River sand blocking dam flood control alarm method

Abstract

The invention discloses a river sand blocking dam flood control alarm method, and relates to the technical field of flood control alarm, and the method comprises the steps: recognizing a target region with a high flood risk in a low-lying region if the weather condition in the low-lying region is abnormal; the method comprises the following steps: identifying risk points in a plurality of monitoring points in a target area according to flood data, constructing a risk aggregation degree according to related information of the risk points, if the risk aggregation degree exceeds an expectation, taking an area covering all the risk points as a high-risk area, and taking real-time climate condition data as input to predict flood risk in the high-risk area. If the risk is in an increased state, issuing early warning information to the outside; and according to the relationship between the risk degree and the warning threshold value, giving an alarm of a corresponding level to the outside, and giving corresponding emergency response measures for the high-risk area. By enabling the warning threshold to enter a fluctuation state, dynamic early warning is performed on the flood risk, and the destructiveness of the flood risk is reduced.

Description

Technical Field

[0001] The invention relates to the technical field of flood prevention alarms, and in particular to a flood prevention alarm method for a river sediment retaining dam. Background Art

[0002] The flood control and alarm system for the sediment-retention dam uses various sensors installed on the dam and its surrounding areas to monitor key data, including hydrological, meteorological, and dam structural safety data, in real time. If flood levels exceed preset warning levels, cracks appear in the dam, or other safety hazards are detected, the system immediately triggers an alarm.

[0003] A variety of alarm methods are available, including but not limited to audio and video alarms, SMS notifications, email alerts, and automatic transmission of emergency reports to the flood control command center, ensuring that flood control managers can quickly obtain emergency information and respond. The system also automatically records and transmits key information such as the time, location, and specific conditions of the emergency, providing important evidence for subsequent flood control and disaster assessment.

[0004] In the Chinese invention patent application publication number CN116229684A, a risk warning system for waterlogging backflow into subway platforms is disclosed, including patrol cameras, surveillance cameras, scene waterlogging monitoring cameras, meteorological and flood control information collection modules, and data analysis and warning modules; patrol cameras or surveillance cameras are arranged at the entrance of the subway station to monitor the waterlogging conditions at the entrance of the subway station; scene monitoring cameras are arranged in the subway station to monitor the waterlogging conditions in the subway station; the meteorological and flood control information collection module is used to obtain meteorological information and flood control data of the subway station; the data analysis and warning module is connected to the patrol cameras, surveillance cameras, scene waterlogging monitoring cameras, and meteorological and flood control information collection modules to judge the waterlogging conditions based on video data; the working modes of the risk warning system include primary monitoring mode, secondary monitoring mode, warning mode, and alarm mode.

[0005] Combined with the above application and the contents of the prior art:

[0006] When there is continuous rainfall in the upstream basin of the river sediment-blocking dam, the water level may gradually rise. However, due to the limited drainage capacity in the urban construction area, when the accumulated water is difficult to drain, it will gather in the low-lying areas, causing water accumulation and waterlogging upstream of the sediment-blocking dam and even in the adjacent areas, which will have a great impact on the traffic conditions in the low-lying areas.

[0007] When meteorological conditions are abnormal and there is continuous high-frequency and high-intensity rainfall, waterlogging is more likely to occur in low-lying areas. Especially when there is a high flow of people and traffic in low-lying areas, the harm caused by flood disasters is greater.

[0008] When issuing alarms for flood control at sediment-blocking dams, existing alarm methods are usually based on relevant data such as the river's water level, flow velocity and flow rate. By constructing a prediction model to output the prediction results, an alarm is then issued to the outside world based on the prediction results. However, when there are certain anomalies in climatic conditions, especially when climatic parameters such as rainfall intensity, groundwater level and air humidity continue to produce anomalies, or even when the degree of anomalies continues to increase, the risk of waterlogging and flooding in low-lying areas within the dam basin is greater. In this case, if the alarm is still issued only based on changes in various data, it may be difficult to take early action to reduce the losses caused by waterlogging and flooding.

[0009] To this end, the present invention provides a flood prevention alarm method for a river sediment dam. Summary of the Invention

[0010] (1) Technical problems solved

[0011] In response to the shortcomings of the existing technology, the present invention provides a flood prevention and alarm method for river sediment dams. The method identifies target areas with high flood risk within low-lying areas; identifies risk points at several monitoring points within the target area based on flood data; and constructs a risk concentration based on the relevant information of the risk points. If the risk concentration exceeds expectations, the area covering all risk points is designated as a high-risk area. Real-time climate condition data is used as input to predict the flood risk within the high-risk area. If the risk is increasing, an early warning message is issued to the outside world. Based on the relationship between the risk level and the warning threshold, an alarm of the corresponding level is issued to the outside world, and corresponding emergency response measures are given to the high-risk area. By causing the warning threshold to enter a fluctuating state, a dynamic early warning of flood risk is provided; thereby solving the technical problems raised in the background technology.

[0012] (2) Technical solution

[0013] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0014] A flood control alarm method for a river sediment dam includes selecting a low-lying area within a river basin covered by the flood control sediment dam, and if abnormal weather conditions exist in the low-lying area, identifying a target area with a high flood risk within the low-lying area based on environmental condition data;

[0015] Based on flood data, risk points are identified at several monitoring points within the target area. Risk concentration is constructed based on the relevant information of the risk points. If the risk concentration exceeds expectations, a risk prediction instruction is issued to the outside world.

[0016] The area covering all risk points is designated as a high-risk area. Real-time climate data is used as input to predict flood risk in the high-risk area. If the risk is increasing, an early warning message is issued to the outside world.

[0017] Based on the relationship between the risk level and the warning threshold, an alarm of corresponding level is issued to the outside, and corresponding emergency response measures are given for high-risk areas based on different alarm levels.

[0018] Furthermore, after executing emergency response measures, the effectiveness is constructed after collecting emergency feedback data. If the effectiveness is lower than expected, an optimization plan is given by the emergency measures optimization knowledge graph based on the emergency feedback data.

[0019] Furthermore, the meteorological condition data in the low-lying area and its adjacent areas are monitored and collected, and the meteorological condition data set is generated after aggregation; the anomaly degree is generated from the meteorological condition data set, and if the anomaly degree exceeds the anomaly threshold, a data collection instruction is issued to the outside.

[0020] Further, after receiving the data collection instruction, environmental condition data is collected in the low-lying area to generate an environmental condition data set of the low-lying area;

[0021] If a low-lying area is currently experiencing continuous rainfall, the environmental condition data of the low-lying area is used as input, and the trained low-lying risk identification model is used to identify the flood risk in the low-lying area. If the flood risk exceeds expectations, the corresponding low-lying area will be used as the target area.

[0022] Furthermore, several evenly distributed monitoring points are set up in the target area, and real-time flood data are collected at the monitoring points. After obtaining the real-time data, the corresponding historical data are collected and aggregated to generate a flood data set;

[0023] Taking real-time flood data as input, the trained risk area identification model is used to predict the probability of flood risk in the monitoring point and its affected area, and the part where the probability of flood risk exceeds the expected level is regarded as the risk point.

[0024] Furthermore, after obtaining the location information of the risk point, the risk point is marked on the electronic map, and the risk concentration is calculated based on the location information of the risk point and the probability of flood risk. If the risk concentration exceeds the risk threshold, a risk prediction instruction is issued to the outside.

[0025] Furthermore, after receiving the risk prediction instruction, the prediction period is constrained according to the risk concentration Ftx, and the prediction period that meets the constraint conditions is used as the target prediction period, which includes a number of prediction nodes;

[0026] Taking the real-time climate condition data in the high-risk area as input, the trained flood risk identification model is used for prediction and evaluation. At each prediction node within the target prediction period, the flood risk value of the risk point is evaluated and obtained.

[0027] Furthermore, the risk level is constructed after obtaining the flood risk value of each risk point in the high-risk area;

[0028] After obtaining several risk levels continuously, if the current risk level increases compared to the previous value, early warning information will be released to the outside through multiple channels.

[0029] Furthermore, after continuously obtaining the risk levels of several prediction nodes, a warning threshold is constructed based on the risk levels;

[0030] If the risk level exceeds the warning threshold, a Level 1 warning will be issued to the outside world, and traffic arteries will be closed and warning signs will be set up to provide detour routes. An evacuation plan for vehicles and personnel will be developed, and evacuation routes and assembly points will be marked.

[0031] If the risk level is within the warning threshold, a Level 2 warning will be issued to the outside world, flood control equipment will be inspected and maintained, and flood control materials will be prepared in advance at locations close to the risk point;

[0032] If the risk level is lower than the warning threshold, a level 3 warning will be issued to the outside world, drainage ditches and pipes in low-lying areas will be inspected and cleaned, and mobile pumping equipment will be deployed at corresponding locations to carry out emergency drainage in waterlogged areas.

[0033] Furthermore, each real-time flood and traffic data, response measures and corresponding feedback data are recorded at each feedback node and aggregated to generate an emergency feedback data set;

[0034] The emergency feedback data in the emergency feedback data set is used as input, and the trained effectiveness evaluation model is used to output the effectiveness. If the effectiveness is lower than the preset effectiveness threshold, an optimization instruction is issued to the outside.

[0035] Furthermore, after receiving the optimization instruction and obtaining the feedback data of the emergency response measures, the corresponding feedback features are extracted through feature engineering;

[0036] When there is a long period of rainfall in a low-lying area, the optimization of emergency response measures for flood risks in the low-lying area is used as the target word, and a knowledge graph for emergency measures optimization is pre-constructed.

[0037] A flood control alarm system for a river sediment dam includes a region identification unit that selects low-lying areas within a river basin covered by the flood control sediment dam and, if abnormal weather conditions exist in the low-lying areas, identifies target areas with high flood risk within the low-lying areas based on environmental condition data;

[0038] The risk detection unit identifies risk points at several monitoring points within the target area based on flood data, and constructs risk concentration based on the relevant information of the risk points. If the risk concentration exceeds expectations, a risk prediction instruction is issued to the outside world;

[0039] The prediction unit considers the area covering all risk points as a high-risk area, uses real-time climate condition data as input, and predicts the flood risk in the high-risk area. If the risk is increasing, it will issue an early warning information to the outside world;

[0040] The alarm unit, based on the relationship between risk and warning threshold, issues external alarms of corresponding levels and provides corresponding emergency response measures for high-risk areas according to different alarm levels;

[0041] After executing emergency response measures, the optimization unit collects emergency feedback data and constructs effectiveness. If the effectiveness is lower than expected, an optimization plan is given based on the emergency feedback data and the emergency measure optimization knowledge graph.

[0042] (3) Beneficial effects

[0043] The present invention provides a flood prevention alarm method for a river sediment dam, which has the following beneficial effects:

[0044] 1. Determine flood risk based on various data in low-lying areas. If the identified flood risk exceeds expectations, it will be used as a target area for early treatment to reduce the risk of waterlogging or flooding in the low-lying area. After flood risk identification and prediction, several risk points can be screened out in the low-lying area, and pre-treatment can be carried out on the risk points before flooding or waterlogging occurs.

[0045] 2. Determine the risk within the target area. If there is continuous rainfall or abnormal weather conditions in the target area, and the risk of flooding is high, it is necessary to predict the corresponding risk level and scale, and take appropriate treatment measures based on the obtained prediction results.

[0046] 3. Based on the flood risk value, we can determine what emergency measures to take, and by constructing the risk level based on the flood risk value at each risk point, we can make an overall evaluation of the high-risk areas in the low-lying areas based on the risk level. At the same time, based on the changing trend of the risk level, we can issue warning information to the outside world to ensure the reliability of the warning.

[0047] 4. Determine the alarm level based on the relationship between the risk level and the warning threshold. When there is a flood risk in a low-lying area, implement a multi-level in-depth early warning. By making the warning threshold enter a fluctuating state, a dynamic early warning of the flood risk can be achieved. When a sudden abnormal flood risk occurs, the destructiveness of the flood risk can be reduced.

[0048] 5. Provide corresponding emergency response measures based on different alarm levels. When different degrees of waterlogging and flooding risks occur in high-risk areas, take targeted emergency response measures to minimize corresponding safety hazards and reduce losses. At the same time, it can also avoid excessive and disorderly preparations and improve efficiency when emergency treatment is needed.

[0049] 6. Evaluate the effectiveness of emergency response measures. When the effectiveness is insufficient, optimize or replace the current effective emergency response strategy to ensure the handling of waterlogging and flooding risks. Provide an optimization plan based on the emergency measures optimization knowledge graph to achieve targeted optimization of emergency response measures. When flood hazards occur again in low-lying areas, they can be better handled, achieving in-depth early warning of flood risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 This is a flow chart of the flood prevention alarm method for a river sediment dam according to the present invention;

[0051] Figure 2 This is a structural diagram of the river sediment dam flood prevention alarm system of the present invention. DETAILED DESCRIPTION

[0052] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0053] See also Figure 1 The present invention provides a flood prevention alarm method for a river sediment dam, comprising:

[0054] Step 1: Select low-lying areas within the catchment area of ​​the flood control and sediment retention dam. If there are abnormal weather conditions in the low-lying areas, identify target areas with high flood risk within the low-lying areas based on environmental condition data.

[0055] The step 1 includes the following:

[0056] Step 101: Screen low-lying areas within the watershed covered by the flood control and sediment retention dam, monitor and collect meteorological condition data in the low-lying areas and their adjacent areas, including accumulated rainfall, rainfall intensity, groundwater level, and air humidity, and aggregate the data to generate a meteorological condition data set;

[0057] After completing the linear normalization processing on each data, the abnormality degree Tto describing the abnormality degree of meteorological conditions is generated from the meteorological condition data set in the following way:

[0058]

[0059] Weight coefficient, 0≤F1≤1, 0≤F2≤1, and F1+F2=1; the weight coefficient can be obtained by referring to the hierarchical analysis method; T is the length of the monitoring period, Rr is the rainfall intensity, Xr is the groundwater level, and Sr is the groundwater level;

[0060] Pre-set anomaly thresholds based on historical data and management expectations of the degree of climate anomalies;

[0061] If the anomaly degree Tto exceeds the anomaly threshold, it means that the climate state in the low-lying area is abnormal and the risk of flooding is high. At this time, a data collection instruction is issued to the outside world;

[0062] When in use, by real-time monitoring of meteorological conditions in low-lying areas, flood warning can be achieved when anomalies occur;

[0063] Step 102: After receiving the data collection instruction, environmental condition data is collected in the low-lying area, including the drainage volume of the drainage system in the low-lying area, historical flood data, including flood depth and duration, length of roads along the river, number of bridges, average daily traffic flow and pedestrian density, etc., and the environmental condition data set of the low-lying area is generated;

[0064] Step 103: If the low-lying area is currently experiencing continuous rainfall, a convolutional neural network is trained using the labeled sample data to obtain a low-lying risk identification model for assessing the risk of low-lying areas within the low-lying area.

[0065] The environmental condition data of low-lying areas is used as input, and the trained low-lying risk identification model is used to identify the flood risk in the low-lying areas. If the flood risk exceeds the expected level, the corresponding low-lying areas are designated as target areas.

[0066] When using, combine the contents in steps 101 to 103:

[0067] After determining the low-lying areas related to the dam, if the meteorological conditions in the low-lying areas are abnormal, the flood risk can be judged based on the various data in the low-lying areas. If the identified flood risk exceeds expectations, it will be used as a target area, and advance processing can be carried out on the target area to reduce the risk of water accumulation or flooding in the low-lying areas.

[0068] When meteorological conditions are abnormal and there is continuous high-frequency and high-intensity rainfall, waterlogging is more likely to occur in low-lying areas. Especially when there is a high flow of people and traffic in low-lying areas, the harm caused by flood disasters is greater.

[0069] When issuing alarms for flood control at sediment-blocking dams, existing alarm methods are usually based on relevant data such as the river's water level, flow velocity and flow rate. By constructing a prediction model to output the prediction results, an alarm is then issued to the outside world based on the prediction results. However, when there are certain anomalies in climatic conditions, especially when climatic parameters such as rainfall intensity, groundwater level and air humidity continue to produce anomalies, or even when the degree of anomalies continues to increase, the risk of waterlogging and flooding in low-lying areas within the dam basin is greater. In this case, if the alarm is still issued only based on changes in various data, it may be difficult to take early action to reduce the losses caused by waterlogging and flooding.

[0070] Step 2: Based on flood data, risk points are identified at several monitoring points within the target area. Based on the relevant information of the risk points, a risk concentration degree is constructed. If the risk concentration degree exceeds the expectation, a risk prediction instruction is issued to the outside world.

[0071] The second step includes the following:

[0072] Step 201: Set up a number of evenly distributed monitoring points in the target area, collect real-time flood data at the monitoring points, and collect corresponding historical data after obtaining the real-time data, and aggregate them to generate a flood data set;

[0073] For example, remote automatic weather stations, including rain gauges, anemometers, and thermometers, are installed at key locations to monitor real-time rainfall, rainfall intensity, wind speed, and temperature. Water level gauges and flow meters are deployed at different sections of the river. Soil moisture sensors are installed to monitor soil moisture content. Groundwater level monitoring wells are deployed to monitor groundwater dynamics. Geological sensors are deployed in areas where landslides may occur. Settlement monitoring devices are installed around dams and flood control facilities.

[0074] Step 202: Train a convolutional neural network using the labeled sample data to obtain a trained flood risk prediction model. Using the real-time flood data as input, the trained risk area identification model is used to predict the probability of flood risk occurring at the monitoring point and its affected area. The portion with a flood risk probability exceeding the expected value is identified as a risk point.

[0075] When in use, a target area is selected from several low-lying areas. After flood risk identification and prediction, several risk points can be screened out in the low-lying areas. The risk points can be pre-processed before flooding and waterlogging occur, for example, emergency measures can be prepared in advance, etc., to reduce corresponding safety hazards.

[0076] Step 203: Construct an electronic map covering the target area. After obtaining the location information of the risk points, mark the risk points on the electronic map. Calculate the risk concentration Ftx based on the location information of the risk points and the probability of flood risk as follows:

[0077]

[0078] Among them, fx i y i ) is the grid point (x i y i )’s risk concentration, A i is the grid point (x i y i ), m is the density of the grid in the low-lying area, n is the total number of risk points, σ ​​is the bandwidth parameter, which controls the smoothness of the kernel function; p j Risk point F j Flood risk probability; risk point F j and grid points (x i y i ) between the two;

[0079] Pre-set risk thresholds based on historical data and expectations for flood risk management within the target area;

[0080] If the risk concentration Ftx exceeds the risk threshold, it means that the risk of flooding in the current area is high and needs to be dealt with in a timely manner, which may cause great economic risks. At this time, a risk prediction instruction is issued to the outside world;

[0081] When using, combine the contents of steps 201 to 203:

[0082] When there are multiple risk points, the risk concentration Ftx is obtained by calculating the location of the risk points and the probability of flood risk. Based on the risk concentration Ftx, the risk in the target area can be judged. When there is continuous rainfall or abnormal weather conditions in the target area, if the risk of flooding is large, it is necessary to predict the corresponding risk level and scale, and take corresponding treatment measures based on the obtained prediction results.

[0083] Step 3: The area covering all risk points is designated as a high-risk area. Using real-time climate data as input, the flood risk in the high-risk area is predicted. If the risk is increasing, an early warning is issued to the outside world.

[0084] The step three includes the following:

[0085] Step 301: After receiving the risk prediction instruction, the area covering all risk points is regarded as a high-risk area. The prediction period is constrained according to the risk concentration Ftx, and the prediction period that meets the constraint conditions is regarded as the target prediction period. The target prediction period includes a number of prediction nodes. The constraint method is as follows:

[0086]

[0087] Weight coefficient, 0≤α≤1, 0≤β≤1; n is the number of prediction nodes, C ij is the time interval from the i-th prediction node to the j-th prediction node, C a It is the average value of the time interval. When making equal interval predictions, the number of predictions is limited to constrain the length of the prediction cycle.

[0088] When in use, when it is necessary to predict the risk of waterlogging and flooding in high-risk areas, the risk concentration degree is used to constrain the prediction period length, prediction frequency and prediction nodes, so that the risk and prediction time remain positively correlated. When the risk is high, there can be sufficient processing time to reduce the corresponding hidden dangers.

[0089] Step 302: training a convolutional neural network using the labeled sample data to obtain a trained flood risk identification model;

[0090] Using real-time climate data such as rainfall, river water levels, and soil moisture in high-risk areas as input, the trained flood risk identification model is used for prediction and assessment. At each prediction node within the target prediction period, the flood risk value at the risk point is evaluated and obtained.

[0091] Step 303: Obtain the flood risk value of each risk point in the high-risk area and construct the risk degree Fxo in the following manner:

[0092]

[0093] Among them, the weight coefficient w(xy) is proportional to the inverse of the distance from the risk point to the fixed point (x0y0), and fxy) is the flood risk value at the risk point (xy);

[0094] After obtaining several consecutive risk levels Fxo, if the current risk level Fxo increases compared to the previous value, early warning information is released to the outside world through multiple channels, such as SMS or APP message push, including flood status data in high-risk areas, the scope of areas with flood risk, road traffic status data, etc.

[0095] When using, combine the contents in steps 301 to 303:

[0096] After limiting the prediction period, the output flood risk value can be used to evaluate the risk level of waterlogging and flooding at each risk point. The emergency response measures can be determined based on the flood risk value. The risk degree Fxo can be constructed based on the flood risk value at each risk point. Based on the risk degree Fxo, an overall evaluation of high-risk areas in low-lying areas can be carried out. At the same time, prompt information can be issued to the outside based on the changing trend of the risk degree to ensure the reliability of the prompt.

[0097] Step 4: Based on the relationship between risk and warning threshold, issue an external alarm of corresponding level, and provide corresponding emergency response measures for high-risk areas according to different alarm levels;

[0098] The step 4 includes the following contents:

[0099] Step 401: After continuously obtaining the risk Fxo of several prediction nodes, construct the warning threshold [Ua, Ub] based on the risk Fxo in the following manner:

[0100]

[0101] Where i = 1, 2, ..., k, k is the number of prediction nodes, Fxo i is the risk degree on the i-th prediction node, Fxo a is the mean value of risk;

[0102] Step 402: Determine the alarm level based on the relationship between the risk level and the warning threshold, and provide corresponding emergency response measures based on different alarm levels; the details are as follows:

[0103] If the risk Fxo i If the level exceeds the warning threshold [Ua, Ub], a Level 1 warning will be issued to the outside world. After closing the main traffic arteries, warning signs will be set up to guide detour routes. An evacuation plan for vehicles and personnel will be formulated, and evacuation routes and assembly points will be marked.

[0104] If the risk Fxo i Within the warning threshold [Ua, Ub], a Level 2 warning is issued to the outside world, flood control equipment is inspected and maintained, and flood control materials are prepared in advance near the risk point, such as sandbags, flood barriers, pumps, and tarpaulins;

[0105] If the risk Fxo i If the water level falls below the warning threshold [Ua, Ub], a Level 3 warning will be issued to the outside world, drainage ditches and pipes in low-lying areas will be inspected and cleaned, and mobile pumping equipment will be deployed at corresponding locations to carry out emergency drainage in waterlogged areas;

[0106] It should be noted that after an early warning is issued, one or a combination of the emergency response measures of the current level and the higher level may be implemented. For example, after a level 1 early warning is issued, in addition to the emergency response measures corresponding to the level 1 early warning, the response measures of the level 2 and level 3 early warnings may also be implemented simultaneously or in combination.

[0107] When using, combine the contents in steps 401 and 402:

[0108] By constructing warning thresholds after continuously obtaining several risk levels, and determining the alarm level based on the relationship between the risk level and the warning threshold, a multi-level in-depth early warning can be achieved when flood risks exist in low-lying areas. By making the warning threshold enter a fluctuating state, a dynamic early warning of flood risks can be achieved. When sudden abnormalities in flood risks occur, they are more likely to be noticed, which can reduce the destructiveness of flood risks.

[0109] Corresponding emergency response measures are given according to different alarm levels. When different degrees of waterlogging and flooding risks occur in high-risk areas, targeted emergency response measures are taken to minimize corresponding safety hazards and reduce losses. At the same time, it can also avoid excessive and disorderly preparations and improve efficiency when emergency treatment is needed.

[0110] Step 5: After executing the emergency response measures, collect emergency feedback data and build an effectiveness evaluation system to evaluate the effectiveness of the emergency response. If the effectiveness is lower than expected, optimize the emergency response measures based on the emergency feedback data and the emergency measures optimization knowledge graph.

[0111] The step five includes the following:

[0112] Step 501: pre-set a feedback cycle including several feedback nodes, record each real-time flood and traffic data, response measures and corresponding feedback data at each feedback node, and aggregate them to generate an emergency feedback data set;

[0113] The labeled sample data is used to train a machine learning algorithm to obtain a trained effectiveness evaluation model for flood control emergency response measures. The emergency feedback data in the emergency feedback data set is used as input, and the trained effectiveness evaluation model is used to output the effectiveness. The effectiveness of the early warning system and emergency response measures is evaluated based on the effectiveness.

[0114] Pre-set effectiveness thresholds based on historical data and management expectations for emergency response measures;

[0115] If the effectiveness is lower than the preset effectiveness threshold, it means that the current emergency response measures have failed to achieve the desired effect and need to be optimized or replaced. At this time, an optimization instruction is issued to the outside world;

[0116] When in use, after flooding occurs in low-lying areas and is handled, effectiveness is established based on the collected feedback data. The effectiveness of the emergency treatment measures implemented is evaluated based on the effectiveness. When the effectiveness is insufficient, the current effective emergency response strategy can be optimized or replaced to ensure the handling of waterlogging and flooding risks.

[0117] Step 502: After receiving the optimization instruction, after obtaining feedback data on the execution of emergency response measures, extract corresponding feedback features through feature engineering;

[0118] When a low-lying area experiences prolonged rainfall, the optimization of emergency response measures for flood risk in the low-lying area is used as the target word. After deep retrieval and entity relationship building, a knowledge graph for emergency measure optimization is pre-built. Based on the correspondence between the emergency measure optimization scheme and the feedback features, an optimization scheme is given by the emergency measure optimization knowledge graph, and the optimization scheme is executed to optimize the emergency response measures.

[0119] When using, combine the contents in steps 501 and 502:

[0120] When emergency response needs to be optimized, the emergency measures optimization knowledge graph provides an optimization plan based on feedback data, which can achieve targeted optimization of emergency response measures. When flood hazards occur again in low-lying areas, they can be better handled, achieving in-depth early warning of flood risks.

[0121] The Analytic Hierarchy Process (AHP) is a decision-making method that breaks down decision-related elements into a hierarchy of objectives, criteria, and options, and then conducts qualitative and quantitative analysis based on this hierarchy. It is particularly well-suited for target systems with hierarchical and interleaved evaluation indicators, and is an effective decision-making tool when target values ​​are difficult to quantify.

[0122] The core of the AHP lies in breaking down the decision problem into multiple levels, forming a hierarchical structure. This structure typically includes the objective level, the criteria level, the sub-criteria level, and the solution level. By solving the eigenvectors of the judgment matrix, the priority weight of each element at each level relative to the element at the previous level is determined. Finally, a weighted sum method is used to recursively combine the final weights of each alternative solution relative to the overall goal, thereby identifying the optimal solution.

[0123] The method for constructing a knowledge graph for optimizing emergency response measures for flood risk in low-lying areas is as follows:

[0124] Determine the goals and scope of the knowledge graph: Clarify the goals: The knowledge graph to be constructed should focus on flood risks in low-lying areas, including risk identification, assessment, early warning, emergency response and post-disaster recovery.

[0125] Define the scope: Determine the geographical area covered by the knowledge graph, the time span, the departments and institutions involved, the specific flood types and emergency measures, etc.

[0126] Data Collection and Preprocessing: Data Collection: Literature: Collect books, journals, reports, and other materials related to flood disasters to gain theoretical knowledge. Online Resources: Utilize search engines, industry knowledge bases, and online encyclopedias to obtain the latest data and case studies. Field Research: Conduct on-site inspections of low-lying areas to gather firsthand information. Obtain relevant documents issued by government departments regarding flood disaster warnings, emergency response, and recovery efforts.

[0127] Data preprocessing: Cleaning: Remove duplicate, erroneous, or irrelevant data. Deduplication: Combine identical or similar data items. Classification and summarization: Organize and summarize data by different categories, such as risk type, emergency response measures, and departmental responsibilities.

[0128] Knowledge Extraction and Representation: Entity Extraction: Automatically identify entities related to flood risk emergency response from text data, such as place names, organization names, and measure names. Relationship Extraction: Identify relationships between entities, such as the "prone to" relationship in "a certain area is prone to flood disasters" and the "responsible for" relationship in "a certain department is responsible for a certain emergency measure." Knowledge Representation: Use graphical methods (such as RDF and Property Graph) to represent entities and their relationships, building a knowledge graph framework.

[0129] Knowledge fusion and verification: Knowledge fusion: Integrate knowledge from different sources, structures, and formats to form a unified knowledge system. Knowledge verification: Verify the accuracy and completeness of knowledge through expert review, data comparison, and other methods.

[0130] Knowledge Graph Construction and Optimization: Building a knowledge graph: Based on the data collected and processed in the above steps, use a graph database or specialized knowledge graph construction tools (such as gStore) to build a knowledge graph. Optimization and Iteration: Based on actual application scenarios, continuously optimize and iterate the knowledge graph, adding new knowledge points and relationships and correcting incorrect information.

[0131] See also Figure 2 The present invention provides a river sediment dam flood prevention alarm system, comprising:

[0132] The regional identification unit selects low-lying areas within the watershed covered by the flood control and sediment retention dam. If there are abnormal weather conditions in the low-lying areas, the target areas with high flood risk are identified within the low-lying areas based on environmental condition data.

[0133] The risk detection unit identifies risk points at several monitoring points within the target area based on flood data, and constructs risk concentration based on the relevant information of the risk points. If the risk concentration exceeds expectations, a risk prediction instruction is issued to the outside world;

[0134] The prediction unit considers the area covering all risk points as a high-risk area, uses real-time climate condition data as input, and predicts the flood risk in the high-risk area. If the risk is increasing, it will issue an early warning information to the outside world;

[0135] The alarm unit, based on the relationship between risk and warning threshold, issues external alarms of corresponding levels and provides corresponding emergency response measures for high-risk areas according to different alarm levels;

[0136] After executing emergency response measures, the optimization unit collects emergency feedback data and constructs effectiveness. If the effectiveness is lower than expected, an optimization plan is given based on the emergency feedback data and the emergency measure optimization knowledge graph.

[0137] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0138] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0139] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0140] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is only for some logical functions. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0141] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0142] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0143] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0144] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A flood prevention alarm method for a river sediment dam, characterized by: include, Select low-lying areas within the catchment area of ​​the flood control and sediment retention dam. If there are abnormal weather conditions in the low-lying areas, identify target areas with high flood risk within the low-lying areas based on environmental condition data. Based on flood data, risk points are identified at several monitoring points within the target area. Risk concentration is constructed based on the relevant information of the risk points. If the risk concentration exceeds expectations, a risk prediction instruction is issued to the outside world. The area covering all risk points is designated as a high-risk area. Real-time climate data is used as input to predict flood risk in the high-risk area. If the risk is increasing, an early warning message is issued to the outside world. Based on the relationship between the risk level and the warning threshold, an alarm of corresponding level is issued to the outside, and corresponding emergency response measures are given for high-risk areas based on different alarm levels.

2. A flood prevention alarm method for a river sediment dam according to claim 1, characterized in that: After executing emergency response measures, the effectiveness is constructed after collecting emergency feedback data. If the effectiveness is lower than expected, an optimization plan is given based on the emergency feedback data and the emergency measures optimization knowledge graph.

3. A flood prevention alarm method for a river sediment dam according to claim 2, characterized in that: Monitor and collect meteorological condition data in low-lying areas and their adjacent areas, and generate a meteorological condition data set after aggregation; generate anomaly degree from the meteorological condition data set, and if the anomaly degree exceeds the anomaly threshold, issue a data collection instruction to the outside.

4. A flood prevention alarm method for a river sediment dam according to claim 3, characterized in that: After receiving the data collection instruction, environmental condition data is collected in the low-lying area to generate an environmental condition data set for the low-lying area; If a low-lying area is currently experiencing continuous rainfall, the environmental condition data of the low-lying area is used as input, and the trained low-lying risk identification model is used to identify the flood risk in the low-lying area. If the flood risk exceeds expectations, the corresponding low-lying area will be used as the target area.

5. A flood prevention alarm method for river sediment dams according to claim 4, characterized in that: Several evenly distributed monitoring points are set up in the target area to collect real-time flood data at the monitoring points. After obtaining the real-time data, the corresponding historical data is collected and aggregated to generate a flood data set. Taking real-time flood data as input, the trained risk area identification model is used to predict the probability of flood risk in the monitoring point and its affected area, and the part where the probability of flood risk exceeds the expected level is regarded as the risk point.

6. A flood prevention alarm method for river sediment dams according to claim 5, characterized in that: After obtaining the location information of the risk points, mark the risk points on the electronic map, and calculate the risk concentration based on the location information of the risk points and the probability of flood risk; If the risk concentration exceeds the risk threshold, a risk prediction instruction will be issued to the outside.

7. A flood prevention alarm method for a river sediment dam according to claim 6, characterized in that: After receiving the risk prediction instruction, the prediction period is constrained according to the risk concentration, and the prediction period that meets the constraint conditions is used as the target prediction period, which includes several prediction nodes; Using real-time climate data in high-risk areas as input, the trained flood risk identification model is used for prediction and assessment. At each prediction node within the target prediction period, the flood risk value at the risk point is evaluated and obtained. Obtain the flood risk value of each risk point in the high-risk area and construct the risk degree; After obtaining several risk levels continuously, if the current risk level increases compared to the previous value, early warning information will be released to the outside through multiple channels.

8. A flood prevention alarm method for river sediment dams according to claim 7, characterized in that: After continuously obtaining the risk levels of several prediction nodes, the warning threshold is constructed based on the risk levels; If the risk level exceeds the warning threshold, a Level 1 warning will be issued to the outside world, and traffic arteries will be closed and warning signs will be set up to provide detour routes. An evacuation plan for vehicles and personnel will be developed, and evacuation routes and assembly points will be marked. If the risk level is within the warning threshold, a Level 2 warning will be issued to the outside world, flood control equipment will be inspected and maintained, and flood control materials will be prepared in advance at locations close to the risk point; If the risk level is lower than the warning threshold, a level 3 warning will be issued to the outside world, drainage ditches and pipes in low-lying areas will be inspected and cleaned, and mobile pumping equipment will be deployed at corresponding locations to carry out emergency drainage in waterlogged areas.

9. A flood prevention alarm method for river sediment dams according to claim 8, characterized in that: Record each real-time flood and traffic data, response measures and corresponding feedback data at each feedback node, and aggregate them to generate an emergency feedback data set; The emergency feedback data in the emergency feedback data set is used as input, and the trained effectiveness evaluation model is used to output the effectiveness. If the effectiveness is lower than the preset effectiveness threshold, an optimization instruction is issued to the outside.

10. A flood prevention alarm method for a river sediment dam according to claim 9, characterized in that: After receiving the optimization instruction and obtaining the feedback data of the execution of emergency response measures, the corresponding feedback features are extracted through feature engineering; when there is a long period of rainfall in the low-lying area, the optimization of emergency response measures for flood risks in the low-lying area is used as the target word, and an emergency measure optimization knowledge graph is pre-constructed.

Images