Early warning method and system for extreme rainfall geological disasters and operation method of system

Abstract

The invention discloses an extreme rainfall geological disaster early warning method and system and an operation method of the system. The early warning method comprises the following steps: acquiring topographic data, soil data, vegetation data and historical rainfall data of a monitoring area; the method comprises the following steps: determining a threshold XT corresponding to each recurrence period of extreme rainfall in a monitoring area according to historical rainfall data of the monitoring area, generalizing rainfall in the monitoring area into multiple rain patterns, determining the occurrence probability and time history distribution characteristics of each rain pattern, and forming a rain pattern database; inputting the topographic data, the soil data, the vegetation data and the rain pattern database data into a Step-Tramm model, establishing a disaster prediction model of the possibility of geological disasters caused by various rain patterns in different recurrence periods in the rain pattern database, and forming a disaster database; and when rainfall occurs in the monitoring area, rainfall data is input into the rainfall pattern database to compare the rainfall pattern and judge the recurrence period, and meanwhile, a disaster prediction model result is obtained according to the disaster database and early warning information is issued. The system can provide powerful support for disaster prevention and reduction work.

Description

Technical Field

[0001] The present invention belongs to the technical field of geological disaster early warning, and relates to early warning of geological disasters caused by extreme rainfall, and in particular to an extreme rainfall geological disaster early warning method and system and an operation method of the system. Background Art

[0002] As global climate change intensifies and extreme rainfall events become more frequent, flash floods and mudslides are becoming the leading types of natural disasters worldwide. Rainfall-induced geological disasters are characterized by rapid onset, short response times, and sudden, rapid onset, and severe localized damage. Strip projects, such as mountain highways, are particularly vulnerable to geological disasters caused by extreme rainfall. In recent years, these disasters have resulted in significant direct and indirect losses, including road damage and traffic congestion. Monitoring rainfall and developing effective early warning systems for extreme rainfall-induced geological disasters are crucial for supporting decision-makers in developing effective geological disaster prevention and mitigation plans.

[0003] Existing geological disaster early warning systems generally use buried sensors to achieve early warning of geological disaster risks in the monitoring area. The geological disaster early warning system will only sound an alarm when a geological disaster occurs in the monitoring area. It cannot provide real-time monitoring and early warning, and can only warn of the disaster point where the disaster occurs, but not the entire area with disaster risks. It has little guiding significance. Summary of the Invention

[0004] In view of the above-mentioned deficiencies in the prior art, the purpose of the present invention is to provide an extreme rainfall geological disaster early warning method and system and a method for operating the system. The present invention can provide early warning of geological disasters in an entire area caused by extreme rainfall, achieve accurate prediction and safety warning of geological disasters, and provide strong support for disaster prevention and mitigation work.

[0005] The technical solution of the present invention is achieved as follows:

[0006] An extreme rainfall geological disaster early warning method specifically comprises the following steps:

[0007] Step 1: Collect topographic data, soil data, vegetation data and historical rainfall data of the monitoring area;

[0008] Step 2: Based on the historical rainfall data of the monitoring area, determine the threshold value X corresponding to each return period of extreme rainfall in the monitoring area. T , and generalize the rainfall in the monitoring area into multiple rain types, determine the probability and time distribution characteristics of each rain type, and form a rain type database;

[0009] Step 3: Input terrain data, soil data, vegetation data and rainfall database data into the Step-Tramm model, establish a disaster prediction model for the possibility of geological disasters caused by various rainfall types in the rainfall database at different recurrence periods, and form a disaster database;

[0010] Step 4: When rainfall occurs in the monitoring area, the measured rainfall data is input into the rainfall type database to compare the rainfall type and determine the recurrence period. At the same time, the disaster prediction model results are obtained based on the disaster database and early warning information is issued.

[0011] Furthermore, the terrain data includes a terrain DEM file; the soil data includes soil type, soil cohesion, vegetation root strength, vegetation root friction angle and soil moisture content; and the vegetation data includes vegetation type and vegetation coverage area.

[0012] Furthermore, the historical rainfall data includes daily rainfall data and hourly rainfall data in the monitoring area, which should include no less than 30 years of daily rainfall data.

[0013] Furthermore, the threshold value X corresponding to each return period of rainfall in the monitoring area is determined according to the daily rainfall. T , X T Determined according to the following formula:

[0014]

[0015] Where: u is the shape parameter; α is the location parameter; k is the scale parameter; T is the recurrence period, years.

[0016] Furthermore, rain types are divided according to the hourly rainfall in the monitoring area, and the rain types include rain peak position forward and concentrated type, rain peak position forward and dispersed type, rain peak position in the middle and concentrated type, rain peak position in the middle and dispersed type, rain peak position backward and concentrated type, rain peak position backward and dispersed type and uniform type.

[0017] Furthermore, the rain types are divided according to the rain peak position coefficient γ and the central tendency index CTI, specifically:

[0018] a. The rain peak position coefficient γ is calculated according to formula (2):

[0019]

[0020] Where: t max is the time when the maximum rainfall occurs in a unit period, h; t is the total duration of the rainfall, h;

[0021] If 0<γ≤0.33, it is identified as the rain peak position biased to the front type; if 0.33<γ≤0.67, it is identified as the rain peak position biased to the middle type; if 0.67<γ≤1, it is identified as the rain peak position biased to the back type;

[0022] b. The central tendency degree CTI is calculated by formula (3):

[0023]

[0024] Where: i is the number of rainfall periods; P i is the rainfall in each period, mm; P tr is the rainfall at the peak moment, mm; P tr-i and P tr+i are the rainfall amounts before and after the peak, mm; is the total rainfall in the total time t, in mm.

[0025] If CTI≤0.50, it is judged as uniform type; if 0.50<CTI<0.67, it is judged as dispersed type; if CTI≥0.67, it is judged as concentrated type.

[0026] Furthermore, the logistic function is used to fit all rainfall events in each rain type, thereby obtaining the temporal distribution characteristics of each rain type.

[0027] The present invention also provides an extreme rainfall geological disaster early warning system, which includes a central control platform and several monitoring platforms.

[0028] All monitoring platforms are distributed at various geological disaster monitoring points within the monitoring area, and are used to collect rainfall data and soil moisture data in the monitoring area in real time.

[0029] The central control platform is integrated with the extreme rainfall geological disaster early warning method described above.

[0030] All monitoring platforms are connected to the central control platform, which facilitates the monitoring platforms to transmit the collected rainfall data and soil moisture data to the central control platform for data processing and to issue early warning information.

[0031] Furthermore, the monitoring platform includes a data acquisition unit, a numerical control unit, a power management unit, a solar power generation unit and a battery pack unit; the data acquisition unit includes a rain gauge and a soil moisture sensor, which are used to collect rainfall data and soil moisture data respectively; the power management unit is connected to the solar power generation unit and the battery pack unit, and the numerical control unit is connected to the data acquisition unit, and the solar power generation unit supplies power to the numerical control unit and the data acquisition unit and charges the battery pack unit through the power management unit.

[0032] The central control platform includes a processor and a display and memory connected to the processor; the memory stores the extreme rainfall geological disaster warning method described above, and the processor runs the calculation program in the memory to issue warning information, and the warning information is displayed on the display screen.

[0033] The numerical control unit is communicatively connected to the processor.

[0034] The aforementioned method for operating an extreme rainfall geological disaster early warning system specifically includes the following steps:

[0035] (1) The central control platform collects 24-hour weather forecast rainfall data for the monitoring area. If the weather forecast shows that there will be rainfall in the monitoring area in the next 24 hours and the predicted rainfall is greater than the warning threshold, the warning system is activated and the process goes to step (2); otherwise, the monitoring platform goes into hibernation.

[0036] (2) After the early warning system is activated, the monitoring platform collects hourly rainfall data and soil moisture data at the beginning of rainfall in the monitoring area, and transmits the rainfall data and soil moisture data to the central control platform;

[0037] (3) The central control platform reads the soil moisture data and the accumulated rainfall data every three hours, inputs the measured rainfall data into the rainfall type database to determine the rainfall type and recurrence period, then inputs the rainfall type and recurrence period into the disaster database, calls the corresponding prediction model in the disaster database, issues the first warning based on the prediction model results, and indicates the risk level and scope of the disaster that may occur in the monitoring area;

[0038] (4) The central control platform reads the accumulated 6-hour rainfall data, inputs the measured rainfall data into the rain type database to modify the rainfall type and recurrence period, then inputs the rainfall type and recurrence period into the disaster database, calls the corresponding prediction model in the disaster database, issues a second warning based on the prediction model results, and indicates the danger level and scope of the disaster that may occur in the monitoring area;

[0039] (5) The central control platform reads the accumulated rainfall data for each 9 hours. If there is a continuous rainless period of more than 3 hours at the end of the series, a temporary lifting of the danger message is issued. If there is still rainfall, the measured rainfall data is input into the rain type database to correct the rainfall type and recurrence period. The rainfall type and recurrence period are then input into the disaster database, and the corresponding prediction model in the disaster database is called. The third warning is issued based on the model prediction results, and the danger level and scope of the possible disaster in the monitoring area are indicated.

[0040] (6) The central control platform reads the accumulated rainfall data for each 12 hours. If there is a continuous rainless period of more than 3 hours at the end of the series, a temporary lifting of the danger message is issued. If there is still rainfall, the measured rainfall data is input into the rain type database to correct the rainfall type and recurrence period. The rainfall type and recurrence period are then input into the disaster database, and the corresponding prediction model in the disaster database is called. The fourth warning is issued based on the model prediction results, and the danger level and scope of the possible disaster in the monitoring area are indicated.

[0041] (7) The central control platform reads the accumulated rainfall data for each 24 hours. If there is a continuous rainless period of more than 3 hours at the end of the series, a temporary lifting of the danger message is issued. If there is still rainfall, the measured rainfall data is input into the rain type database to correct the rainfall type and recurrence period. The rainfall type and recurrence period are then input into the disaster database, and the corresponding prediction model in the disaster database is called. The fifth warning is issued based on the model prediction results, and the danger level and scope of the possible disaster in the monitoring area are indicated.

[0042] (8) The central control platform reads the accumulated rainfall data for each 24 hours in a rolling manner. If there is a continuous rainless period of more than 3 hours at the end of the series, a temporary lifting of the danger message is issued; if there is still rainfall, the measured rainfall data is input into the rain type database to correct the rainfall type and recurrence period; then the rainfall type and recurrence period are input into the disaster database, and the corresponding prediction model in the disaster database is called. The sixth warning is issued based on the model prediction results, and the danger level and scope of the possible disaster in the monitoring area are indicated;

[0043] (9) When there is no rain for more than 12 consecutive hours at the end of the rainfall series, the central control platform will announce that the danger has been completely eliminated and the monitoring platform will go into hibernation.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] 1. This invention, through real-time monitoring of rainfall within the monitoring area, combined with numerical simulation methods and precise rainfall pattern classification, can quickly and accurately issue geological disaster warning information during rainfall. This combination of real-time performance and precision effectively addresses the issues of untimely and inaccurate warnings found in traditional geological disaster warning methods, providing strong support for disaster prevention and mitigation efforts. Furthermore, the invention can provide targeted warnings based on rainfall patterns, making the warning information more instructive and helpful for decision makers in developing more scientific and rational disaster prevention and mitigation plans.

[0046] 2. The present invention has a wider coverage area, enabling early warnings to be issued across the entire monitoring area. This not only improves the comprehensiveness of early warnings but also enhances their efficiency through numerical simulation methods. By intelligently analyzing real-time monitoring data and historical data, early warning information can be dynamically adjusted to ensure the accuracy and timeliness of early warnings. Furthermore, the present invention is widely applicable to various areas with the potential for geological disasters caused by rainfall, particularly large-scale infrastructure projects such as mountain highways. It can effectively ensure traffic safety and reduce losses caused by geological disasters.

[0047] 3. This invention models and analyzes a vast amount of terrain, soil, vegetation data, and historical rainfall information, providing a reliable basis for early warning decision-making through a scientific, data-driven approach. It not only intelligently determines rainfall patterns and recurrence periods, but also dynamically adjusts early warning information based on real-time rainfall data, ensuring the scientific and rationality of early warnings. Furthermore, this invention boasts a high degree of stability and reliability, capable of stable operation in harsh environments and providing continuous and reliable early warning services. Through data-driven and scientific decision-making, it can provide a powerful guarantee for effectively preventing and mitigating losses caused by geological disasters. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 - Flowchart of the early warning method of the present invention.

[0049] Figure 2 -Statistical results of rain patterns in the examples of the present invention.

[0050] Figure 3 -Model diagram of the early warning system of the present invention.

[0051] Figure 4 -Structural block diagram of the early warning system of the present invention.

[0052] Figure 5 - Flowchart of the method for operating the early warning system of the present invention. DETAILED DESCRIPTION

[0053] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0054] An extreme rainfall geological disaster early warning method specifically comprises the following steps:

[0055] Step 1: Collect terrain data, soil data, vegetation data and historical rainfall data of the monitoring area.

[0056] Here, the terrain data includes terrain DEM files, and the DEM elevation data is used to determine the terrain of the monitoring area, such as elevation, altitude, etc. The soil data includes soil type, soil cohesion, vegetation root strength, vegetation root friction angle and soil moisture content. The soil data is mainly used to determine the soil classification of the monitoring area, such as clay, loam, etc., and is used to calculate soil depth, obtain soil internal friction angle, determine vegetation root strength, soil cohesion, etc. The vegetation data includes vegetation types and vegetation coverage area. The historical rainfall data includes daily rainfall data and hourly rainfall data in the monitoring area, which should include no less than 30 years of daily rainfall data for subsequent determination of the threshold values ​​corresponding to each recurrence period of rainfall in the monitoring area; it should include hourly rainfall data for as long as possible for subsequent classification of rainfall types.

[0057] Step 2: Based on the historical rainfall data of the monitoring area, the generalized extreme value distribution function (GEV) is used to determine the threshold value X corresponding to each return period of extreme rainfall in the monitoring area. T , and generalize the rainfall in the monitoring area into multiple rain types, determine the probability and time distribution characteristics of each rain type, and form a rain type database.

[0058] The probability density function of the extreme value distribution (GEV) is:

[0059]

[0060] Among them, u, α, and k are the shape parameter, location parameter, and scale parameter of the generalized extreme value distribution, respectively.

[0061] The distribution function of the generalized extreme value distribution (GEV) is:

[0062]

[0063] Among them, when k=0, it is extreme value type I, that is, Gumbel distribution; when k<0, it is extreme value type II, that is, Fréchet distribution; when k>0, it is extreme value type III, that is, Weibull distribution.

[0064] The inverse function form of the generalized extreme value distribution (GEV) is:

[0065]

[0066] At present, the parameter estimation in the GEV method mainly includes the ordinary moment method, the ordinary probability weighted moment method and the high-order probability weighted moment method. However, the ordinary moment method and the ordinary probability weighted moment method have certain limitations when simulating the flood values ​​of the high tail part of the flood sequence, and cannot fully reflect the characteristics of extreme rainfall, thereby affecting the accuracy of extreme rainfall monitoring and early warning. Compared with the ordinary moment method and the ordinary probability weighted moment method, the high-order probability weighted moment method has higher accuracy and can more accurately reflect the characteristics of extreme rainfall, thereby significantly improving the accuracy of extreme rainfall monitoring and early warning. Therefore, the present invention adopts the high-order probability weighted moment method to estimate the shape parameter u, location parameter α and scale parameter k in the GEV distribution.

[0067] Therefore, let the distribution function of the random variable x be F(x)=P(X≤x), and its probability weight moment expression is:

[0068]

[0069] Among them, x(F) is the inverse function of the distribution function, p, r, s are real numbers, when p = 1, s = 0, the probability weight moment β r for:

[0070]

[0071] Assume that a sample of length n (maximum daily rainfall in the monitoring area) is given, which obeys the F distribution, and the sample in the order statistic satisfies x (1) ≤x (2) ≤…≤x (n) , its probability weight moment β r An unbiased estimate of b r It can be calculated by the following formula:

[0072]

[0073] in:

[0074] x (0) represents a specific threshold, b r The order of is determined by the number of parameters of the distribution function. If there are p parameters of the distribution function, then p equations are solved simultaneously, r = 0, 1, ..., p-1. The way to give a higher weight to the random variable x is to increase the order r. When r increases, F r will increase at an accelerated rate, that is, the higher-order probability weight moments are more dependent on larger values ​​of the random variable.

[0075] When k≠0, the probability weight moment of the GEV distribution is:

[0076]

[0077] In the above formula, let u=-lnF, we can deduce:

[0078]

[0079] In the above formula, let x=(r+1)u, we can deduce:

[0080]

[0081] in, is the gamma function.

[0082] When k = 0, the probability weight moment of the GEV distribution is:

[0083] (r+1)β r =u+a[ε+ln(r+1)] (10)

[0084] Among them, ε = 0.5772156649… is the Euler constant.

[0085] From the probability weight moment expression, we can see that in order to solve the three parameters u, a, and k in the GEV distribution function, we need to list three equations to solve them.

[0086] For k≠0, let the order of the probability weight moment r=η, η+1, η+2, and substitute them into formula (9) respectively, we have:

[0087]

[0088] Formula (12) - Formula (11) can be obtained:

[0089]

[0090] Formula (13) - Formula (11) can be obtained:

[0091]

[0092] Dividing formula (14) by formula (15) yields:

[0093]

[0094] For k = 0, let r = η, η + 1, η + 2, and substitute them into formula (10) respectively, and we have:

[0095] (η+1)β η =u+a[ε+ln(η+1)] (17)

[0096] (η+2)β η+1 =u+a[ε+ln(η+2)] (18)

[0097] (η+3)β η+2=u+a[ε+ln(η+3)] (19)

[0098] Formula (18) - Formula (17) can be obtained:

[0099] (η+2)β η+1 -(η+1)β η =a[ln(η+2)-ln(η+1)] (20)

[0100] Formula (19) - Formula (17) can be obtained:

[0101] (η+3)β η+2 -(η+1)β η =a[ln(η+3)-ln(η+1)] (21)

[0102] Dividing formula (20) by formula (21) yields:

[0103]

[0104] In the above formula, the sample probability weight moment β of the GEV distribution is η , β η+1 , β η+2 Available probability weight moment β r The calculated unbiased estimator b η , b η+1 , b η+2 Since there is no direct expression for k, the conventional iterative method is very computationally intensive, so a computer numerical algorithm can be used to solve it.

[0105] Subtract both sides of formula (16) have:

[0106]

[0107] The right side of formula (23) is

[0108]

[0109] Furthermore, given the value range of k is -0.5≤k≤0.5, let η=0,1,2,3,4, that is, r=0,1,2 (that is, ordinary probability weight moment), r=1,2,3, r=2,3,4, r=3,4,5, r=4,5,6 respectively, calculate the z value corresponding to each group of values ​​respectively, and fit the curve of the quadratic function according to formula (25) to obtain the coefficients a0, a1, a2 in formula (25) with different orders.

[0110] k=a0+a1z+a2z 2 (25)

[0111] When η=0,1,2,3,4 (i.e. r=0,1,2,r=1,2,3,r=2,3,4,r=3,4,5,r=4,5,6), the left side of equation (23) is calculated as follows:

[0112]

[0113] Where, β r Use its unbiased estimator b r (It can be calculated using formula (6) instead), combined with the coefficient values ​​of the corresponding order, substitute into formula (25) to obtain the estimated value of parameter k

[0114] The estimated value of parameter k Substituting into equations (11) and (12), we can calculate the estimated value of the scale parameter α of the GEV distribution: and the estimated value of the location parameter u The parameter expressions are:

[0115]

[0116] Furthermore, the return period T and F(X T )as follows:

[0117]

[0118] Where T is the return period.

[0119] Substituting Equation (29) into Equation (3), the return period T and the inverse function of the generalized extreme value distribution (GEV) X T The relationship is as follows:

[0120]

[0121] Substituting the calculated shape parameter u, location parameter α, and scale parameter k into formula (30), we can estimate the corresponding threshold value X for different recurrence periods T. T .

[0122] According to the maximum and second largest 1-day rainfall series in a certain monitoring area, the extreme rainfall value of the monitoring area is calculated to be 166.02 mm. The GEV formula is used to estimate the different return periods. The threshold values ​​X corresponding to different return periods are T See the table below:

[0123] Table 1. The corresponding threshold value X of the extreme rainfall return period in the monitoring area T

[0124]

[0125] In addition, the rain types are divided according to the hourly rainfall in the monitoring area. The rain types include the forward and concentrated type, the forward and dispersed type, the central and concentrated type, the central and dispersed type, the backward and concentrated type, the backward and dispersed type, and the uniform type. The rain types are divided specifically according to the rain peak position coefficient γ and the degree of central tendency CTI:

[0126] a. The rain peak position coefficient γ is calculated according to formula (31):

[0127]

[0128] Where: t max is the time when the maximum rainfall occurs in a unit period, h; t is the total duration of the rainfall, h;

[0129] If 0<γ≤0.33, it is identified as the rain peak position biased to the front type; if 0.33<γ≤0.67, it is identified as the rain peak position centered type; if 0.67<γ≤1, it is identified as the rain peak position biased to the back type.

[0130] b. The central tendency index (CTI) is calculated by formula (32):

[0131]

[0132] Where: i is the number of rainfall periods; P i is the rainfall in each period, mm; P tr is the rainfall at the peak moment, mm; P tr-i and P tr+i are the rainfall amounts before and after the peak, mm; is the total rainfall in the total time t, in mm. For example, if a rain lasts for 120 minutes, the peak time occurs between 60 and 80 minutes, and the rainfall in the 60-80 minutes is P tr ,and For 40-60 minutes of rainfall, The rainfall lasts for 80-100 minutes.

[0133] If CTI ≤ 0.50, the rainfall is classified as uniform; if 0.50 < CTI < 0.67, the rainfall is classified as dispersed; and if CTI ≥ 0.67, the rainfall is classified as concentrated. Because the peak rainfall amount in uniform rainfall is not prominent and the rainfall amount at each moment in the entire rainfall process is relatively close, the peak rainfall position coefficient is not used to distinguish the rainfall peak location.

[0134] In this embodiment, 24 long-duration rainfall processes lasting more than 24 hours and with rainfall reaching the rainstorm level are extracted from the rainfall data of the monitoring area as the basic data for rain type classification. The calculated extreme rainfall characteristic parameters of the monitoring area are shown in Table 2.

[0135] Table 2 Calculation results of extreme rainfall characteristic parameters in the monitoring area

[0136]

[0137] According to the rain peak position coefficient γ and the degree of central tendency CTI, the 50 extreme rainfall events in the monitoring area are divided into 7 types. The number of events of each rainfall type is as follows: Figure 2 As shown, within the monitoring area, there were 13 rainfall events with an early and concentrated peak, 5 with an early and dispersed peak, 10 with a central and concentrated peak, 5 with a central and dispersed peak, 9 with a late and concentrated peak, 4 with a late and dispersed peak, and 4 with a uniform peak. The rainfall corresponding to each extreme rainfall return period calculated in Table 1 is distributed according to the distribution ratio of different rainfall types to obtain the rainfall time series distribution of different rainfall types under each return period. The time series distribution of one rainfall type under different return periods is shown in Table 3.

[0138] Table 3 Time distribution of concentrated rainfall with peak ahead under different return periods Unit (mm)

[0139]

[0140] Step 3: Input terrain data, soil data, vegetation data and rainfall database data into the Step-Tramm model, establish a disaster prediction model for the possibility of geological disasters caused by various rainfall types in the rainfall database at different recurrence periods, and form a disaster database;

[0141] In this example, DEM elevation data is input to construct a three-dimensional terrain surface, generating slope, aspect, and contour lines. Soil data is input to determine that the soil type in the monitoring area is primarily silt clay. Average values ​​and internal friction angles, vegetation root strength, and soil cohesion are calculated for different soil geologies. Vegetation data is input to determine the forest coverage in the monitoring area, thereby quantifying the effects of vegetation on rainfall interception and surface runoff mitigation.

[0142] The topographic data, soil data, vegetation data and various rainfall type data under seven different return periods of the monitoring area were input into the Step-Tramm model to simulate the debris flow discharge of various rainfall types under different return periods to form a disaster database. The analysis results based on rainfall types are shown in Table 4.

[0143] Table 4. Predicted debris flow discharge under different return periods for various types of rainfall in the monitoring area (unit: m 3 )

[0144]

[0145] Step 4: When rainfall occurs in the monitoring area, the measured rainfall data is input into the rainfall type database to compare the rainfall type and determine the recurrence period. At the same time, the disaster prediction model results are obtained based on the disaster database and early warning information is issued.

[0146] The measured rainfall data of the monitoring area is input into the rainfall type database, and it is judged that the rainfall type of the monitoring area is the peak-front concentrated type, and a 500-year rainfall will occur. By comparing it with the disaster database, the possible disaster model is determined as follows: Figure 3 shown.

[0147] The present invention also provides an extreme rainfall geological disaster early warning system, see Figure 4 , including a central control platform and several monitoring platforms.

[0148] All monitoring platforms are distributed at various geological disaster monitoring points within the monitoring area, and are used to collect rainfall data and soil moisture data in the monitoring area in real time.

[0149] The central control platform is integrated with the extreme rainfall geological disaster early warning method described above.

[0150] All monitoring platforms are connected to the central control platform, which facilitates the monitoring platforms to transmit the collected rainfall data and soil moisture data to the central control platform for data processing and to issue early warning information.

[0151] During specific implementation, the monitoring platform includes a data acquisition unit, a numerical control unit, a power management unit, a solar power generation unit and a battery pack unit; the data acquisition unit includes a rain gauge and a soil moisture sensor, which are used to collect rainfall data and soil moisture data respectively; the power management unit is connected to the solar power generation unit and the battery pack unit, and the numerical control unit is connected to the data acquisition unit. The solar power generation unit supplies power to the numerical control unit and the data acquisition unit and charges the battery pack unit through the power management unit.

[0152] The central control platform includes a processor and a display and memory connected to the processor; the memory stores the extreme rainfall geological disaster warning method described above, and the processor runs the calculation program in the memory to issue warning information, and the warning information is displayed on the display screen.

[0153] The numerical control unit is connected to the processor via a wireless communication unit to transmit data from the monitoring platform to the central control platform.

[0154] The present invention also provides a method for operating an extreme rainfall geological disaster early warning system, which specifically includes the following steps:

[0155] (1) The central control platform collects 24-hour weather forecast rainfall data for the monitoring area. If the weather forecast shows that there will be rainfall in the monitoring area in the next 24 hours and the predicted rainfall is greater than the warning threshold, the warning system is activated and the process goes to step (2); otherwise, the monitoring platform goes into hibernation.

[0156] (2) After the early warning system is activated, the monitoring platform collects hourly rainfall data and soil moisture data at the beginning of rainfall in the monitoring area, and transmits the rainfall data and soil moisture data to the central control platform. Under extreme rainfall conditions, the soil moisture content will quickly reach saturation, so it is only necessary to monitor the soil moisture content at the beginning of rainfall. The soil moisture content during the rainfall process is considered to be saturated.

[0157] (3) The central control platform reads the soil moisture data and the accumulated rainfall data every three hours, inputs the measured rainfall data into the rainfall type database to determine the rainfall type and recurrence period, then inputs the rainfall type and recurrence period into the disaster database, calls the corresponding prediction model in the disaster database, issues the first warning based on the prediction model results, and indicates the risk level and scope of the disaster that may occur in the monitoring area;

[0158] (4) The central control platform reads the accumulated 6-hour rainfall data, inputs the measured rainfall data into the rain type database to modify the rainfall type and recurrence period, then inputs the rainfall type and recurrence period into the disaster database, calls the corresponding prediction model in the disaster database, issues a second warning based on the prediction model results, and indicates the danger level and scope of the disaster that may occur in the monitoring area;

[0159] (5) The central control platform reads the accumulated rainfall data for each 9 hours. If there is a continuous rainless period of more than 3 hours at the end of the series, a temporary lifting of the danger message is issued. If there is still rainfall, the measured rainfall data is input into the rain type database to correct the rainfall type and recurrence period. The rainfall type and recurrence period are then input into the disaster database, and the corresponding prediction model in the disaster database is called. The third warning is issued based on the model prediction results, and the danger level and scope of the possible disaster in the monitoring area are indicated.

[0160] (6) The central control platform reads the accumulated rainfall data for each 12 hours. If there is a continuous rainless period of more than 3 hours at the end of the series, a temporary lifting of the danger message is issued. If there is still rainfall, the measured rainfall data is input into the rain type database to correct the rainfall type and recurrence period. The rainfall type and recurrence period are then input into the disaster database, and the corresponding prediction model in the disaster database is called. The fourth warning is issued based on the model prediction results, and the danger level and scope of the possible disaster in the monitoring area are indicated.

[0161] (7) The central control platform reads the accumulated rainfall data for each 24 hours. If there is a continuous rainless period of more than 3 hours at the end of the series, a temporary lifting of the danger message is issued. If there is still rainfall, the measured rainfall data is input into the rain type database to correct the rainfall type and recurrence period. The rainfall type and recurrence period are then input into the disaster database, and the corresponding prediction model in the disaster database is called. The fifth warning is issued based on the model prediction results, and the danger level and scope of the possible disaster in the monitoring area are indicated.

[0162] (8) The central control platform reads the accumulated rainfall data for each 24 hours in a rolling manner. If there is a continuous rainless period of more than 3 hours at the end of the series, a temporary lifting of the danger message is issued; if there is still rainfall, the measured rainfall data is input into the rain type database to correct the rainfall type and recurrence period; then the rainfall type and recurrence period are input into the disaster database, and the corresponding prediction model in the disaster database is called. The sixth warning is issued based on the model prediction results, and the danger level and scope of the possible disaster in the monitoring area are indicated;

[0163] (9) When there is no rain for more than 12 consecutive hours at the end of the rainfall series, the central control platform will announce that the danger has been completely eliminated and the monitoring platform will go into hibernation.

[0164] Finally, it should be noted that the above embodiments of the present invention are merely examples for illustrating the present invention and are not intended to limit the embodiments of the present invention. Those skilled in the art will appreciate that other variations and modifications can be made based on the above description. It is not possible to enumerate all embodiments here. Any obvious variations or modifications arising from the technical solution of the present invention remain within the scope of protection of the present invention.

Claims

1. A method for early warning of extreme rainfall geological disasters, characterized in that: The specific steps include: Step 1: Collect topographic data, soil data, vegetation data and historical rainfall data of the monitoring area; Step 2: Based on the historical rainfall data of the monitoring area, determine the threshold value X corresponding to each return period of extreme rainfall in the monitoring area. T , and generalize the rainfall in the monitoring area into multiple rain types, determine the probability and time distribution characteristics of each rain type, and form a rain type database; Step 3: Input terrain data, soil data, vegetation data and rainfall database data into the Step-Tramm model, establish a disaster prediction model for the possibility of geological disasters caused by various rainfall types in the rainfall database at different recurrence periods, and form a disaster database; Step 4: When rainfall occurs in the monitoring area, the measured rainfall data is input into the rainfall type database to compare the rainfall type and determine the recurrence period. At the same time, the disaster prediction model results are obtained based on the disaster database and early warning information is issued.

2. The extreme rainfall geological disaster early warning method according to claim 1, characterized in that: The terrain data includes a terrain DEM file; the soil data includes soil type, soil cohesion, vegetation root strength, vegetation root friction angle and soil moisture content; and the vegetation data includes vegetation type and vegetation coverage area.

3. The extreme rainfall geological disaster early warning method according to claim 1, characterized in that: The historical rainfall data include daily rainfall data and hourly rainfall data in the monitoring area, which should include no less than 30 years of daily rainfall data.

4. The extreme rainfall geological disaster early warning method according to claim 3, characterized in that: Determine the threshold value X corresponding to each return period of rainfall in the monitoring area based on the daily rainfall T , X T Determined according to the following formula: Where: u is the shape parameter; α is the location parameter; k is the scale parameter; T is the recurrence period, years.

5. The extreme rainfall geological disaster early warning method according to claim 3, characterized in that: The rain types are divided according to the hourly rainfall in the monitoring area, and the rain types include the rain peak position is forward and concentrated type, the rain peak position is forward and dispersed type, the rain peak position is in the middle and concentrated type, the rain peak position is in the middle and dispersed type, the rain peak position is backward and concentrated type, the rain peak position is backward and dispersed type and uniform type.

6. The extreme rainfall geological disaster early warning method according to claim 5, characterized in that: Rainfall types are classified according to the rain peak position coefficient γ and the central tendency index CTI, specifically: a. The rain peak position coefficient γ is calculated according to formula (2): Where: t max is the time when the maximum rainfall occurs in a unit period, h; t is the total duration of the rainfall, h; If 0<γ≤0.33, it is identified as the rain peak position biased to the front type; if 0.33<γ≤0.67, it is identified as the rain peak position biased to the middle type; if 0.67<γ≤1, it is identified as the rain peak position biased to the back type; b. The central tendency degree CTI is calculated by formula (3): Where: i is the number of periods of rainfall; P i is the rainfall in each period, mm; P tr is the rainfall at the peak moment, mm; P tr-i and P tr+i are the rainfall amounts before and after the peak, mm; is the total rainfall in the total duration t, mm; If CTI≤0.50, it is judged as uniform type; if 0.50<CTI<0.67, it is judged as dispersed type; if CTI≥0.67, it is judged as concentrated type.

7. The extreme rainfall geological disaster early warning method according to claim 1, characterized in that: The logistic function is used to fit all rainfall events in each rainfall type to obtain the temporal distribution characteristics of each rainfall type.

8. An extreme rainfall geological disaster early warning system, characterized in that: Including a central control platform and several monitoring platforms; All monitoring platforms are distributed at various geological disaster monitoring points within the monitoring area, and are used to collect real-time rainfall data and soil moisture data in the monitoring area; The central control platform is integrated with an extreme rainfall geological disaster early warning method according to any one of claims 1 to 7; All monitoring platforms are connected to the central control platform, which facilitates the monitoring platforms to transmit the collected rainfall data and soil moisture data to the central control platform for data processing and to issue early warning information.

9. The extreme rainfall geological disaster early warning system according to claim 8, characterized in that: The monitoring platform includes a data acquisition unit, a numerical control unit, a power management unit, a solar power generation unit, and a battery unit; the data acquisition unit includes a rain gauge and a soil moisture sensor, which are used to collect rainfall data and soil moisture data respectively; the power management unit is connected to the solar power generation unit and the battery unit, and the numerical control unit is connected to the data acquisition unit. The solar power generation unit supplies power to the numerical control unit and the data acquisition unit and charges the battery unit through the power management unit; The central control platform includes a processor, a display connected to the processor, and a memory; the memory stores an extreme rainfall geological disaster early warning method according to any one of claims 1 to 7; the processor runs the computing program in the memory to issue early warning information, and the early warning information is displayed on the display screen; The numerical control unit is communicatively connected to the processor.

10. The method for operating an extreme rainfall geological disaster early warning system according to claim 8 or 9, characterized in that: The specific steps include: (1) The central control platform collects 24-hour weather forecast rainfall data for the monitoring area. If the weather forecast shows that there will be rainfall in the monitoring area in the next 24 hours and the predicted rainfall is greater than the warning threshold, the warning system is activated and the process goes to step (2); otherwise, the monitoring platform goes into hibernation. (2) After the early warning system is activated, the monitoring platform collects hourly rainfall data and soil moisture data at the beginning of rainfall in the monitoring area, and transmits the rainfall data and soil moisture data to the central control platform; (3) The central control platform reads the soil moisture data and the accumulated rainfall data every three hours, inputs the measured rainfall data into the rainfall type database to determine the rainfall type and recurrence period, then inputs the rainfall type and recurrence period into the disaster database, calls the corresponding prediction model in the disaster database, issues the first warning based on the prediction model results, and indicates the risk level and scope of the disaster that may occur in the monitoring area; (4) The central control platform reads the accumulated 6-hour rainfall data, inputs the measured rainfall data into the rain type database to modify the rainfall type and recurrence period, then inputs the rainfall type and recurrence period into the disaster database, calls the corresponding prediction model in the disaster database, issues a second warning based on the prediction model results, and indicates the danger level and scope of the disaster that may occur in the monitoring area; (5) The central control platform reads the accumulated rainfall data for each 9 hours. If there is a continuous rainless period of more than 3 hours at the end of the series, a temporary lifting of the danger message is issued. If there is still rainfall, the measured rainfall data is input into the rain type database to correct the rainfall type and recurrence period. The rainfall type and recurrence period are then input into the disaster database, and the corresponding prediction model in the disaster database is called. The third warning is issued based on the model prediction results, and the danger level and scope of the possible disaster in the monitoring area are indicated. (6) The central control platform reads the accumulated rainfall data for each 12 hours. If there is a continuous rainless period of more than 3 hours at the end of the series, a temporary lifting of the danger message is issued. If there is still rainfall, the measured rainfall data is input into the rain type database to correct the rainfall type and recurrence period. The rainfall type and recurrence period are then input into the disaster database, and the corresponding prediction model in the disaster database is called. The fourth warning is issued based on the model prediction results, and the danger level and scope of the possible disaster in the monitoring area are indicated. (7) The central control platform reads the accumulated rainfall data for each 24 hours. If there is a continuous rainless period of more than 3 hours at the end of the series, a temporary lifting of the danger message is issued. If there is still rainfall, the measured rainfall data is input into the rain type database to correct the rainfall type and recurrence period. The rainfall type and recurrence period are then input into the disaster database, and the corresponding prediction model in the disaster database is called. The fifth warning is issued based on the model prediction results, and the danger level and scope of the possible disaster in the monitoring area are indicated. (8) The central control platform reads the accumulated rainfall data for each 24 hours in a rolling manner. If there is a continuous rainless period of more than 3 hours at the end of the series, a temporary lifting of the danger message is issued; if there is still rainfall, the measured rainfall data is input into the rain type database to correct the rainfall type and recurrence period; then the rainfall type and recurrence period are input into the disaster database, and the corresponding prediction model in the disaster database is called. The sixth warning is issued based on the model prediction results, and the danger level and scope of the possible disaster in the monitoring area are indicated; (9) When there is no rain for more than 12 consecutive hours at the end of the rainfall series, the central control platform will announce that the danger has been completely eliminated and the monitoring platform will go into hibernation.

Images