Flood early warning comprehensive dynamic index determination method

Abstract

The invention discloses a flood early warning comprehensive dynamic index determination method, which comprises the following steps: collecting target index data, and normalizing each index value; calculating a subjective weight value and an objective weight value of each index data according to a subjective and objective combined weight method; the subjective weight value and the objective weight value are combined to calculate the consistency coefficient of subjective and objective weights, a long-term low-correlation index replacement mechanism is established, and then the weights are dynamically adjusted; determining a comprehensive dynamic index in combination with the corresponding index normalization data; risk grades are divided according to the numerical range of the comprehensive dynamic indexes, and a judgment basis is provided for decision makers. According to the method, various flood related factors are comprehensively considered, index weight values at different moments are calculated, a systematic early warning index determination method is constructed, the limitation of a traditional method is effectively overcome, flexible adaptation can be achieved according to real-time data and flood scenes, more accurate and more adaptive technical means are provided for flood early warning, and disaster prevention and reduction work is assisted.

Description

Technical Field

[0001] The present invention relates to the technical field of flood early warning, and in particular to a method for determining a comprehensive dynamic index of flood early warning. Background Art

[0002] Accurate and timely flood warnings are crucial for disaster prevention and mitigation. With the continuous advancement of science and technology, flood warning technology is also constantly developing. However, there are still some problems with existing flood warning methods.

[0003] Traditional flood warning methods typically rely on a single indicator, such as water level or rainfall, to assess flood risk. While simple and intuitive, this approach has significant limitations. The formation and development of floods is a complex natural process influenced by a variety of factors. A single indicator cannot fully reflect the true extent of flood risk. For example, assessing flood risk based solely on water level may overlook the role of other important factors, such as rainfall intensity, soil moisture, and topography, resulting in reduced accuracy and reliability of warnings. In practice, single-indicator approaches often fail to capture early signs of flood development. For example, even if the water level has not yet reached the warning line, abnormal changes in rainfall intensity and soil moisture may already indicate an impending flood. Focusing solely on water levels can miss the optimal opportunity for early warning, preventing the implementation of effective flood prevention measures.

[0004] Most existing flood warning methods use fixed warning indicators and thresholds, which cannot be dynamically adjusted based on real-time data and changes in flood scenarios. Different regions have significant differences in topography, climate conditions, and population distribution. Fixed warning indicators and thresholds are difficult to adapt to these differences, resulting in insufficient adaptability and flexibility in warnings. In highly urbanized areas, due to the high degree of ground hardening, the large runoff coefficient of rainfall, and the rapid rise of floods, more sensitive warning indicators are needed. In rural areas, due to the more complex terrain and diverse flood propagation paths, warnings need to be comprehensively considered based on multiple factors. Fixed warning indicators cannot meet these diverse needs, resulting in insufficient accuracy and timeliness of warnings.

[0005] Existing flood warning methods often fail to fully consider the spatiotemporal distribution characteristics of rainfall and flooding when assessing flood risk. In reality, flood risk depends not only on the absolute values ​​of rainfall and water levels but also on their temporal and spatial distribution. For example, concentrated rainfall over a short period of time and prolonged rainfall have different impacts on flood risk, and flood risks also differ between urban cores and suburban areas. Failure to fully account for these spatiotemporal distribution characteristics can lead to inaccurate and delayed warnings, hindering timely and effective support for disaster prevention and mitigation. For example, even with the same rainfall, concentrated rainfall in a city center poses a much greater risk of flooding than concentrated rainfall in the suburbs. Failure to distinguish these differences in warning methods can result in untimely warnings for high-risk areas, increasing the risk of casualties and property damage.

[0006] In view of the above-mentioned deficiencies of existing flood early warning methods, the present invention proposes a method for determining comprehensive dynamic indicators for flood early warning, which can comprehensively consider multiple factors and dynamically adjust indicator weights. Summary of the Invention

[0007] The purpose of the present invention is to provide a method for determining comprehensive dynamic indicators of flood warning, which can comprehensively consider multiple factors and dynamically adjust the indicator weights, so as to overcome the shortcomings of the existing technology and improve the accuracy and adaptability of flood warning.

[0008] To achieve the above functions, the present invention designs a method for determining a comprehensive dynamic index for flood warning, which performs the following steps S1 to S4 to complete the determination of the comprehensive dynamic index for flood warning and flood warning:

[0009] Step S1: for the target watershed, select multiple indicators related to the natural environment and human activities of the target watershed, collect data for each indicator, and classify the collected indicator data into a normal processing set and a to-be-replaced set;

[0010] Step S2: normalizing the indicator data in the normal processing set, and calculating the subjective weight value and objective weight value of each indicator data in the normal processing set according to the subjective and objective combined weight method;

[0011] Step S3: Based on the subjective and objective weight values ​​of each indicator data, the consistency coefficient of the subjective and objective weights is calculated to determine the consistency degree of the subjective and objective weights of the indicator. The correlation coefficient between the indicator and the flood peak flow is calculated, the indicator is reclassified into a normal processing set and a set to be replaced, and the weight of the indicator is dynamically adjusted.

[0012] Step S4: Multiply the normalized index by its corresponding dynamically adjusted weight and sum them to obtain a comprehensive dynamic index. Flood risk levels are divided according to the values ​​of the comprehensive dynamic index to issue flood warnings.

[0013] As a preferred technical solution of the present invention: the indicators selected in step S1 include hydrology, meteorology, topography, human activities and extreme flood events.

[0014] As a preferred technical solution of the present invention: Step S2 specifically comprises the following steps:

[0015] Step S2.1: Perform Min-Max normalization on the indicator data in the normal processing set to obtain the normalized value x of indicator n at time m nm , n=1,2,3…N, where N is the total number of indicators, m=1,2,3…M, where M is the total number of data for indicator n;

[0016] Step S2.2: Construct a judgment matrix based on the analytic hierarchy process and the 1-9 scaling method to obtain the subjective weight value α of indicator n at time m nm ;

[0017] Step S2.3: Based on the entropy weight method, calculate the entropy value of indicator n at time m, and determine its corresponding objective weight value β according to the entropy value nm .

[0018] As a preferred technical solution of the present invention: Step S2.2 specifically comprises the following steps:

[0019] Step S2.2.1: Evaluate the size of each indicator based on the 1-9 scale method and construct the judgment matrix P;

[0020] Step S2.2.2: Find the sum of the relative importance levels of each row in the judgment matrix P and the sum of all relative importance levels:

[0021]

[0022] Where a ij is the data in the i-th row and j-th column of the matrix P; r is the number of rows of the matrix P, c is the number of columns of the matrix P; V nm (i) V is the sum of the relative importance levels of index n in row i of matrix P at time m; nm is the sum of all relative importance of index n in the matrix P at time m.

[0023] Step S2.2.3: Calculate the subjective weight of each indicator:

[0024]

[0025] Where, α nm is the subjective weight value of indicator n at time m.

[0026] As a preferred technical solution of the present invention: Step S2.3 specifically comprises the following steps:

[0027] Step S2.3.1: Calculate the weight of indicator n at time m according to the following formula:

[0028]

[0029] Where, τ nm is the proportion of index n at time m, M is the total number of data of index n, x nm is the normalized value of index n at time m;

[0030] According to the following formula, the entropy value of indicator n at time m is calculated:

[0031]

[0032] Where, e nm is the entropy value of index n at time m, and M is the total number of data of index n;

[0033] Step S2.3.2: Based on the entropy value calculated in step S3.2.1, calculate the coefficient of variation for each indicator:

[0034] d nm =1-e nm

[0035] Where, d nm is the difference coefficient of indicator n at time m;

[0036] Calculate the objective weight value of each indicator according to the following formula:

[0037]

[0038] Where, β nm is the objective weight value of indicator n at time m.

[0039] As a preferred technical solution of the present invention: Step S3 specifically comprises the following steps:

[0040] Step S3.1: Based on the subjective weight value and objective weight value of each indicator data, calculate the consistency coefficient of the subjective and objective weights:

[0041]

[0042] Where c n is the consistency coefficient of the subjective and objective weights of indicator n, if c n ≥0.8, the subjective and objective weights are considered to be highly consistent; if 0.5≤c n <0.8, it is considered that the subjective and objective weights are basically consistent; if c n<0.5, it is considered that the subjective and objective weights are in significant conflict;

[0043] Step S3.2: Dynamically adjust the weight according to the consistency coefficient, specifically including the following steps S3.2.1-S3.2.3:

[0044] Step S3.2.1: If the subjective and objective weights of indicator n calculated in step S3.1 are highly consistent, use the Pearson correlation coefficient method to calculate the correlation coefficient r between indicator n and the peak flow of the most recent flood. n ;

[0045] If r n ≥0.7, the weight assigned to indicator n is calculated according to the following formula:

[0046]

[0047] Where μ nm is the weight after dynamic adjustment;

[0048] If 0.3≤r n <0.7, the weight assigned to indicator n is calculated according to the following formula:

[0049]

[0050] If r n <0.3, then mark the index n as the index to be replaced, add it to the set to be replaced, and go to step S3.3;

[0051] Step S3.2.2: If the subjective and objective weights of the indicator n calculated in step S3.1 conflict significantly, use the Pearson correlation coefficient method to calculate the correlation coefficient r between the indicator n and the peak flow of the most recent flood. n If r n ≥0.7, the weight of indicator n at this moment is considered to be the objective weight β nm , μ nm =β nm Otherwise, the weight assigned to indicator n is calculated according to the following formula:

[0052]

[0053] Where μ nm is the weight after dynamic adjustment;

[0054] Step S3.2.3: If the subjective and objective weights of indicator n calculated in step S3.1 are substantially consistent, the weight assigned to indicator n is calculated according to the following formula:

[0055]

[0056] Where μ nm is the weight after dynamic adjustment;

[0057] Step S3.3: Establish a long-term low-correlation indicator replacement mechanism for the indicators to be replaced: Calculate the correlation coefficient between the indicator data in the set to be replaced and the flood peak flow. If the indicator data in the normal processing set meets c n ≥0.8 and r n <0.3, an indicator with a higher correlation coefficient with flood events is selected from the set to be replaced and the weight is recalculated.

[0058] As a preferred technical solution of the present invention: Step S4 specifically comprises the following steps:

[0059] Step S4.1: Calculate the comprehensive dynamic index according to the following formula:

[0060]

[0061] Where, I m Represents the comprehensive dynamic index, μ nm is the dynamically adjusted weight, x nm is the normalized value of index n at time m;

[0062] Step S4.2: The numerical range of the comprehensive dynamic index is clearly defined as 0-1, and it is divided into three levels of risk: low risk, medium risk, and high risk according to the numerical value of the comprehensive dynamic index. Low risk means that the possibility of flooding is small and the basin is in a relatively safe state. Medium risk means that the possibility of flooding is gradually increasing, and attention should be paid and certain preventive measures should be taken. High risk means that the possibility of flooding is very high, and emergency warnings and response measures need to be initiated immediately.

[0063] As a preferred technical solution of the present invention: if 0≤I m <0.3, it is classified as low risk. If 0.3≤I m <0.6, it is classified as medium risk. If 0.6≤I m <1, it is classified as high risk.

[0064] Beneficial effects: Compared with the prior art, the advantages of the present invention include:

[0065] The present invention comprehensively considers a variety of flood-related factors and provides a systematic method for determining flood warning indicators by calculating dynamic indicator weights, thereby avoiding the limitations of traditional methods that rely on a single indicator and being able to flexibly adapt to changes in real-time data and flood scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 This is a flow chart of a method for determining a comprehensive dynamic index of flood warning provided by an embodiment of the present invention;

[0067] Figure 2 is a schematic diagram of a research basin provided according to an embodiment of the present invention;

[0068] Figure 3 2 is a schematic diagram of dynamic changes in comprehensive indicators provided according to an embodiment of the present invention. DETAILED DESCRIPTION

[0069] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and are not intended to limit the scope of protection of the present invention.

[0070] A method for determining a comprehensive dynamic index of flood warning provided by an embodiment of the present invention is provided. Figure 1 , execute the following steps S1 to S4 to complete the determination of the comprehensive dynamic index of flood warning and flood warning:

[0071] Step S1: For a target watershed, multiple indicators related to the natural environment and human activities of the target watershed are selected. In one embodiment, the selected indicators include hydrology, meteorology, topography, human activities, and extreme flood events; data for each indicator is collected, and the collected indicator data is classified into a normal processing set and a to-be-replaced set;

[0072] Step S2: normalizing the indicator data in the normal processing set, and calculating the subjective weight value and objective weight value of each indicator data in the normal processing set according to the subjective and objective combined weight method;

[0073] The specific steps of step S2 are as follows:

[0074] Step S2.1: Perform Min-Max normalization on the indicator data in the normal processing set to obtain the normalized value x of indicator n at time m nm :

[0075]

[0076] Among them, x nm is the data value of the nth indicator at time m; x nmax and x nmin are the maximum and minimum values ​​of the nth indicator respectively; N is the total number of indicators, and M is the total number of data for indicator n;

[0077] Step S2.2: Construct a judgment matrix based on the analytic hierarchy process and the 1-9 scaling method to obtain the subjective weight value α of indicator n at time m nm ;

[0078] The specific steps of step S2.2 are as follows:

[0079] Step S2.2.1: Evaluate the size of each indicator based on the 1-9 scale method and construct the judgment matrix P;

[0080] Step S2.2.2: Find the sum of the relative importance levels of each row in the judgment matrix P and the sum of all relative importance levels:

[0081]

[0082] Where a ij is the data in the i-th row and j-th column of the matrix P; r is the number of rows of the matrix P, c is the number of columns of the matrix P; V nm (i) V is the sum of the relative importance levels of index n in row i of matrix P at time m; nm is the sum of all relative importance of index n in the matrix P at time m.

[0083] Step S2.2.3: Calculate the subjective weight of each indicator:

[0084]

[0085] Where, α nm is the subjective weight value of indicator n at time m.

[0086] Step S2.3: Based on the entropy weight method, calculate the entropy value of indicator n at time m, and determine its corresponding objective weight value β according to the entropy value nm .

[0087] The specific steps of step S2.3 are as follows:

[0088] Step S2.3.1: Calculate the weight of indicator n at time m according to the following formula:

[0089]

[0090] Where, τ nm is the proportion of index n at time m, M is the total number of data of index n, x nm is the normalized value of index n at time m;

[0091] According to the following formula, the entropy value of indicator n at time m is calculated:

[0092]

[0093] Where, e nm is the entropy value of index n at time m, and M is the total number of data of index n;

[0094] Step S2.3.2: Based on the entropy value calculated in step S3.2.1, calculate the coefficient of variation for each indicator:

[0095] d nm =1-e nm

[0096] Where, d nm is the difference coefficient of indicator n at time m;

[0097] Calculate the objective weight value of each indicator according to the following formula:

[0098]

[0099] Where, β nm is the objective weight value of indicator n at time m.

[0100] Step S3: Based on the subjective and objective weight values ​​of each indicator data, the consistency coefficient of the subjective and objective weights is calculated to determine the consistency degree of the subjective and objective weights of the indicator. The correlation coefficient between the indicator and the flood peak flow is calculated, the indicator is reclassified into a normal processing set and a set to be replaced, and the weight of the indicator is dynamically adjusted.

[0101] The specific steps of step S3 are as follows:

[0102] Step S3.1: Based on the subjective weight value and objective weight value of each indicator data, calculate the consistency coefficient of the subjective and objective weights:

[0103]

[0104] Where c n is the consistency coefficient of the subjective and objective weights of indicator n, if c n ≥0.8, the subjective and objective weights are considered to be highly consistent; if 0.5≤c n <0.8, it is considered that the subjective and objective weights are basically consistent; if c n <0.5, it is considered that the subjective and objective weights are in significant conflict;

[0105] Step S3.2: Dynamically adjust the weight according to the consistency coefficient, specifically including the following steps S3.2.1-S3.2.3:

[0106] Step S3.2.1: If the subjective and objective weights of the indicator n calculated in step S3.1 are highly consistent (i.e., c n ≥0.8), and the correlation coefficient r between the index n and the peak flow of the most recent flood was calculated using the Pearson correlation coefficient method. n ;

[0107] If r n ≥0.7, the weight assigned to indicator n is calculated according to the following formula, that is, the weight of indicator n is increased:

[0108]

[0109] Where μ nm is the weight after dynamic adjustment;

[0110] If 0.3≤r n <0.7, the weight assigned to indicator n is calculated according to the following formula, that is, the weight of indicator n is maintained:

[0111]

[0112] If r n <0.3, then mark the index n as the index to be replaced, add it to the set to be replaced, and go to step S3.3;

[0113] Step S3.2.2: If the subjective and objective weights of the indicator n calculated in step S3.1 conflict significantly (i.e., c n <0.5), and the correlation coefficient r between the index n and the peak flow of the most recent flood was calculated using the Pearson correlation coefficient method. n If r n ≥0.7, the weight of indicator n at this moment is considered to be the objective weight β nm , μ nm =β nm Otherwise, the weight assigned to indicator n is calculated according to the following formula, that is, the weight of indicator n is reduced:

[0114]

[0115] Where μ nm is the weight after dynamic adjustment;

[0116] Step S3.2.3: If the subjective and objective weights of the indicator n calculated in step S3.1 are basically consistent (i.e. 0.5≤c n <0.8), the weight assigned to indicator n is calculated according to the following formula, that is, the weight of maintenance indicator n:

[0117]

[0118] Where μ nm is the weight after dynamic adjustment;

[0119] Step S3.3: Establish a long-term low-correlation indicator replacement mechanism for the indicators to be replaced: Calculate the correlation coefficient between the indicator data in the set to be replaced and the flood peak flow. If the indicator data in the normal processing set meets c n ≥0.8 and r n <0.3, an indicator with a higher correlation coefficient with flood events is selected from the set to be replaced and the weight is recalculated.

[0120] Step S4: Multiply the normalized index by its corresponding dynamically adjusted weight and sum them to obtain a comprehensive dynamic index. Flood risk levels are divided according to the values ​​of the comprehensive dynamic index to issue flood warnings.

[0121] The specific steps of step S4 are as follows:

[0122] Step S4.1: Calculate the comprehensive dynamic index according to the following formula:

[0123]

[0124] Where, I m Represents the comprehensive dynamic index, μ nm is the dynamically adjusted weight, x nm is the normalized value of index n at time m;

[0125] Step S4.2: Define the value range of the comprehensive dynamic index as 0-1, and divide it into three levels: low risk, medium risk, and high risk according to the value of the comprehensive dynamic index; if 0≤I m <0.3, it is classified as low risk, which means that the possibility of flood occurrence is small and the basin is in a relatively safe state; if 0.3≤I m <0.6, it is classified as medium risk, which means that the possibility of flooding is gradually increasing, and attention should be paid and certain preventive measures should be taken; if 0.6≤I m If <1, it is classified as high risk; high risk means that the possibility of flooding is very high and emergency warning and response measures need to be initiated immediately.

[0126] The following is an application embodiment of the present invention:

[0127] Taking the Tunxi River Basin as an example, the study area is as follows Figure 2 The basin is located in the southern Anhui mountainous area in the southeast of Anhui Province. It is an important part of the Xin'an River Basin and has a drainage area of ​​approximately 2670m 2 . The Tunxi River Basin is located in the main area of ​​the Huangshan Mountains. It is adjacent to the Yangtze River system on the west and north sides, and connected to the Tianmu Mountains and Baiji Mountains on the southeast side. There are large-scale mountain basins and valleys in the basin, such as Xiuning, Tunxi, and Shexian. The basin belongs to the subtropical monsoon climate zone, with an average annual precipitation of about 1,800 mm and an average annual temperature of 15.4℃~16.8℃. The vegetation coverage is good, mainly including evergreen coniferous forests, deciduous broad-leaved forests, mixed forests, shrub forests, pastures and cultivated land. The elevation of the Tunxi River Basin decreases from west to east, with a maximum elevation of 1,611 meters, a minimum elevation of 92 meters, and an average elevation of 850.525692 meters. The terrain in the basin is high in the west and low in the east, with steep slopes and rapid currents, and the maximum drop is 1,519 meters.

[0128] Step 1: Collect relevant data; take time t as an example, the flow rate is Q, the soil saturated water content θs ; Rainfall R, cumulative rainfall R cum ; Elevation data Z; Population density data ρ; Peak flow of the most recent flood Q m The normal treatment set indicators are flow rate Q, soil saturated water content θ s ; Rainfall R, the data set to be processed is the cumulative rainfall R cum ; Elevation data Z; Population density data ρ.

[0129] Step 2: Normalize the data. 1t , soil saturated water content x 2t , rainfall x 3t , cumulative rainfall x 4t , average elevation x 5t , population density x 6t ;

[0130] Calculate weights;

[0131] 1) Calculate subjective weights using the analytic hierarchy process;

[0132] Construct the judgment matrix P:

[0133]

[0134] According to the formula:

[0135]

[0136] The weights of the corresponding indicators are: flow α 1t , soil saturated water content α 2t , rainfall α 3t .

[0137] 2) The entropy weight method is used to calculate the objective weight.

[0138] According to the formula:

[0139]

[0140] d nm =1-e nm

[0141]

[0142] Get the weight of each indicator: traffic β 1t , soil saturated water content β 2t , rainfall β 3t .

[0143] Step 3: Adjust weights and screen indicators:

[0144] 1) Calculate the consistency coefficient of subjective and objective weights;

[0145]

[0146] The consistency coefficients of each indicator are: flow c 1t , soil saturated water content c 2t , rainfall c 3t If c n ≥0.8, the subjective and objective weights are considered to be highly consistent; if 0.5≤c n <0.8, it is considered that the subjective and objective weights are basically consistent; if c n <0.5, it is considered that the subjective and objective weights are in significant conflict;

[0147] 2) Dynamically adjust weights based on consistency coefficients:

[0148] a. The subjective and objective weights are highly consistent (c n ≥0.8):

[0149] The Pearson correlation coefficient method was used to calculate the relationship between this index and the peak flow rate Q of the most recent flood. m Correlation coefficient r n ,

[0150] If r n ≥0.7, then strengthen the weight:

[0151]

[0152] If 0.3≤r n <0.7, then maintain the weight:

[0153]

[0154] If r n <0.3 (low correlation), then mark it as an indicator to be replaced and go to step d;

[0155] b. Significant conflict between subjective and objective weights (c n <0.5):

[0156] If the correlation coefficient between this indicator and the peak flow of the most recent flood is r n ≥0.7, the weight at this moment is considered to be the objective weight β nm Otherwise, the weight is reduced to:

[0157]

[0158] c. The subjective and objective weights are basically consistent (0.5≤c n <0.8);

[0159] Maintain the original weight of the portfolio:

[0160]

[0161] d. Establish a long-term low-correlation indicator replacement mechanism:

[0162] Calculate the correlation coefficient between the data of the replacement set and the peak flow. If the normal processing set index meets c n ≥0.8 and r n <0.3, an indicator with a higher correlation coefficient with flood events is selected from the set to be replaced and the weight is recalculated.

[0163] Step 4: Determine comprehensive dynamic indicators and issue flood warnings.

[0164] According to the formula:

[0165]

[0166] Get comprehensive dynamic indicators I t .

[0167] Combined with the flood warning classification, risk warning information is issued. The dynamic change process is shown in Figure 3 .

[0168] The embodiments of the present invention are described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Various changes can be made within the scope of knowledge possessed by ordinary technicians in this field without departing from the spirit of the present invention.

Claims

1. A method for determining comprehensive dynamic indicators of flood warning, characterized in that: Execute the following steps S1 to S4 to complete the determination of the comprehensive dynamic index of flood warning and flood warning: Step S1: for the target watershed, select multiple indicators related to the natural environment and human activities of the target watershed, collect data for each indicator, and classify the collected indicator data into a normal processing set and a to-be-replaced set; Step S2: normalizing the indicator data in the normal processing set, and calculating the subjective weight value and objective weight value of each indicator data in the normal processing set according to the subjective and objective combined weight method; Step S3: Based on the subjective and objective weight values ​​of each indicator data, the consistency coefficient of the subjective and objective weights is calculated to determine the consistency degree of the subjective and objective weights of the indicator. The correlation coefficient between the indicator and the flood peak flow is calculated, the indicator is reclassified into a normal processing set and a set to be replaced, and the weight of the indicator is dynamically adjusted. Step S4: Multiply the normalized index by its corresponding dynamically adjusted weight and sum them to obtain a comprehensive dynamic index. Flood risk levels are divided according to the values ​​of the comprehensive dynamic index to issue flood warnings.

2. A method for determining a comprehensive dynamic index of flood early warning according to claim 1, characterized in that: The indicators selected in step S1 include hydrology, meteorology, topography, human activities and extreme flood events.

3. A method for determining a comprehensive dynamic index of flood early warning according to claim 1, characterized in that: The specific steps of step S2 are as follows: Step S2.1: Perform Min-Max normalization on the indicator data in the normal processing set to obtain the normalized value x of indicator n at time m nm , n=1,2,3…N, where N is the total number of indicators, m=1,2,3…M, where M is the total number of data for indicator n; Step S2.2: Construct a judgment matrix based on the analytic hierarchy process and the 1-9 scaling method to obtain the subjective weight value α of indicator n at time m nm ; Step S2.3: Based on the entropy weight method, calculate the entropy value of indicator n at time m, and determine its corresponding objective weight value β according to the entropy value nm .

4. A method for determining a comprehensive dynamic index of flood early warning according to claim 3, characterized in that: The specific steps of step S2.2 are as follows: Step S2.2.1: Evaluate the size of each indicator based on the 1-9 scale method and construct the judgment matrix P; Step S2.2.2: Find the sum of the relative importance levels of each row in the judgment matrix P and the sum of all relative importance levels: Where a ij is the data in the i-th row and j-th column of the matrix P; r is the number of rows of the matrix P, c is the number of columns of the matrix P; V nm (i) V is the sum of the relative importance levels of index n in row i of matrix P at time m; nm is the sum of all relative importance of index n in the matrix P at time m. Step S2.2.3: Calculate the subjective weight of each indicator: Where, α nm is the subjective weight value of indicator n at time m.

5. A method for determining a comprehensive dynamic index of flood early warning according to claim 3, characterized in that: The specific steps of step S2.3 are as follows: Step S2.3.1: Calculate the weight of indicator n at time m according to the following formula: Where, τ nm is the proportion of index n at time m, M is the total number of data of index n, x nm is the normalized value of index n at time m; According to the following formula, the entropy value of indicator n at time m is calculated: Where, e nm is the entropy value of index n at time m, and M is the total number of data of index n; Step S2.3.2: Based on the entropy value calculated in step S3.2.1, calculate the coefficient of variation for each indicator: d nm =1-e nm Where, d nm is the difference coefficient of indicator n at time m; Calculate the objective weight value of each indicator according to the following formula: Where, β nm is the objective weight value of indicator n at time m.

6. A method for determining a comprehensive dynamic index of flood early warning according to claim 1, characterized in that: The specific steps of step S3 are as follows: Step S3.1: Based on the subjective weight value and objective weight value of each indicator data, calculate the consistency coefficient of the subjective and objective weights: Where c n is the consistency coefficient of the subjective and objective weights of indicator n, if c n ≥0.8, the subjective and objective weights are considered to be highly consistent; if 0.5≤c n <0.8, it is considered that the subjective and objective weights are basically consistent; if c n <0.5, it is considered that the subjective and objective weights are in significant conflict; Step S3.2: Dynamically adjust the weight according to the consistency coefficient, specifically including the following steps S3.2.1-S3.2.3: Step S3.2.1: If the subjective and objective weights of the indicator n calculated in step S3.1 are highly consistent, the correlation coefficient r between the indicator n and the peak flow of the most recent flood is calculated using the Pearson correlation coefficient method. n ; If r n ≥0.7, the weight assigned to indicator n is calculated according to the following formula: Where μ nm is the weight after dynamic adjustment; If 0.3≤r n <0.7, the weight assigned to indicator n is calculated according to the following formula: If r n <0.3, then mark the index n as the index to be replaced, add it to the set to be replaced, and go to step S3.3; Step S3.2.2: If the subjective and objective weights of the indicator n calculated in step S3.1 conflict significantly, use the Pearson correlation coefficient method to calculate the correlation coefficient r between the indicator n and the peak flow of the most recent flood. n If r n ≥0.7, the weight of indicator n at this moment is considered to be the objective weight β nm , μ nm =β nm Otherwise, the weight assigned to indicator n is calculated according to the following formula: Where μ nm is the weight after dynamic adjustment; Step S3.2.3: If the subjective and objective weights of indicator n calculated in step S3.1 are substantially consistent, the weight assigned to indicator n is calculated according to the following formula: Where μ nm is the weight after dynamic adjustment; Step S3.3: Establish a long-term low-correlation indicator replacement mechanism for the indicators to be replaced: Calculate the correlation coefficient between the indicator data in the set to be replaced and the flood peak flow. If the indicator data in the normal processing set meets c n ≥0.8 and r n <0.3, an indicator with a higher correlation coefficient with flood events is selected from the set to be replaced and the weight is recalculated.

7. A method for determining a comprehensive dynamic index of flood early warning according to claim 1, characterized in that: The specific steps of step S4 are as follows: Step S4.1: Calculate the comprehensive dynamic index according to the following formula: Where, I m Represents the comprehensive dynamic index, μ nm is the dynamically adjusted weight, x nm is the normalized value of index n at time m; Step S4.2: The numerical range of the comprehensive dynamic index is 0-1, and the risk level is divided into three levels: low risk, medium risk, and high risk. Low risk means that the possibility of flooding is small and the basin is in a relatively safe state. Medium risk means that the possibility of flooding is gradually increasing, and attention should be paid and certain preventive measures should be taken; high risk means that the possibility of flooding is very high, and emergency warnings and response measures need to be initiated immediately.

8. A method for determining a comprehensive dynamic index of flood early warning according to claim 7, characterized in that: If 0≤I m <0.3, it is classified as low risk. If 0.3≤I m <0.6, it is classified as medium risk. If 0.6≤I m <1, it is classified as high risk.

Images