Water conservancy data analysis method and system for flood control monitoring

Abstract

The invention discloses a water conservancy data analysis method and system for flood control monitoring, and belongs to the technical field of flood control monitoring. The system comprises a flood control monitoring module, a data analysis module, an operation management module and a risk early warning module. The flood control monitoring module is used for collecting a GIS map, submission information and return data of each monitoring device in a water area; the data analysis module is used for identifying abnormal equipment according to the returned data and analyzing the submitted information to establish a prediction set for the abnormal equipment; the operation management module is used for calculating a deviation index of each prediction set so as to establish a reference scheme for each monitoring device; and the risk early warning module is used for analyzing the flood control risk of each reference scheme so as to carry out early warning on the water area corresponding to each reference scheme. According to the invention, the data integrity in a heavy rainfall environment is guaranteed through combination of dual-communication redundancy transmission and intelligent analysis, accurate grading of flood control threats is realized by adopting a multi-index risk assessment algorithm, and the monitoring reliability and emergency response efficiency in extreme weather are improved.

Description

Technical Field

[0001] This invention relates to the field of flood control monitoring technology, specifically to a water conservancy data analysis method and system for flood control monitoring. Background Technology

[0002] Against the backdrop of frequent extreme heavy rainfall events, flood control monitoring systems face unprecedented challenges. Heavy rainfall not only triggers direct flood risks such as sudden rises in water levels and river overloading, but also severely interferes with the network communication quality of monitoring equipment, leading to interruptions in critical water conservancy data transmission or the loss of some parameters. Therefore, ensuring real-time acquisition, highly reliable transmission, and intelligent repair of monitoring data under heavy rainfall conditions has become a core issue in improving the robustness of flood control monitoring systems.

[0003] Currently, traditional flood monitoring systems have several drawbacks in heavy rainfall scenarios: First, monitoring equipment relying on a single communication method is prone to signal interruption due to rainstorm interference, resulting in a sharp drop in data transmission success rate. Furthermore, the lack of a dual-channel redundant transmission mechanism prevents timely data retransmission via backup links such as satellite communication. Second, methods for repairing missing parameters are crude, often using fixed thresholds or historical averages as substitutes, without dynamically calculating missing values ​​based on the spatial correlation and historical relationships of hydraulic parameters under heavy rainfall. This leads to repaired data deviating from the actual scenario. Finally, risk assessment models have low tolerance for missing parameters; when some key parameters are lost, the model may misjudge the flood's evolution trend, exacerbating decision-making risks. Existing technologies struggle to balance communication reliability, data integrity, and analytical accuracy under heavy rainfall conditions, necessitating innovative solutions. Summary of the Invention

[0004] The purpose of this invention is to provide a water conservancy data analysis method and system for flood control monitoring, so as to solve the problems mentioned in the background art.

[0005] To address the aforementioned technical problems, this invention provides a method for analyzing water conservancy data for flood control monitoring, the method comprising: S100, collect GIS maps and report information, as well as the data returned by various monitoring devices in the water area; S200: Identify abnormal devices based on the returned data, and analyze the reported information to establish a prediction set for abnormal devices; S300: Calculate the deviation index of each prediction set to establish a reference scheme for each monitoring device; S400: Analyze the flood control risks of each reference scheme, and thus issue early warnings for the water areas corresponding to each reference scheme.

[0006] Furthermore, in S100, the GIS map includes the location and coverage of the water area, as well as the location of each monitoring device; the reporting information includes the reporting records of each monitoring device, and each reporting record includes the reporting time, reporting method, reporting template, success rate, and water parameters; the reporting methods include cellular network and satellite communication; each monitoring device periodically collects water parameters according to a preset time interval and transmits them back to the monitoring center; the reporting template includes all water parameters that should be collected; the success rate refers to the ratio of the number of water parameters received at the reporting time to the number of water parameters that should be collected; the transmitted data includes all water parameters.

[0007] Furthermore, the S200 includes: S201. Obtain the water conservancy parameters returned by each monitoring device, count the number of water conservancy parameters received and calculate the success rate after comparing with the reporting template, and set the status of monitoring devices with a success rate of less than 100% as abnormal. S202. Abnormal devices simultaneously transmit water conservancy parameters using both cellular network and satellite communication. When the success rate is 100%, the abnormal status of the corresponding abnormal device is cancelled. S203. Obtain the reporting records of each abnormal device, and use the various water conservancy parameters in the reporting records as independent variables and the success rate as dependent variables to fit the cellular relationship and satellite relationship of the corresponding abnormal device. S204. Analyze the reporting template of abnormal equipment, and treat the received hydraulic parameters as known parameters and the unreceived hydraulic parameters as unknown parameters. S205. Establish P prediction sets for abnormal devices. All unknown parameters are placed in each prediction set. Assign values ​​to each unknown parameter in each prediction set according to the cellular relation and the satellite relation.

[0008] Furthermore, S203 includes: S2031. Classify all reporting records of the abnormal device E1 according to the reporting method; count the number T of all reporting records in the cellular network category x. x And the total number of reporting records T in satellite communication category y. y ; S2032. Analyze each reporting record in cellular network class x, using various water conservancy parameters as independent variables and success rate as the dependent variable, and package them into a sample. x The samples were used to form a cellular training set; S2033. Analyze each reporting record in satellite communication category y, using various water conservancy parameters as independent variables and success rate as the dependent variable, and package them into a sample. y The sample data was used to construct a satellite training set. S2034. Substitute the cellular training set and satellite training set into the formula respectively for training, and fit the cellular relationship and satellite relationship of the abnormal device E1; the formula is as follows: ; In the formula, H is the success rate, β0 is the intercept term, k is the number of all hydraulic parameters, and β i Z is the influence coefficient of the i-th hydraulic parameter. i Let i be the i-th water conservancy parameter; S2035, and so on, respectively, to obtain cellular and satellite relational formulas for each abnormal device.

[0009] Furthermore, S205 includes: S2051. Count the number of known parameters and unknown parameters under the two reporting methods respectively, and calculate the success rate G of the cellular network reporting method. x And the success rate of satellite communication reporting method G y ; S2052, All known parameters and success rate G x Substituting these into the cellular relational formula yields the first relational formula; all known parameters and success rate G. y Substituting into the satellite relation, we obtain the second relation; S2053. Assign values ​​to each unknown parameter in each prediction set. The values ​​of all unknown parameters in each prediction set simultaneously satisfy the first relation and the second relation. The values ​​of all unknown parameters in different prediction sets are not completely the same.

[0010] Furthermore, the S300 includes: S301. Mark the location of each monitoring device on the GIS map and associate the abnormal device with the monitoring device that is closest to it and is not in an abnormal state; S302. Obtain all reporting records for each pair of associated devices, and match the reporting records with the same reporting time pairwise; analyze the values ​​of each unknown parameter in each pair of matched records, and fit the influence relationship of each unknown parameter; S303. Analyze the various water conservancy parameters received by the monitoring equipment under the associated equipment, substitute them into the corresponding influence relationship formulas, calculate the predicted values ​​of various unknown parameters of the abnormal equipment, and then calculate the deviation index of each prediction set under the abnormal equipment. S304. For abnormal equipment, the prediction set of the minimum deviation index is selected as the reference set. Based on the values ​​of each known parameter and the values ​​of each unknown parameter in the reference set, a reference scheme is established for the corresponding abnormal equipment. S305. The monitoring equipment establishes a reference scheme based on the values ​​of the various water conservancy parameters that have been received.

[0011] Furthermore, each pair of associated devices includes one abnormal device and one monitoring device; each pair of matching records includes two reporting records, which are reported at the same time and belong to the corresponding abnormal device and monitoring device respectively; S302 includes: S3021. Count the number of matching record pairs d under the associated device Q1, obtain the value of unknown parameter C0 in the two reporting records in each pair of matching records, and package them as independent and dependent variables into a sample. S3022. Input these d samples into the linear regression model for training and fit the relationship between the unknown parameter C0 and the model: M=θ+vN; where M is the dependent variable, N is the independent variable, θ is the intercept, and v is the regression coefficient. S3023, and so on, are respectively the fitting influence formulas of other unknown parameters of abnormal equipment in associated equipment Q1, and the fitting influence formulas of other unknown parameters of abnormal equipment in each pair of associated equipment.

[0012] Furthermore, S303 includes: obtaining the predicted values ​​of various unknown parameters of the abnormal device, and the values ​​of various unknown parameters in each prediction set, and substituting them into the formula to calculate the deviation index BI for each prediction set: ; In the formula, α is a constant, b is the number of all unknown parameter terms under abnormal equipment, and YX a PRE is the weighting coefficient for the a-th unknown parameter. a SET is the predicted value of the a-th unknown parameter. a Let be the value of the a-th unknown parameter in the prediction set.

[0013] Furthermore, in S400, a reference scheme for each abnormal device and monitoring device is obtained, and the values ​​of various hydraulic parameters in the reference scheme are analyzed. A multi-index weighted comprehensive scoring algorithm is used to analyze the flood risk level of each reference scheme. The reference schemes are ranked in descending order of flood risk level, thereby providing early warning for the water areas corresponding to each reference scheme.

[0014] The present invention also provides a water conservancy data analysis system for flood control monitoring, including a flood control monitoring module, a data analysis module, an operation management module, and a risk early warning module.

[0015] The flood control monitoring module is used to collect GIS maps and report information, as well as the data transmitted back from various monitoring devices in the water area.

[0016] The system collects GIS maps, reports information, and data transmitted from monitoring equipment. Reporting records are dynamically updated through a "preliminary generation" and "post-update supplementation" mechanism to ensure consistency between the number of water conservancy parameters and the template. Data transmission utilizes both cellular networks and satellite communication, and malfunctioning equipment must use both methods simultaneously.

[0017] By employing multi-source data acquisition and dynamic data completion mechanisms, the integrity and real-time nature of the data are ensured, providing a reliable foundation for subsequent analysis. The dual-communication design enhances the redundancy of data transmission from malfunctioning devices, reducing the risk of data loss due to a single communication failure.

[0018] The data analysis module is used to identify abnormal devices based on the returned data and to analyze the reported information to build a prediction set for abnormal devices.

[0019] Abnormal devices are identified using a success rate threshold. Based on cellular and satellite communication reporting records, cellular and satellite training sets are constructed respectively, and a relational formula is fitted. Multiple prediction sets are then generated for the unknown parameters of the abnormal devices using this formula, requiring the predicted values ​​to simultaneously satisfy the relational constraints of both communication methods.

[0020] By quantifying the impact of hydraulic parameters on success rates using machine learning algorithms, multi-scenario predictions for missing data from faulty equipment are achieved, improving the scientific rigor of data repair. Dual relational constraints ensure that the prediction results conform to actual patterns under different communication methods, enhancing the predictive validity.

[0021] The operation management module is used to calculate the deviation index of each prediction set, thereby establishing a reference scheme for each monitoring device.

[0022] The abnormal equipment is associated with nearby normal equipment, and records with the same reporting time are matched. A linear regression model is used to fit the influence relationship of unknown parameters. The reliability of each prediction set is evaluated using the deviation index formula, and finally the set with the smallest deviation is selected as the reference scheme.

[0023] By leveraging the correlation between spatially adjacent devices, predicted values ​​are optimized, reducing errors from isolated predictions. A deviation index is used to quantify and assess prediction quality, ensuring that the reference scheme closely reflects real-world scenarios and providing high-confidence data support for flood control decision-making.

[0024] The risk warning module is used to analyze the flood control risks of each reference scheme, thereby providing early warnings for the corresponding water areas of each reference scheme.

[0025] A multi-indicator weighted comprehensive scoring algorithm, combined with the water conservancy parameter values ​​in the reference scheme, is used to calculate the flood control risk level of each scheme and rank them according to risk level. Real-time images and water conservancy parameter details of high-risk water areas are displayed.

[0026] Quantitative risk assessment helps to quickly locate flood control weaknesses, enabling managers to intuitively grasp the risk distribution and prioritize high-threat areas, thereby enhancing emergency response capabilities.

[0027] Compared with the prior art, the beneficial effects achieved by the present invention are: Dual-communication redundancy ensures data integrity: The technical solution forces malfunctioning devices to simultaneously transmit data via both cellular network and satellite communication. When heavy rainfall causes low transmission quality in a single communication channel, missing parameters are supplemented through the backup link, significantly improving data transmission success rate. Compared to existing reliance on a single communication channel, this effectively reduces the risk of data loss under extreme weather conditions.

[0028] Dynamic data imputation and scientific prediction: Based on historical reporting records, a relationship between water conservancy parameters and communication success rate is constructed. Combined with spatial correlation analysis of nearby normal equipment, multiple sets of predictions are dynamically generated to fill in missing data. Compared with traditional manual interpolation or fixed threshold imputation, this method utilizes machine learning and spatial patterns to improve the accuracy and rationality of data repair.

[0029] Multi-parameter collaborative risk assessment: Employing a multi-indicator weighted scoring algorithm, this system dynamically quantifies flood control risks by integrating multiple parameters and prioritizing them. It displays real-time images of high-risk areas, providing more comprehensive support for quickly locating threats and developing precise emergency response strategies compared to traditional single-threshold alarms.

[0030] Closed-loop system linkage and adaptive optimization: Through closed-loop collaboration of flood monitoring, data analysis, operation management, and risk early warning modules, communication strategies, forecast parameters, and risk assessment rules are iterated in real time. Compared with existing isolated module designs, the system's ability to adapt to changes in heavy rainfall environments is significantly enhanced, and overall response efficiency and decision reliability are improved simultaneously. Attached Figure Description

[0031] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart illustrating a water conservancy data analysis method for flood control monitoring according to the present invention. Figure 2 This is a flowchart of abnormal data processing in a water conservancy data analysis method for flood control monitoring according to the present invention; Figure 3 This is a data filling and prediction model diagram of a water conservancy data analysis method for flood control monitoring according to the present invention; Figure 4 This is a schematic diagram of the structure of a water conservancy data analysis system for flood control monitoring according to the present invention. Detailed Implementation

[0032] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0033] Please see Figure 1 This invention provides a water conservancy data analysis method for flood control monitoring, comprising: S100 collects GIS maps and reports information, as well as data transmitted back from various monitoring devices in the water area.

[0034] The GIS map includes the location and coverage of the water area, as well as the location of each monitoring device. The reported information includes the reporting records of each monitoring device, and each reporting record includes the reporting time, reporting method, reporting template, success rate, and water parameters.

[0035] Reporting methods include cellular networks and satellite communication. Each monitoring device periodically collects water parameters according to preset intervals and transmits them back to the monitoring center. The reporting template includes all water parameters that should be collected.

[0036] Success rate refers to the ratio of the number of water conservancy parameters received at the reporting time to the number of water conservancy parameters that should have been collected. The returned data includes all water conservancy parameters.

[0037] By integrating GIS maps, reported information, and data transmitted from monitoring equipment, the integrity and real-time nature of multi-source data (such as location, water parameters, and communication methods) required for flood control monitoring are ensured.

[0038] The reporting records include those generated in the early stages and those supplemented later: Preliminary generation: The monitoring center calculates the success rate based on the various water conservancy parameters received by the monitoring equipment within the reporting time, compares them with the reporting template, and generates reporting records by combining the reporting time and reporting type.

[0039] Subsequent Supplements: Based on changes in the status of the monitoring equipment, re-acquire and fill in the various water conservancy parameters that should have been collected but were not received within the corresponding reporting time in each reporting record, ensuring that the number of water conservancy parameters in each reporting record is the same as the number of water conservancy parameters that should be collected in the reporting template.

[0040] By dynamically filling in missing hydraulic parameters through a "preliminary generation" and "post-supplementation" mechanism, the problem of incomplete data caused by communication failures is avoided, providing a reliable data foundation for subsequent analysis.

[0041] S200: Identify abnormal devices based on the returned data, and analyze the reported information to establish a prediction set for the abnormal devices. Please refer to [link / reference]. Figure 2Specifically, it includes: S201. Obtain the water parameters returned by each monitoring device, count the number of water parameters received and calculate the success rate after comparing with the reporting template, and set the status of monitoring devices with a success rate of less than 100% as abnormal.

[0042] S202. Abnormal devices simultaneously transmit water conservancy parameters using both cellular network and satellite communication. When the success rate is 100%, the abnormal status of the corresponding abnormal device is cancelled.

[0043] Abnormal equipment always transmits hydraulic parameters using two reporting methods simultaneously, and the success rate of the combined reporting methods is less than 100%. Normal monitoring equipment transmits hydraulic parameters using one or two reporting methods, and the success rate is 100%.

[0044] The status analysis and settings of the monitoring equipment will immediately provide feedback to the corresponding monitoring equipment if the success rate is not 100% after each received data transmission, and require it to be re-uploaded in another reporting method.

[0045] If the other method has a 100% success rate, then the same reporting method will be used to upload data next time, and the abnormal status of the monitoring device will be canceled.

[0046] If the success rate is still less than 100% after combining the two methods, the status of the monitoring device will be set to abnormal, and the data will be uploaded using both methods in the next attempt.

[0047] If the success rate is 100% when the other method is combined with the current method, then the abnormal status of the monitoring device is cancelled, and the two methods will be used together to upload data in the next instance.

[0048] By using a success rate threshold (<100%), abnormal devices can be quickly identified, and these devices can be forced to use dual-channel redundant transmission via cellular network and satellite communication to reduce the risk of failure of a single communication method.

[0049] S203. Obtain the reporting records of each abnormal device. Using the various hydraulic parameters in the reporting records as independent variables and the success rate as the dependent variable, fit the cellular and satellite relationship formulas for the corresponding abnormal devices. Specifically, this includes: S2031, please refer to Figure 3 All reporting records of the abnormal device E1 are categorized according to the reporting method. The number of reporting records T in each cellular network category x is then counted. x And the total number of reporting records T in satellite communication category y. y .

[0050] S2032. Analyze each reporting record in cellular network class x, using various water conservancy parameters as independent variables and success rate as the dependent variable, and package them into a sample.x The samples were used to form a cellular training set.

[0051] S2033. Analyze each reporting record in satellite communication category y, using various water conservancy parameters as independent variables and success rate as the dependent variable, and package them into a sample. y The sample data were used to form a satellite training set.

[0052] S2034. Substitute the cellular training set and satellite training set into the formulas respectively for training, and fit to obtain the cellular and satellite relational expressions for the abnormal device E1. The formulas are as follows: ; In the formula, H is the success rate, β0 is the intercept term, k is the number of all hydraulic parameters, and β i Z is the influence coefficient of the i-th hydraulic parameter. i Let be the i-th hydraulic parameter.

[0053] Water parameters include meteorological parameters, sediment parameters, water quality parameters, and hydrological parameters.

[0054] Meteorological parameters include rainfall, evaporation, and surface air temperature and humidity. Sediment parameters include sediment content and sediment transport rate. Water quality parameters include turbidity, conductivity, pH value, and nutrient concentration. Hydrological parameters include water level, flow velocity, discharge rate, water temperature, and water depth.

[0055] Different monitoring devices collect different types of water parameters and therefore use different collection methods, including surface collection and underwater collection.

[0056] Real-time monitoring of these hydraulic parameters relies on sensors combined with wireless communication, but the hydraulic environment itself can affect signal transmission. Specific ways in which this influence occurs include: Water body itself: water depth, water quality (salinity) → electromagnetic wave attenuation (e.g., a 2.4 GHz signal can penetrate <0.5m in fresh water).

[0057] Dynamic changes: a sudden rise in water level → submerging communication equipment; rapid currents → destroying sensors.

[0058] Meteorological factors: Rainfall → signal attenuation; humidity → change in air dielectric constant.

[0059] Hydraulic parameters encompass four major categories of physicochemical quantities: meteorological, sediment, water quality, and hydrological parameters. They form the data foundation for water resource management and flood control monitoring. In IoT applications, it is necessary to adaptively select appropriate data transmission methods based on parameter characteristics and hydraulic environmental challenges (rainfall, water quality changes) to ensure reliable data transmission.

[0060] The success rate of water conservancy parameter return indicates the transmission quality of the network link. The value of each water conservancy parameter has a certain impact on the transmission quality under different links. The impact relationship is analyzed by fitting the relationship formula.

[0061] S2035, and so on, respectively, to obtain cellular and satellite relational formulas for each abnormal device.

[0062] S204. Analyze the reporting template of abnormal equipment, and treat the received hydraulic parameters as known parameters and the unreceived hydraulic parameters as unknown parameters.

[0063] S205. Establish P prediction sets for the abnormal device, each prediction set containing all unknown parameters. Assign values ​​to each unknown parameter in each prediction set according to the cellular and satellite relation formulas. Specifically, this includes: S2051. Count the number of known parameters and unknown parameters under the two reporting methods respectively, and calculate the success rate of cellular network reporting method and satellite communication reporting method.

[0064] S2052, All known parameters and success rate G x Substituting these into the cellular relational formula, we obtain the first relational formula. All known parameters and the success rate G. y Substituting into the satellite relation, we obtain the second relation.

[0065] S2053. Assign values ​​to each unknown parameter in each prediction set. The values ​​of all unknown parameters in each prediction set simultaneously satisfy the first relation and the second relation. The values ​​of all unknown parameters in different prediction sets are not completely the same.

[0066] By fitting the relationship between cellular and satellite communications using historical reporting records, multiple prediction sets are generated for unknown parameters, ensuring that the predicted values ​​simultaneously satisfy the regular constraints of both communication methods, thereby improving the scientific rigor and rationality of data repair.

[0067] S300: Calculate the deviation index of each prediction set to establish a reference scheme for each monitoring device. Specifically, this includes: S301. Mark the location of each monitoring device on the GIS map and associate the abnormal device with the monitoring device that is closest to it and is not in an abnormal state.

[0068] S302. Obtain all reporting records for each pair of associated devices, and match reporting records with the same reporting time pairwise. Analyze the values ​​of various unknown parameters in each pair of matched records, and fit the influence relationship of each unknown parameter.

[0069] Each pair of associated devices includes one faulty device and one monitoring device. Each pair of matching records includes two reporting records, which are reported at the same time and belong to the corresponding faulty device and monitoring device, respectively. Specifically, this includes: S3021. Count the number of matching record pairs d under the associated device Q1, obtain the value of unknown parameter C0 in the two reporting records in each pair of matching records, and package them as independent and dependent variables into a sample.

[0070] S3022. Input these d samples into the linear regression model for training and fit the relationship between the unknown parameter C0 and the model: M=θ+vN; where M is the dependent variable, N is the independent variable, θ is the intercept, and v is the regression coefficient.

[0071] S3023, and so on, are respectively the fitting influence formulas of other unknown parameters of abnormal equipment in associated equipment Q1, and the fitting influence formulas of other unknown parameters of abnormal equipment in each pair of associated equipment.

[0072] By associating abnormal devices with nearby normal devices, the influence relationship of unknown parameters (such as a linear regression model) is fitted using spatiotemporally matched reporting records, and the accuracy of the predicted values ​​is optimized by leveraging spatial proximity.

[0073] S303. Analyze the various hydraulic parameters received by the monitoring equipment under the associated equipment, substitute them into the corresponding influence relationship formulas, calculate the predicted values ​​of various unknown parameters of the abnormal equipment, and then calculate the deviation index of each prediction set under the abnormal equipment. Specifically, this includes: Obtain the predicted values ​​of various unknown parameters of the abnormal equipment, as well as the values ​​of various unknown parameters in each prediction set, and substitute them into the formula to calculate the deviation index BI for each prediction set: ; In the formula, α is a constant, b is the number of all unknown parameter terms under abnormal equipment, and YX a PRE is the weighting coefficient for the a-th unknown parameter. a SET is the predicted value of the a-th unknown parameter. a Let be the value of the a-th unknown parameter in the prediction set.

[0074] Quantify the degree of deviation between the assigned values ​​of unknown parameters and the theoretical predictions in each prediction set, and select the prediction set that is closest to the actual situation (with the smallest deviation index).

[0075] S304. For abnormal equipment, the prediction set of the minimum deviation index is selected as the reference set. Based on the values ​​of each known parameter and the values ​​of each unknown parameter in the reference set, a reference scheme is established for the corresponding abnormal equipment.

[0076] S305. The monitoring equipment establishes a reference scheme based on the values ​​of the various water conservancy parameters that have been received.

[0077] The reliability of the prediction set is quantified by the deviation index, and the set with the smallest deviation is selected as the reference scheme to ensure that the data repair results of abnormal equipment are close to the real scenario, providing high-confidence data support for flood control decision-making.

[0078] S400: Analyze the flood control risks of each reference scheme, and thus issue early warnings for the water areas corresponding to each reference scheme.

[0079] Obtain reference solutions for each abnormal device and monitoring device, and analyze the values ​​of various hydraulic parameters in the reference solutions.

[0080] A multi-index weighted comprehensive scoring algorithm was used to analyze the flood control risk level of each reference scheme. The reference schemes were ranked in descending order of flood control risk level, and real-time images of the water areas corresponding to each reference scheme were displayed.

[0081] The system dynamically assesses the flood risk level of each plan and sorts them from highest to lowest risk. It displays real-time images and parameter details of high-risk water areas, helping managers quickly locate weak points in flood control and improve emergency response efficiency.

[0082] Please see Figure 4 The present invention provides a water conservancy data analysis system for flood control monitoring, including a flood control monitoring module, a data analysis module, an operation management module, and a risk early warning module.

[0083] The flood control monitoring module is used to collect GIS maps and report information, as well as the data transmitted back from various monitoring devices in the water area.

[0084] The system collects GIS maps (including water area location, coverage area, and monitoring equipment location), reports information (including time, method, template, success rate, and water parameters), and data transmitted back from monitoring equipment. Reporting records are dynamically updated through a "preliminary generation" and "post-supplementation" mechanism to ensure the number of water parameters matches the template. Data transmission methods include both cellular networks and satellite communication, and abnormal equipment must use both methods simultaneously for data transmission.

[0085] By employing multi-source data acquisition and dynamic data completion mechanisms, the integrity and real-time nature of the data are ensured, providing a reliable foundation for subsequent analysis. The dual-communication design enhances the redundancy of data transmission from malfunctioning devices, reducing the risk of data loss due to a single communication failure.

[0086] The data analysis module is used to identify abnormal devices based on the returned data and to analyze the reported information to build a prediction set for abnormal devices.

[0087] Abnormal devices are identified by a success rate threshold (<100%). Based on cellular and satellite communication reporting records, cellular and satellite training sets are constructed respectively, and a relationship (success rate versus various hydraulic parameters) is fitted. Multiple prediction sets are generated for the unknown parameters of the abnormal devices using this relationship, requiring the predicted values ​​to simultaneously satisfy the relationship constraints of both communication methods.

[0088] By quantifying the impact of hydraulic parameters on success rates using machine learning algorithms, multi-scenario predictions for missing data from faulty equipment are achieved, improving the scientific rigor of data repair. Dual relational constraints ensure that the prediction results conform to actual patterns under different communication methods, enhancing the predictive validity.

[0089] The operation management module is used to calculate the deviation index of each prediction set, thereby establishing a reference scheme for each monitoring device.

[0090] The abnormal equipment is associated with nearby normal equipment, and records with the same reporting time are matched. A linear regression model is used to fit the influence relationship of unknown parameters. The reliability of each prediction set is evaluated by the deviation index formula (combining the weighted difference between the predicted value and the actual associated equipment parameters), and finally the set with the smallest deviation is selected as the reference scheme.

[0091] By leveraging the correlation between spatially adjacent devices, predicted values ​​are optimized, reducing errors from isolated predictions. A deviation index is used to quantify and assess prediction quality, ensuring that the reference scheme closely reflects real-world scenarios and providing high-confidence data support for flood control decision-making.

[0092] The risk warning module is used to analyze the flood control risks of each reference scheme, thereby providing early warnings for the corresponding water areas of each reference scheme.

[0093] A multi-indicator weighted comprehensive scoring algorithm, combined with the water conservancy parameter values ​​in the reference scheme, is used to calculate the flood control risk level of each scheme and rank them according to risk level. Real-time images and water conservancy parameter details of high-risk water areas are displayed.

[0094] Quantitative risk assessment helps to quickly locate flood control weaknesses, enabling managers to intuitively grasp the risk distribution and prioritize high-threat areas, thereby enhancing emergency response capabilities.

[0095] Example 1: Assume the abnormal equipment has an unknown parameter, "precipitation intensity," with a predicted value of 125 mm / h; there are three prediction sets, A1, A2, and A3, with precipitation intensities of 110 mm / h, 150 mm / h, and 130 mm / h, respectively; when the constant α is 3 and the weighting coefficient for precipitation intensity is 0.8, substitute these values ​​into the formula to calculate the deviation index for each prediction set: A1 prediction set: ; A2 prediction set: ; A3 prediction set: ; The abnormal device then selects the A3 prediction set as the reference set.

[0096] Example 2: Latest test data analysis for card number 186xxx: Transmission mechanism: Test data is reported every 5 minutes, with one report sent via the BeiDou link and one report sent via the 4G link simultaneously.

[0097] Reporting status: During the test, a total of 967 4G messages and 1033 BeiDou short messages were received; There were 62 messages received by the 4G link but not by the BeiDou link, and 9 messages received by the BeiDou link but not by the 4G link. The consistency between messages received by the 4G link and messages received by the BeiDou link reached over 96.45%.

[0098] Consistency rate of received messages between 4G link and BeiDou link: ; Reporting rate statistics: During the testing period, the 4G link missed 1 scheduled report, with a reporting rate of 99.3%. During the testing period, the BeiDou link missed 55 scheduled reports, with a reporting rate of 82.37%.

[0099] 4G link reporting rate: ; BeiDou link reporting rate: ; By investigating the serial numbers of 4G link messages, it was found that there were 16 messages with the same serial number at 8 different times, which were suspected to be duplicate messages.

[0100] By examining the serial numbers of BeiDou link messages, it was found that 56 messages with the same serial number were sent repeatedly at 28 different times.

[0101] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0102] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A water conservancy data analysis method for flood control monitoring, characterized in that: The method comprises: S100, collecting GIS maps and reporting information, and backhaul data of each monitoring device in the water area; S200, identifying abnormal devices according to the backhaul data, and analyzing the reporting information to establish a prediction set for the abnormal devices; S300, calculating the deviation index of each prediction set, thereby establishing a reference scheme for each monitoring device; S400, analyzing the flood control risk of each reference scheme, thereby warning the water area corresponding to each reference scheme. 2.The water conservancy data analysis method for flood control monitoring of claim 1, wherein: In S100, the GIS map includes the position and coverage of the water area, and the position of each monitoring device; the reporting information includes the reporting records of each monitoring device, each reporting record including reporting time, reporting mode, reporting template, success rate and water conservancy parameters; the reporting mode includes cellular network and satellite communication; each monitoring device periodically collects water conservancy parameters and backhauls to the monitoring center according to a preset time length; the reporting template includes each item of water conservancy parameters that should be collected; the success rate refers to the ratio of the number of received water conservancy parameters to the number of water conservancy parameters that should be collected at the reporting time; the backhaul data includes each item of water conservancy parameters.

3. The water conservancy data analysis method for flood control monitoring of claim 2, wherein: S200 comprises: S201, obtaining the water conservancy parameters backhauled by each monitoring device at the current time, counting the number of received water conservancy parameters, and calculating the success rate after comparing with the reporting template; setting the state of the monitoring device with a success rate less than 100% as abnormal; S202, the abnormal device backhauls water conservancy parameters using both cellular network and satellite communication, and when the success rate is equal to 100%, the abnormal state of the corresponding abnormal device is canceled; S203, obtaining the reporting records of each abnormal device, taking each item of water conservancy parameters in the reporting record as the independent variable, and taking the success rate as the dependent variable, to fit the cellular relationship formula and the satellite relationship formula of the corresponding abnormal device; S204, analyzing the reporting template of the abnormal device, taking the received water conservancy parameters as known parameters, and taking the uncollected water conservancy parameters as unknown parameters; S205, establishing P prediction sets for the abnormal device, putting all unknown parameters into each prediction set, and assigning values to each unknown parameter in each prediction set according to the cellular relationship formula and the satellite relationship formula.

4. The water conservancy data analysis method for flood control monitoring of claim 3, wherein: S203 comprises: S2031. Classify all reporting records of the abnormal device E1 according to the reporting method; count the number T of all reporting records in the cellular network category x. x And the total number of reporting records T in satellite communication category y. y ; S2032、Analysis of each report record in cellular network class x, with each water conservancy parameter as the independent variable, success rate as the dependent variable and packaged as a sample, and these T x samples are grouped into a cellular training set; S2033. Analyze each reporting record in satellite communication category y, using various water conservancy parameters as independent variables and success rate as the dependent variable, and package them into a sample. y The sample data was used to construct a satellite training set. S2034, substituting the cellular training set and the satellite training set into the formula respectively for training, and fitting the cellular relationship formula and the satellite relationship formula of the abnormal device E1; the formula is as follows: ； where H is the success rate, β0is the intercept term, k is the number of all water parameters, β i is the influence coefficient of the i-th water parameter, Z i is the i-th water parameter; S2035, by analogy, the cellular relationship formula and the satellite relationship formula are fitted for each abnormal device.

5. The water conservancy data analysis method for flood control monitoring of claim 3, wherein: S205 comprises: S2051, respectively statistics two kinds of report mode under the known parameter quantity and unknown parameter, the success rate G of the calculation of cellular network report mode x , and the success rate G of satellite communication report mode y ; S2052, all known parameters and success rate G x Substitute into the cellular relation, get the first relation; all known parameters and success rate G y Substitute into the satellite relation, get the second relation; S2053, assigning values to each unknown parameter in each prediction set, the values of all unknown parameters in each prediction set satisfying the first relationship formula and the second relationship formula at the same time, and the values of all unknown parameters in different prediction sets being not completely the same.

6. The water conservancy data analysis method for flood control monitoring of claim 3, wherein: S300 comprises: S301, marking the position of each monitoring device on the GIS map, and associating the abnormal device with the monitoring device closest to it and in a non-abnormal state; S302, obtaining all reporting records of each pair of associated devices, matching the reporting records with the same reporting time two by two; analyzing the values of each unknown parameter in each pair of matched records, and fitting the influence relationship formula of each unknown parameter; S303, analyze the correlation device under the monitoring device has received each water conservancy parameters, respectively, into the corresponding influence relationship, calculate the abnormal device each unknown parameter prediction value, thereby calculating each prediction set of abnormal device deviation index; S304, the abnormal device selects the minimum deviation index prediction set as the reference set, according to the value of each known parameter, combined with the value of each unknown parameter in the reference set, for the corresponding abnormal device to establish a reference scheme; S305, the monitoring device according to the received each water conservancy parameters value to establish the reference scheme.

7. The water conservancy data analysis method for flood control monitoring of claim 6, wherein: Each pair of correlation equipment includes an abnormal device and a monitoring device; each pair of matching records includes two reporting records, and the reporting time of the two reporting records is the same and belongs to the corresponding abnormal device and monitoring device respectively; S302 includes: S3021, statistics of all matching records of the correlation device Q1 pair d, the value of the unknown parameter C0 in the two reporting records in each pair of matching records is obtained as the independent variable and the dependent variable, and then packed as a sample; S3022, the d samples are input into the linear regression model in turn, and the influence relationship of the unknown parameter C0 is fitted: M=θ+vN; wherein, M is the dependent variable, N is the independent variable, θ is the intercept, and v is the regression coefficient; S3023, by analogy, the influence relationship of each unknown parameter of the abnormal device in the correlation device Q1 is fitted, and the influence relationship of each unknown parameter of the abnormal device in each pair of correlation equipment is fitted. 8.The water conservancy data analysis method for flood control monitoring of claim 6, wherein: S303 includes: obtaining the prediction value of each unknown parameter of the abnormal device, and the value of each unknown parameter in each prediction set, substituting into the formula to calculate the deviation index BI of each prediction set respectively: ； wherein a is a constant, b is the number of unknown parameter terms under abnormal equipment, YX a is the weight coefficient of the a-th unknown parameter, PRE a is the predicted value of the a-th unknown parameter, SET a is the value of the a-th unknown parameter in the predicted set. 9.The water conservancy data analysis method for flood control monitoring of claim 6, wherein: In S400, the reference scheme of each abnormal device and monitoring device is obtained, and the value of each water conservancy parameter in the reference scheme is analyzed; The flood control risk degree of each reference scheme is analyzed by using multi-index weighted comprehensive score algorithm, and each reference scheme is sorted in order from large to small according to the flood control risk degree, thereby warning the water area corresponding to each reference scheme.

10. A water data analysis system for flood monitoring, characterized by: The system includes a flood control monitoring module, a data analysis module, an operation management module and a risk warning module; The flood control monitoring module is used for collecting GIS map and reporting information, and the backhaul data of each monitoring device in the water area; The data analysis module is used for identifying abnormal devices according to the backhaul data, and analyzing the reporting information to establish prediction sets for abnormal devices; The operation management module is used for calculating the deviation index of each prediction set, thereby establishing a reference scheme for each monitoring device; The risk warning module is used for analyzing the flood control risk of each reference scheme, thereby warning the water area corresponding to each reference scheme.

Images