform will be improved and strengthened;
[0049] When the safety factor is unqualified, the operation of the monitoring platform shall be stopped immediately, and the existing problems of the monitoring platform shall be fully checked and rectified; after the rectification is completed, the safety factor shall be re-evaluated.
[0050] The present invention adjusts the monitoring platform based on the safety factor, so that managers can reasonably allocate resources according to actual conditions; for monitoring platforms with excellent performance, the existing configuration can be maintained or fine-tuned to maintain efficient operation; while for monitoring platforms with problems, more resources and energy need to be invested in improvement and optimization. This targeted resource allocation helps to improve the overall safety monitoring efficiency.
[0051] Compared with the prior art, the present invention has the following beneficial effects:
[0052] 1. The present invention includes three modules: data acquisition module, intelligent analysis module and platform evaluation module. The data acquisition module collects data to cover all links and aspects of the water conservancy network, ensures the comprehensiveness of the data and provides a solid foundation for subsequent analysis. The real-time data collection ensures the timeliness of the data, which helps to timely discover and respond to potential safety risks. The intelligent analysis module determines whether anomalies occur in the data types of real-time data and historical data, and makes corresponding processing, which is conducive to timely discovery and processing of abnormal situations, thereby improving the safety and stability of the entire system. The safety factor is calculated by the platform evaluation module, and the monitoring platform is adjusted according to the safety factor, which is conducive to setting targeted resource allocation, thereby improving the overall safety monitoring efficiency.
[0053] 2. Through the collaborative work of the three major modules, the present invention enables the water conservancy network safety monitoring system to have powerful data collection, analysis and evaluation capabilities, and can comprehensively and accurately monitor the safety status of the water conservancy network; when safety risks or problems are discovered, the system can respond quickly and generate emergency plans to provide managers with timely decision-making support and reduce losses and impacts; based on the results of big data and intelligent analysis, managers can make more scientific and reasonable decisions to promote the safe and efficient operation of the water conservancy network. BRIEF DESCRIPTION OF THE DRAWINGS
[0054] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0055] Figure 1 is a schematic diagram of the modules included in the present invention;
[0056] Figure 2 It is a schematic diagram of the specific process of safety monitoring of the present invention;
[0057] Figure 3 It is a schematic diagram of the specific steps of the intelligent analysis of the present invention;
[0058] Figure 4 Schematic diagram of specific steps for evaluating the monitoring platform of the present invention. DETAILED DESCRIPTION
[0059] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0060] See also Figure 1-Figure 2 The first embodiment of the present invention provides a water conservancy network safety monitoring system based on big data, including:
[0061] Data collection module: obtain real-time data of the water conservancy network; obtain historical data of the water conservancy network from the database; real-time data includes real-time behavior data, real-time flow data and real-time log data; the data types of historical data and real-time data are the same;
[0062] Intelligent analysis module: build a behavior baseline based on historical behavior data in historical data; determine whether the real-time behavior data is abnormal based on the behavior baseline; if yes, issue an early warning signal to predict the trend of the behavior baseline; if no, continue to analyze the real-time behavior data; and,
[0063] Retrieve historical flow data and determine whether the real-time flow is abnormal based on the historical flow data; if so, issue a notification signal; if not, continue to monitor the real-time flow data; analyze whether the water conservancy network has a fault based on the real-time log data and historical log data; if so, generate a fault alarm signal; if not, continue to analyze the real-time log data;
[0064] Platform evaluation module: detect anomalies and faults based on processing information to obtain detection data, calculate the safety factor of the monitoring platform based on the detection data; adjust the monitoring platform based on the safety factor; the detection data includes accuracy, false alarm rate and missed alarm rate.
[0065] See also Figure 3 ,The specific steps of intelligent analysis are to retrieve the historical behavior data in the historical data and determine whether the historical behavior data is normal; if it is, the corresponding historical behavior data is integrated to construct a threshold sequence; if not, it is determined to be abnormal historical behavior data and the corresponding historical behavior data is removed; the threshold sequence is used as the behavior baseline;
[0066] Retrieve the real-time behavior data of the real-time data of the water conservancy network, compare the real-time behavior data with the behavior baseline, and determine whether the real-time behavior data deviates from the behavior baseline; if yes, determine that the real-time behavior data has an abnormal situation; if no, determine that the real-time behavior data has no abnormal situation; retrieve a number of real-time behavior data, sort the real-time behavior data in chronological order, mark the n-1th real-time behavior data value as Y(n-1), and mark the n-1th exponential smoothing value as S(n-1);
[0067] The nth exponential smoothing value is calculated by the formula S(n)=a×Y(n-1)+(1-a)×S(n-1); wherein a is a smoothing constant greater than 0 and less than 1; n is a number greater than 1; a series scatter plot is constructed according to the exponential smoothing value; the prediction results are analyzed according to the trend presented by the series scatter plot; the series scatter plot is retrieved to determine whether the points on the scatter plot are distributed in a straight line, if yes, the change trend of the real-time behavior data is linear; if not, the change trend of the real-time data is determined to be nonlinear;
[0068] Determine whether the points on the scatter plot are randomly distributed. If yes, it is determined that the real-time behavior data has a large randomness and the stability of the prediction result is poor. If no, calculate the slope of the curve formed by the points of the scatter plot and predict the strength of the trend based on the slope;
[0069] Retrieve real-time traffic data and historical traffic data; mark the real-time traffic data as Ai, calculate the mean value Bi of the historical traffic data; mark the number of historical traffic data as n; and calculate the value of the historical traffic data by the formula Calculate the standard deviation of historical flow data; where i = 1, 2, 3, ..., n; n is a positive integer greater than 0;
[0070] The absolute value of the flow data difference is calculated by the formula SCi=|BZ-Ai|;
[0071] Obtain the abnormal traffic threshold from the database and mark it as T;
[0072] When SCi ≥ T, the real-time traffic is abnormal and a warning signal is issued;
[0073] When SCi<T, there is no abnormality in the real-time traffic.
[0074] Retrieving real-time log data and historical log data, classifying the historical log data to obtain the type of the historical log data; matching the real-time log data with the historical log data to obtain the type corresponding to the real-time log data; wherein the type of the historical log data includes: normal and abnormal;
[0075] Determine whether the type of real-time log data is abnormal; if yes, determine that the water conservancy network has a fault; if no, determine that the water conservancy network has no fault.
[0076] For example, retrieve a set of historical behavior data from the historical data, remove the abnormal historical behavior data, integrate the remaining normal historical behavior data to construct a threshold sequence; and use the threshold sequence as the behavior baseline; retrieve the real-time behavior data of the water conservancy network real-time data, compare the real-time behavior data with the behavior baseline, determine whether the real-time behavior data deviates from the behavior baseline, and sort the real-time behavior data that does not deviate from the behavior baseline in chronological order; the five real-time behavior data that do not deviate from the behavior baseline are: 1, 2, 4, 6, 8; mark the first exponential smoothing value S(1) as 1, and set a to 0.5; through the formula S(n) = a
[0077] ×Y(n-1)+(1-a)×S(n-1) to calculate: S(2)=1, S(3)=1.5, S(4)=2.75, S(5)=4.375, S(6)=6.1825; construct a series scatter plot based on the exponential smoothing value; the scatter plot is found to be distributed in a straight line;
[0078] There is a set of historical flow data, which are 6, 5, 4, and 5 respectively; the mean value of the historical flow data is Bi = 5. The calculated standard deviation of the historical flow data is 1.414;
[0079] The absolute value of the flow data difference is calculated by the formula SCi=|BZ-Ai|; among them, SC1=4.586, SC2=3.586, SC3=2.586, SC4=3.586, and the abnormal flow threshold T=4 is obtained from the database; if SC2, SC3 and SC4 are all smaller than the threshold, then the corresponding real-time flow has no abnormal situation; if SC1 is greater than the threshold, then the corresponding real-time flow has an abnormal situation and a warning signal is issued.
[0080] See also Figure 4 , the specific steps of monitoring platform evaluation, retrieve the detection data, mark the accuracy rate, false alarm rate and missed alarm rate in the detection data as ZL, ZB, LB respectively; calculate the safety factor of the monitoring platform through the formula AX = θ × [α × (ZL) / (β × ZB + ɡ × LB)]; where α, β, ɡ, θ are all proportional coefficients greater than 0;
[0081] Obtain a level mapping table, and match the safety factor with the level mapping table to obtain the level of the monitoring platform; wherein the levels of the monitoring platform include excellent, good, qualified, and unqualified;
[0082] When the safety factor is excellent, continuous monitoring and regular system maintenance and performance checks on the monitoring platform are performed;
[0083] When the safety factor is good, analyze the key factors that cause the safety factor to be less than excellent; optimize the key factors;
[0084] When the safety factor is qualified, the monitoring platform will be improved and strengthened;
[0085] When the safety factor is unqualified, the operation of the monitoring platform shall be stopped immediately, and the existing problems of the monitoring platform shall be fully checked and rectified; after the rectification is completed, the safety factor shall be re-evaluated.
[0086] For example, the monitoring platform at two times TI and T2 is tested, and two sets of test data are obtained, α=100, β=100, ɡ=100, θ=20. The specific analysis results are as follows:
[0087] At time T1: accuracy rate ZL = 73%, false alarm rate ZB = 17%, missed alarm rate LB = 10%, then safety factor AX = 54.07; 54.07 < 60, the safety factor of the monitoring platform is judged to be unqualified;
[0088] At time T2: the accuracy rate ZL=82%, the false alarm rate ZB=15%, the missed alarm rate LB=3%, then the safety factor AX=91.1; when 91.1>80, the safety factor of the monitoring platform is judged to be excellent.
[0089] It should be noted that the sum of the accuracy rate, false alarm rate, and missed alarm rate is 100%; when the accuracy rate is 100%, there is no need to calculate the safety factor through a formula, and the safety factor of the monitoring platform can be directly judged as excellent;
[0090] When AX ≥ 80, the safety factor of the monitoring platform is judged to be excellent;
[0091] When 70≤AX<80, the safety factor of the monitoring platform is judged to be good;
[0092] When 60≤AX<70, the safety factor of the monitoring platform is determined to be qualified;
[0093] When AX<60, the safety factor of the monitoring platform is judged to be unqualified.
[0094] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.
[0095] The working principle of the present invention is as follows: the present invention obtains real-time data of the water conservancy network; obtains historical data of the water conservancy network from a database; the data type of the historical data is the same as that of the real-time data; constructs a behavior baseline based on the historical behavior data in the historical data; determines whether the real-time behavior data is abnormal based on the behavior baseline; retrieves historical flow data, and determines whether the real-time flow is abnormal based on the historical flow data; analyzes whether a fault occurs in the water conservancy network based on the real-time log data and the historical log data; detects anomalies and faults based on processing information to obtain detection data, and calculates the safety factor of the monitoring platform based on the detection data; and adjusts the monitoring platform based on the safety factor.
[0096] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.
Claims
1. A water conservancy network safety monitoring system based on big data, characterized in that: include: Intelligent analysis module, and the data acquisition module and platform evaluation module connected to it; Data acquisition module: obtain real-time data from the water conservancy network; Obtain historical data of the water conservancy network from the database; real-time data includes real-time behavior data, real-time flow data and real-time log data; the data types of historical data and real-time data are the same; Intelligent analysis module: build a behavior baseline based on historical behavior data in historical data; determine whether the real-time behavior data is abnormal based on the behavior baseline; if yes, issue an early warning signal to predict the trend of the behavior baseline; if no, continue to analyze the real-time behavior data; and, Retrieve historical flow data and determine whether the real-time flow is abnormal based on the historical flow data; if so, issue a notification signal; if not, continue to monitor the real-time flow data; analyze whether the water conservancy network has a fault based on the real-time log data and historical log data; if so, generate a fault alarm signal; if not, continue to analyze the real-time log data; Platform evaluation module: detect anomalies and faults based on processing information to obtain detection data, calculate the safety factor of the monitoring platform based on the detection data; adjust the monitoring platform based on the safety factor; the detection data includes accuracy, false alarm rate and missed alarm rate.
2. According to the big data-based water conservancy network safety monitoring system of claim 1, it is characterized in that: The constructing of a behavior baseline based on the historical behavior data in the historical data includes: Retrieve historical behavior data from historical data to determine whether the historical behavior data is normal; if so, integrate the corresponding historical behavior data to construct a threshold sequence; if not, determine it as abnormal historical behavior data and remove the corresponding historical behavior data; use the threshold sequence as the behavior baseline.
3. According to the big data-based water conservancy network safety monitoring system of claim 1, it is characterized in that: The determining whether the real-time behavior data is abnormal based on the behavior baseline includes: Retrieve the real-time behavior data of the real-time data of the water conservancy network, compare the real-time behavior data with the behavior baseline, and determine whether the real-time behavior data deviates from the behavior baseline; if so, determine that the real-time behavior data has an abnormal situation; if not, determine that the real-time behavior data has no abnormal situation.
4. According to the big data-based water conservancy network safety monitoring system of claim 1, it is characterized in that: The trend of the predicted behavior baseline includes: retrieving a plurality of real-time behavior data, sorting the plurality of real-time behavior data in chronological order, marking the n-1th real-time behavior data value as Y(n-1), and marking the n-1th exponential smoothing value as S(n-1); The nth exponential smoothing value is calculated by the formula S(n)=a×Y(n-1)+(1-a)×S(n-1); where a is a smoothing constant greater than 0 and less than 1; n is a numerical value greater than 1; a series scatter plot is constructed based on the exponential smoothing value; and the prediction results are analyzed based on the trend presented in the series scatter plot.
5. The water conservancy network safety monitoring system based on big data according to claim 4 is characterized in that: The analysis of the prediction results according to the trend presented by the series scatter plot includes: Retrieve a series scatter plot to determine whether the points on the scatter plot are distributed in a straight line. If so, the change trend of the real-time behavior data is linear; if not, the change trend of the real-time data is nonlinear; Determine whether the points on the scatter plot are randomly distributed. If yes, it is determined that the real-time behavior data has a large degree of randomness and the stability of the prediction results is poor. If not, calculate the slope of the curve formed by the points of the scatter plot and predict the strength of the trend based on the slope.
6. The water conservancy network safety monitoring system based on big data according to claim 1 is characterized in that: The determining whether the real-time traffic is abnormal based on the historical traffic data includes: Retrieve real-time traffic data and historical traffic data; mark the real-time traffic data as Ai, calculate the mean value Bi of the historical traffic data; mark the number of historical traffic data as n; By formula Calculate the standard deviation of historical flow data; where i = 1, 2, 3, ..., n; n is a positive integer greater than 0; The absolute value of the flow data difference is calculated by the formula SCi=|BZ-Ai|; Obtain the abnormal traffic threshold from the database and mark it as T; When SCi ≥ T, the real-time traffic is abnormal and a warning signal is issued; When SCi<T, there is no abnormality in the real-time traffic.
7. The water conservancy network safety monitoring system based on big data according to claim 1 is characterized in that: The analyzing whether a water conservancy network fails according to real-time log data and historical log data includes: Retrieving real-time log data and historical log data, classifying the historical log data to obtain the type of the historical log data; matching the real-time log data with the historical log data to obtain the type corresponding to the real-time log data; wherein the type of the historical log data includes: normal and abnormal; Determine whether the type of real-time log data is abnormal; if yes, determine that the water conservancy network has a fault; if no, determine that the water conservancy network has no fault.
8. The water conservancy network safety monitoring system based on big data according to claim 1 is characterized in that: The step of calculating the safety factor of the monitoring platform according to the detection data includes: Retrieve the detection data and mark the accuracy rate, false alarm rate, and missed alarm rate in the detection data as ZL, ZB, and LB respectively; The safety factor of the monitoring platform is calculated by the formula AX = θ × [α × (ZL) / (β × ZB + ɡ × LB)]; where α, β, ɡ, and θ are all proportional coefficients greater than 0; Obtain a grade mapping table, and match the safety factor with the grade mapping table to obtain the grade of the monitoring platform; wherein the grades of the monitoring platform include excellent, good, qualified, and unqualified.
9. The water conservancy network safety monitoring system based on big data according to claim 1 is characterized in that: The adjustment of the monitoring platform based on the safety factor includes: When the safety factor is excellent, continuous monitoring and regular system maintenance and performance checks on the monitoring platform are performed; When the safety factor is good, analyze the key factors that cause the safety factor to be less than excellent; optimize the key factors; When the safety factor is qualified, the monitoring platform will be improved and strengthened; When the safety factor is unqualified, the operation of the monitoring platform shall be stopped immediately, and the existing problems of the monitoring platform shall be fully checked and rectified; after the rectification is completed, the safety factor shall be re-evaluated.
