Water level sequence data monitoring and early warning method based on big data

Abstract

The invention discloses a water level sequence data monitoring and early warning method based on big data, and relates to the technical field of water level monitoring and early warning. An existing method still has some defects, and mainly comprises the problems of data quality and anomaly detection, disaster hidden danger early warning lag, insufficient response speed and decision support capability of an early warning system and the like. Comprising the following steps: 1, acquiring hydrological, geographical and meteorological data of a specific area through a big data platform; 2, performing data screening on the hydrological data, the geographic data and the meteorological data; according to the invention, through feature screening, statistical analysis and anomaly detection of the system, the abnormal data are effectively identified and processed, and the data accuracy and reliability of the early warning system are ensured; by calculating abnormal coefficients of hydrology, geography and meteorology, high-risk time periods and regions are identified in time, early disaster early warning is realized, and the problem of early warning lag is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of water level monitoring and early warning, and in particular to a water level sequence data monitoring and early warning method based on big data. Background Art

[0002] Existing water level series data monitoring and early warning methods based on big data use large-scale data collection and analysis technology to provide timely and accurate flood warnings and risk assessments by deeply mining and predicting trends in real-time data from water level monitoring points. This method usually combines meteorological data, historical water level data, and environmental parameters, and uses machine learning and big data analysis models to model water level changes and predict future water level trends;

[0003] However, this method still has some defects, mainly including being limited by data missing, sensor failure or untimely real-time data processing, failing to detect potential disaster risks in advance when water levels change rapidly, precipitation exceeds the standard or extreme weather, the existing system may lack sufficient decision support or respond quickly enough during emergency response and flood warning, and lack the ability to deal with other natural disasters caused by water levels. Summary of the invention

[0004] In order to solve the technical problems raised by the above background technology, the present invention is proposed. An embodiment of the present invention provides a water level series data monitoring and early warning method based on big data.

[0005] The purpose of the present invention can be achieved through the following technical solution: A water level series data monitoring and early warning method based on big data, comprising:

[0006] Step 1: Obtain hydrological, geographical and meteorological data of a specific area through a big data platform;

[0007] Step 2: Screen the hydrological, geographic and meteorological data, specifically: visualize the hydrological, geographic and meteorological data, display the trend of water level and flow over time through line graphs, and mark the peak points and sharp change sections; display the spatial distribution of precipitation and meteorological data through heat maps, and check whether the heavy rainfall area overlaps with the low-lying terrain; at the same time, calculate basic statistics and identify extreme values ​​and key characteristic points of sharp fluctuations;

[0008] Step 3: Obtain the corresponding time points of the data with obvious feature markers through big data, and obtain all the data at the corresponding time points; if floods or mountain torrents have occurred at the corresponding time points, obtain the comparison data set 1 by statistics; if floods or mountain torrents do not exist at the corresponding time points, obtain the comparison data set 2 by statistics; at each time point, filter out the data of the disaster that has occurred at that time point from the comparison data set 1, and calculate the mean of all the comparison data sets 1 to obtain the mean set, compare the comparison data set 1 with the comparison data set 2 to obtain the difference data set between the two; then compare the difference data set with the mean corresponding to the data, and obtain the difference over-boundary value set for the difference data set that is greater than or equal to the mean corresponding to the data; and obtain the difference critical value set for the difference data set that is less than the mean corresponding to the data;

[0009] Step 4: retrieve the difference data set, difference over-boundary value set and difference critical value set obtained in step 3; then classify the three to obtain the difference data set, difference over-boundary value set and difference critical value corresponding to hydrology, geography and meteorology; analyze the hydrological difference data set, hydrological difference over-boundary value set and hydrological difference critical value set to obtain the hydrological anomaly coefficient value; analyze the geographical difference data set, geographical difference over-boundary value set and geographical difference critical value set to obtain the geographical anomaly coefficient value; analyze the meteorological difference data set, meteorological difference over-boundary value set and meteorological difference critical value set to obtain the meteorological anomaly coefficient value;

[0010] Step 5: Analyze the real-time hydrological data, geographic data and meteorological data collected in the corresponding area to obtain the regional water level coefficient and ratio B.

[0011] As a preferred implementation of the present invention, the specific process of calculating the mean of all comparison data sets to obtain the mean set is as follows:

[0012] If in the comparison data set 1, for a certain feature label data F, at time point t i There is m i Each piece of data represents a vector x ij , x ij Contains multiple attributes; calculates all the time points t at which disasters occur i The mean of all attributes corresponding to the feature labeling data F is obtained by statistics; for a certain attribute A, the mean is calculated as follows: Output the mean of attribute A A ; Among them, A(X ij ) is the vector x ij The value of attribute A in It represents the sum of attribute A at all time points when disasters occur; Indicates the total number of data at all disaster time points.

[0013] As a preferred implementation of the present invention, the specific process of analyzing the hydrological difference data set, the hydrological difference over-limit value set and the hydrological difference critical value set to obtain the hydrological anomaly coefficient value is as follows:

[0014] By establishing the formula: Output the hydrological anomaly coefficient value Ks, where Di1 is the difference value of the i1th record in the hydrological difference data set, Wi1 is the preset weight of the i1th record, and n1 is the total number of records in the hydrological difference data set;

[0015] Then count the contribution values ​​of the hydrological difference over-limit value set and the hydrological difference critical value set: calculate the weighted mean of the hydrological difference over-limit value set: Output weighted super-bound mean μ 超界 ; Calculate the weighted mean for the set of hydrological difference critical values: Output weighted critical mean μ 临界 ; Hydrological anomaly coefficient value Ks = ω1 × μ 超界 +ω2×μ 临界 , ω1 and ω2 are weight factors, satisfying ω1+ω2=1, and ω1 / ω2=λ, λ is the adjustment coefficient.

[0016] As a preferred implementation of the present invention, the specific process of analyzing the geographical difference data set, the geographical difference over-boundary value set and the geographical difference critical value set to obtain the geographical anomaly coefficient value is as follows:

[0017] Get each record data using a vector x i2 , including the slope value S i2 、Altitude H i2 and terrain roughness P i2 , assign a weight W to each record i2 , by establishing the formula: W i2 =a1×S i2 +a2×(Hmax-H i2 )+a3×P i2 Output the weight W assigned to the i2th record i2 , a1, a2, and a3 are all preset weight coefficients, and Hmax is the maximum altitude of the study area;

[0018] By establishing the formula: Output geographic anomaly coefficient value Kd, where is the difference value of the i2th record in the geographic difference dataset, is the absolute difference between the geographical difference over-limit value and the critical value, n2 is the total number of records in the geographical difference data set, λ1 is the adjustment coefficient, and m1 is the number of overlapping records between the geographical difference over-limit value set and the critical value set.

[0019] As a preferred embodiment of the present invention, the specific process of analyzing the meteorological difference data set, the meteorological difference over-limit value set and the meteorological difference critical value set to obtain the meteorological anomaly coefficient value is:

[0020] By establishing the formula:

[0021]

[0022] The meteorological anomaly coefficient value Kq is obtained. Contribute to the super-limit part, is the difference value of the i3th record in the meteorological difference exceeding threshold value set, To perform logarithmic transformation on the over-limit values, W i3 The preset weight of the i3th record, is the total weight of the part exceeding the threshold; Contribute to the critical value part, is the difference value of the j3th record in the meteorological difference critical value set, To perform a square root transformation on the critical value, W j3 The preset weight of the j3rd record, is the total weight of the critical value part, ω3 and ω4 are weight factors, satisfying ω3+ω4=1, ω3 / ω4=λ2, λ2 is the adjustment coefficient.

[0023] As a preferred embodiment of the present invention, the specific process of analyzing the real-time collected hydrological data, geographic data and meteorological data in the corresponding area to obtain the regional water level coefficient is:

[0024] By formula: Output area water level coefficient CSZ, T1 time series length, Q1 is the amount of water flowing in or out per unit time, R1 is the rainfall per unit area, K1 s is the soil permeability coefficient, S1 is the terrain slope, E1 is the evaporation, A1 is the normalization coefficient, which normalizes the net change to unit area; Q1 t1 +R1 t1 -E1 t1 is the net change in water volume, K1 s ×S1 is the correction of water flow velocity due to terrain and soil.

[0025] As a preferred embodiment of the present invention, the specific process of analyzing the real-time collected hydrological data, geographic data and meteorological data in the corresponding area to obtain the ratio B is:

[0026] Compare the regional water level value SWZ calculated in step 4 with the regional water level coefficient CSZ to obtain the ratio between the regional water level coefficient CSZ and the regional water level value SWZ, recorded as B; then compare the ratio B with the preset warning interval, and the preset warning interval includes warning interval one, warning interval two and warning interval three; if the ratio B belongs to warning interval one, it is a level three warning, and a level three alarm is issued to notify relevant personnel and departments to ensure that they pay attention to water level changes and prepare for possible flood risks; if the ratio B belongs to warning interval two, it is a level two warning, and a level two alarm is issued. Relevant personnel and departments should strengthen inspections and initiate emergency response measures; if the ratio B belongs to warning interval three, it is a level one warning, and a level one alarm is issued to immediately notify all relevant departments to conduct a comprehensive emergency response.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] 1. The present invention combines multi-dimensional data such as hydrology, geography and meteorology, and provides an accurate means of flood prediction and early warning through systematic feature screening, statistical analysis and anomaly detection. Steps one to four identify possible high-risk periods through in-depth analysis of historical and real-time data, especially in cases of rapid water level changes, excessive precipitation or extreme weather, so that potential disaster risks can be discovered in advance. By calculating the anomaly coefficients of hydrology, geography and meteorology, a comprehensive regional water level value is obtained, and an alarm is issued in a timely manner according to the set warning interval to ensure that relevant personnel and departments can take appropriate preventive measures according to different warning levels, such as strengthening inspections, deploying flood control facilities, evacuating personnel, etc., thereby effectively reducing disaster losses.

[0029] 2. The present invention has comprehensive hydrological monitoring and early warning capabilities, which not only improves the monitoring accuracy of water level changes, but also enhances the early identification and decision-making support capabilities of disaster risks. Through the comprehensive analysis of multi-source data, it is possible to quickly capture abnormal fluctuations in real-time data changes, and quantify the risk level through accurate abnormal coefficient values. This intelligent early warning system can greatly improve the response speed and accuracy of disaster prevention and mitigation, and avoid huge losses caused by delayed response. In addition, combined with the analysis of regional characteristics, this method is not only suitable for flood warnings, but also can effectively respond to natural disasters caused by other water levels. Overall, this method provides governments at all levels and relevant departments with a set of efficient and scientific disaster prevention and mitigation decision-making tools, improves the efficiency and accuracy of emergency response, and optimizes the allocation and utilization of resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. The following drawings are not intentionally scaled to the actual size, and the focus is on illustrating the main purpose of the present invention.

[0031] Figure 1 It is a diagram of the method steps of the present invention. DETAILED DESCRIPTION

[0032] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present invention.

[0033] As shown in the present invention and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular, but also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list, and the method or device may also include other steps or elements.

[0034] Although the present invention has made various references to certain modules in the system according to an embodiment of the present invention, any number of different modules can be used and run on a user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0035] The present invention uses a flow chart to illustrate the operations performed by the system according to an embodiment of the present invention. It should be understood that the preceding or following operations are not necessarily performed precisely in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. At the same time, other operations may also be added to these processes, or one or more operations may be removed from these processes.

[0036] Below, the exemplary embodiments according to the present invention will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments of the present invention, and it should be understood that the present invention is not limited to the exemplary embodiments described here.

[0037] See also Figure 1 As shown, a water level series data monitoring and early warning method based on big data, the specific steps are:

[0038] Step 1: Obtain hydrological data, geographic data and meteorological data of a certain location area through big data; hydrological data include water level, flow, precipitation, evaporation, sand content and groundwater level, etc.; geographic data include topographic data, land cover type (forest, grassland, bare land, cultivated land, water body, etc.), water system distribution, geological conditions and water conservancy facilities, etc.; meteorological data include temperature, precipitation, wind speed and direction, weather type and historical extreme values, etc.;

[0039] Step 2: Obtain the hydrological data, geographic data and meteorological data obtained in step 1 for data screening, specifically:

[0040] Visualize hydrological data, geographic data, and meteorological data, and draw line graphs to show the time-varying curves of water level, flow, or precipitation in a time series; mark the water level change trend in the graph, and mark the peak points and rapid rising sections; if the water level rises significantly within a few hours, highlight these time periods in red in the graph; and then draw heat maps for spatial distribution analysis: for example, rainfall data is mapped into heat maps through geographic coordinates to show areas with concentrated rainfall; mark the rainfall intensity with different colors in the graph to compare whether the heavy rainfall areas overlap with low-lying areas;

[0041] By calculating statistical indicators, including basic statistics: mean, variance, extreme value; obtaining obvious characteristic standards (obvious characteristic standards are determined by experts in this field combined with professional experience in hydrology, geography and meteorology to determine which values ​​or changes can be directly defined) including: extreme value points, violent fluctuation time periods, and trends;

[0042] Hydrological data: Warning water level screening: extract the time point when the water level exceeds the set warning value (such as 5 meters) and mark it with obvious characteristics; further analyze the flow or rainfall in this period for screening: screen the time points corresponding to the records that exceed P times the historical average (P = 1, 2, 3, ...) and mark them with obvious characteristics;

[0043] Geographic data: Terrain screening: Screening terrain data with a slope greater than the corresponding slope threshold (e.g., slope > 30°) or terrain data with a terrain elevation less than the corresponding elevation may lead to flash flood risks and mark them with obvious features;

[0044] Meteorological data: Extreme weather screening: Screen out time points with high wind speed (such as wind speed > 10m / s) or heavy rainfall (such as rainfall > 50mm / h); mark them with obvious characteristics;

[0045] By analyzing the characteristic changes in continuous time periods, potential high-risk periods can be identified: such as continuous rainfall exceeding the standard detection: detect whether the precipitation for three consecutive hours exceeds the corresponding precipitation threshold (such as the cumulative precipitation for 3 consecutive hours is >100mm), and mark it with obvious characteristics; rapid water level change detection: detect the time period when the water level rises rapidly in a short period of time (such as the water level rises by >1 meter within 1 hour), and mark it with obvious characteristics.

[0046] Step 3: Obtain data corresponding to obvious feature markers and analyze them:

[0047] Obtain the corresponding time points of the data with obvious feature markers through big data, and obtain all the data at the corresponding time points; if floods, mountain torrents or other water level natural disasters have occurred at the corresponding time points, it is recorded as comparison data set 1; if floods, mountain torrents or other water level natural disasters do not exist at the corresponding time points, it is recorded as comparison data set 2;

[0048] For each time point, the data of disasters that have occurred at that time point will be screened out from the comparison data set 1, and the mean of all comparison data sets 1 will be calculated:

[0049] If in the comparison data set 1, for a certain feature label data F, at time point t i (i=1,2,...,n) has m i Each piece of data can be represented as a vector x ij (j=1,2,...,m i ), where x ij Contains multiple attributes (such as rainfall, water level, wind speed, etc.); calculates the time points t at which all disasters occur i The mean of all attributes corresponding to the feature label data F is obtained by statistics; for example, for a certain attribute A (such as rainfall), the mean calculation formula is: Output attribute A mean; where A(X ij ) is the vector x ij The value of attribute A in It represents the sum of attribute A at all time points when disasters occur; Indicates the total number of data at all disaster time points;

[0050] Compare the difference between the comparison data set 1 and the comparison data set 2 to obtain the difference data set between the two; then compare the difference data set with the mean set corresponding to the data, and record the difference data set greater than or equal to the mean set corresponding to the data as the difference over-limit value set; record the difference data set less than the mean set corresponding to the data as the difference critical value set;

[0051] Step 4: Retrieve the difference data set, difference over-limit value set and difference critical value set obtained in step 3; then classify the three to obtain the difference data set, difference over-limit value set and difference critical value corresponding to hydrology, geography and meteorology;

[0052] The hydrological anomaly coefficient value is obtained by analyzing the hydrological difference data set, the hydrological difference over-boundary value set and the hydrological difference critical value set; it should be noted that the hydrological difference data set includes all difference data related to hydrology, the hydrological difference over-boundary value set includes data with hydrological difference values ​​greater than or equal to the corresponding mean, and the hydrological difference critical value set includes data with hydrological difference values ​​less than the corresponding mean;

[0053] The calculation process of the hydrological anomaly coefficient value Ks (used to quantify the degree of anomaly in hydrological data) is as follows: Output the hydrological anomaly coefficient value Ks, where Di1 is the difference value of the i1th record in the hydrological difference data set, Wi1 is the preset weight of the i1th record, which is usually set based on the importance or impact of the data. For example, if the data is associated with a specific warning water level, a higher weight is set; if the data reflects an extreme point or a violent fluctuation segment, a higher weight is set; n1 is the total number of records in the hydrological difference data set;

[0054] Then count the contribution values ​​of the hydrological difference over-limit value set and the hydrological difference critical value set: calculate the weighted mean of the hydrological difference over-limit value set: Output weighted super-bound mean μ 超界 ; Calculate the weighted mean for the set of hydrological difference critical values: Output weighted critical mean μ 临界 ; If the over-limit value contributes more to the total weight, Ks tends to a larger value, indicating the degree of anomaly; if the critical value contributes more, Ks tends to a smaller value, indicating that the degree of anomaly is low or close to the mean; Hydrological anomaly coefficient value Ks = ω1×μ 超界 +ω2×μ 临界 , where ω1 and ω2 are weight factors, satisfying ω1+ω2=1, ω1 / ω2=λ, λ is the adjustment coefficient used to balance the contribution of the two parts;

[0055] The geographic difference data set, geographic difference over-limit value set and geographic difference critical value set are analyzed to obtain the geographic anomaly coefficient value; similarly: each record data is represented by a vector x i2 =(Si2, Hi2, Pi2...) indicates that it contains the slope value S i2 、Altitude H i2 and terrain roughness P i2 etc., and assign a weight W to each record i2 , by establishing the formula: W i2 =a1×S i2 +a2×(Hmax-H i2 )+a3×P i2 Output the weight W assigned to the i2th record i2 , where a1, a2, and a3 are preset weight coefficients, reflecting the relative importance of slope, altitude, and terrain roughness (which can be adjusted according to actual conditions); Hmax is the maximum altitude of the study area;

[0056] The calculation process of the geographic anomaly coefficient value Kd (used to quantify the degree of anomaly in geographic data) is as follows: Output geographic anomaly coefficient value Kd, where is the difference value of the i2th record in the geographic difference dataset, is the absolute difference between the geographical difference over-limit value and the critical value, n2 is the total number of records in the geographical difference data set, λ1 is the adjustment coefficient used to balance the contribution of the two parts, and m1 is the number of overlapping records between the geographical difference over-limit value set and the critical value set;

[0057] The meteorological anomaly coefficient values ​​are obtained by analyzing the meteorological difference data set, the meteorological difference over-limit value set and the meteorological difference critical value set, which are as follows:

[0058] By establishing the formula: Output meteorological anomaly coefficient value Kq, where Contribute to the super-limit part, is the difference value of the i3th record in the meteorological difference exceeding threshold value set, In order to perform logarithmic transformation on the over-limit values, amplify the impact of high-value outliers, and reduce the interference of small values ​​on the whole, W i3 The preset weight of the i3th record, is the total weight of the part exceeding the threshold; Contribute to the critical value part, is the difference value of the j3th record in the meteorological difference critical value set, In order to perform square root transformation on the critical value, reduce the influence of high-value outliers and enhance the relative contribution of small values, W j3 The preset weight of the j3rd record, is the total weight of the critical value part, ω3 and ω4 are weight factors, satisfying ω3+ω4=1, ω3 / ω4=λ2, λ2 is the adjustment coefficient used to balance the contribution of the two parts;

[0059] Then the calculated hydrological anomaly coefficient value Ks, geographical anomaly coefficient value Kd and meteorological anomaly coefficient value Kq are comprehensively calculated: the regional water level value SWZ is output through the established formula: SWZ = Ks × b1 + Kd × b2 + Kq × b3, b1, b2, b3 are all preset weight coefficients, and b1 + b2 + b3 = 1;

[0060] Step 5: Analyze the real-time hydrological data, geographic data and meteorological data collected in the corresponding area to obtain the regional water level coefficient, which is:

[0061] By formula: Output regional water level coefficient CSZ, where T1 is the length of the time series, Q1 is the amount of water flowing in or out per unit time, R1 is the rainfall per unit area, and K1 is the s is the soil permeability coefficient, S1 is the terrain slope, E1 is the evaporation, A1 is the normalization coefficient, which normalizes the net change to unit area; Q1 t1 +R1 t1 -E1 t1is the net change in water volume, K1 s ×S1 is the correction of water velocity due to topography and soil;

[0062] The regional water level value SWZ calculated in step 4 is compared with the regional water level coefficient CSZ to obtain the ratio between the regional water level coefficient CSZ and the regional water level value SWZ, which is recorded as B; the ratio B is then compared with the preset warning interval, and the preset warning interval includes warning interval one, warning interval two and warning interval three; if the ratio B belongs to warning interval one, it is a level three warning, and a level three alarm is issued to notify relevant personnel and departments to ensure that they pay attention to water level changes and prepare for possible flood risks; if the ratio B belongs to warning interval two, it is a level two warning, and a level two alarm is issued. Relevant personnel and departments should strengthen inspections and initiate emergency response measures, such as deploying flood control facilities in advance and restricting the entry and exit of personnel; if the ratio B belongs to warning interval three, it is a level one warning, and a level one alarm is issued to immediately notify all relevant departments to carry out a comprehensive emergency response, including evacuation of personnel, blockade of dangerous areas, emergency rescue, etc.

[0063] In addition, those skilled in the art will appreciate that various aspects of the present invention may be illustrated and described by a number of patentable categories or situations, including any new and useful process, machine, product, or combination of substances, or any new and useful improvements thereto. Accordingly, various aspects of the present invention may be performed entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. The above hardware or software may be referred to as "data blocks", "modules", "engines", "units", "components" or "systems". In addition, various aspects of the present invention may be represented as a computer product located in one or more computer-readable media, which includes computer-readable program code.

[0064] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the invention belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.

[0065] The above is an explanation of the present invention and should not be considered as a limitation thereof. Although several exemplary embodiments of the present invention have been described, it will be readily appreciated by those skilled in the art that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present invention. Therefore, all such modifications are intended to be included within the scope of the present invention as defined in the claims. It should be understood that the above is an explanation of the present invention and should not be considered as being limited to the specific embodiments disclosed, and modifications to the disclosed embodiments and other embodiments are intended to be included within the scope of the appended claims. The present invention is defined by the claims and their equivalents.

Claims

1. A water level series data monitoring and early warning method based on big data, characterized in that: include: Step 1: Obtain hydrological, geographical and meteorological data of a specific area through a big data platform; Step 2: Screen the hydrological, geographic and meteorological data, specifically: visualize the hydrological, geographic and meteorological data, display the trend of water level and flow over time through line graphs, and mark the peak points and sharp change sections; display the spatial distribution of precipitation and meteorological data through heat maps, and check whether the heavy rainfall area overlaps with the low-lying terrain; at the same time, calculate basic statistics and identify extreme values ​​and key characteristic points of sharp fluctuations; Step 3: Obtain the corresponding time points of the data with obvious feature markers through big data, and obtain all the data at the corresponding time points; if floods or mountain torrents have occurred at the corresponding time points, obtain the comparison data set 1 by statistics; if floods or mountain torrents do not exist at the corresponding time points, obtain the comparison data set 2 by statistics; at each time point, filter out the data of the disaster that has occurred at that time point from the comparison data set 1, and calculate the mean of all the comparison data sets 1 to obtain the mean set, compare the comparison data set 1 with the comparison data set 2 to obtain the difference data set between the two; then compare the difference data set with the mean corresponding to the data, and obtain the difference over-boundary value set for the difference data set that is greater than or equal to the mean corresponding to the data; and obtain the difference critical value set for the difference data set that is less than the mean corresponding to the data; Step 4: retrieve the difference data set, difference over-boundary value set and difference critical value set obtained in step 3; classify the three to obtain the difference data set, difference over-boundary value set and difference critical value corresponding to hydrology, geography and meteorology; analyze the hydrological difference data set, hydrological difference over-boundary value set and hydrological difference critical value set to obtain the hydrological anomaly coefficient value; The geographic anomaly coefficient values ​​were obtained by analyzing the geographic difference data set, the geographic difference over-limit value set and the geographic difference critical value set; The meteorological anomaly coefficient value is obtained by analyzing the meteorological difference data set, the meteorological difference over-limit value set and the meteorological difference critical value set; Step 5: Analyze the real-time hydrological data, geographic data and meteorological data collected in the corresponding area to obtain the regional water level coefficient and ratio B.

2. The water level series data monitoring and early warning method based on big data according to claim 1 is characterized in that: And calculate the mean of all comparison data sets to get the specific process of the mean set: If in the comparison data set 1, for a certain feature label data F, at time point t i There is m i Each piece of data represents a vector x ij , x ij Contains multiple attributes; calculates all the time points t at which disasters occur i The mean of all attributes corresponding to the feature labeling data F is obtained by statistics; for a certain attribute A, the mean is calculated as follows: Output the mean of attribute A A ; Among them, A(X ij ) is the vector x ij The value of attribute A in It represents the sum of attribute A at all time points when disasters occur; Indicates the total number of data at all disaster time points.

3. The water level series data monitoring and early warning method based on big data according to claim 1 is characterized in that: The specific process of analyzing the hydrological difference data set, the hydrological difference over-limit value set and the hydrological difference critical value set to obtain the hydrological anomaly coefficient value is as follows: By establishing the formula: Output the hydrological anomaly coefficient value Ks, where Di1 is the difference value of the i1th record in the hydrological difference data set, Wi1 is the preset weight of the i1th record, and n1 is the total number of records in the hydrological difference data set; Then count the contribution values ​​of the hydrological difference over-limit value set and the hydrological difference critical value set: calculate the weighted mean of the hydrological difference over-limit value set: Output weighted super-bound mean μ 超界 ; Calculate the weighted mean for the set of hydrological difference critical values: Output weighted critical mean μ 临界 ; Hydrological anomaly coefficient value Ks = ω1 × μ 超界 +ω2×μ 临界 , ω1 and ω2 are weight factors, satisfying ω1+ω2=1, and ω1 / ω2=λ, λ is the adjustment coefficient.

4. The water level series data monitoring and early warning method based on big data according to claim 3 is characterized in that: The specific process of analyzing the geographic difference data set, the geographic difference over-boundary value set and the geographic difference critical value set to obtain the geographic anomaly coefficient value is as follows: Get each record data using a vector x i2 , including the slope value S i2 、Altitude H i2 and terrain roughness P i2 , assign a weight W to each record i2 , by establishing the formula: W i2 =a1×S i2 +a2×(Hmax-H i2 )+a3×P i2 Output the weight W assigned to the i2th record i2 , a1, a2, and a3 are all preset weight coefficients, and Hmax is the maximum altitude of the study area; By establishing the formula: Output geographic anomaly coefficient value Kd, where is the difference value of the i2th record in the geographic difference dataset, is the absolute difference between the geographical difference over-limit value and the critical value, n2 is the total number of records in the geographical difference data set, λ1 is the adjustment coefficient, and m1 is the number of overlapping records between the geographical difference over-limit value set and the critical value set.

5. The water level series data monitoring and early warning method based on big data according to claim 4 is characterized in that: The specific process of analyzing the meteorological difference data set, the meteorological difference over-boundary value set and the meteorological difference critical value set to obtain the meteorological anomaly coefficient value is as follows: By establishing the formula: Output meteorological anomaly coefficient value Kq, Contribute to the super-limit part, is the difference value of the i3th record in the meteorological difference exceeding threshold value set, To perform logarithmic transformation on the over-limit values, W i3 The preset weight of the i3th record, is the total weight of the part exceeding the threshold; Contribute to the critical value part, is the difference value of the j3th record in the meteorological difference critical value set, To perform a square root transformation on the critical value, W j3 The preset weight of the j3rd record, is the total weight of the critical value part, ω3 and ω4 are weight factors, satisfying ω3+ω4=1, ω3 / ω4=λ2, λ2 is the adjustment coefficient.

6. The water level series data monitoring and early warning method based on big data according to claim 1 is characterized in that: The specific process of analyzing the real-time collected hydrological data, geographic data and meteorological data in the corresponding area to obtain the regional water level coefficient is as follows: By formula: Output area water level coefficient CSZ, T1 time series length, Q1 is the amount of water flowing in or out per unit time, R1 is the rainfall per unit area, K1 s is the soil permeability coefficient, S1 is the terrain slope, E1 is the evaporation, A1 is the normalization coefficient, which normalizes the net change to unit area; Q1 t1 +R1 t1 -E1 t1 is the net change in water volume, K1 s ×S1 is the correction of water flow velocity due to terrain and soil.

7. The water level series data monitoring and early warning method based on big data according to claim 1 is characterized in that: The specific process of analyzing the real-time collected hydrological data, geographic data and meteorological data in the corresponding area to obtain the ratio B is: Compare the regional water level value SWZ calculated in step 4 with the regional water level coefficient CSZ to obtain the ratio between the regional water level coefficient CSZ and the regional water level value SWZ, recorded as B; then compare the ratio B with the preset warning interval, and the preset warning interval includes the preset warning interval 1, the preset warning interval 2 and the warning interval 3; if the ratio B belongs to the warning interval 1, it is a level 3 warning, and a level 3 alarm is issued to notify relevant personnel and departments to ensure that they pay attention to water level changes and prepare for possible flood risks; if the ratio B belongs to the warning interval 2, it is a level 2 warning, and a level 2 alarm is issued. Relevant personnel and departments should strengthen inspections and initiate emergency response measures; If the ratio B belongs to the warning interval three, it is a level one warning, a level one alarm is issued, and all relevant departments are immediately notified to conduct a comprehensive emergency response.

Images