Mountain torrent disaster early warning short message accurate calling and signing feedback system

Abstract

The invention discloses a mountain torrent disaster early-warning short message accurate calling and signing feedback system, and the system comprises an early-warning information generation and release module which is used for transmitting early-warning information to each receiving node; the receiving node is used for receiving the early warning information; the comparison and correction module is used for comparing the difference between the index to be fed back and the feedback index to obtain the instant deviation of the receiving node; superposing the obtained instant deviation to a previous accumulated deviation value to form a deviation accumulation amount; dynamically adjusting the weight of the feedback index according to the deviation cumulant and the delay deviation; and calculating a global deviation sum according to the deviation cumulant, the delay deviation and the weight of the feedback index, judging whether the global deviation sum is lower than a set threshold value or not, and deciding to close a feedback loop or initiate information retransmission. According to the invention, whether the person in charge understands the early warning message can be accurately confirmed, and efficient response and accurate execution in an emergency scene of disaster early warning are ensured.

Description

Technical Field

[0001] The present invention relates to the field of disaster early warning technology, and in particular to a flash flood disaster early warning text message accurate response and receipt feedback system. Background Art

[0002] Flash torrents refer to the rapid floods that occur in mountain streams due to heavy rains and other reasons. The geological disasters caused by flash torrents include mudslides, landslides, mountain collapses, etc. As a natural disaster caused by heavy rainfall, it is characterized by suddenness, great destructive power, and a wide range of impact. It poses a serious threat to the safety of life and property of the people. Therefore, establishing and improving the flash torrent disaster safety early warning system has become an important means to prevent and reduce disaster losses.

[0003] In view of this, the Henan Provincial Department of Water Resources has developed and built a flash flood disaster warning SMS call and response system. The system can track and grasp the release and receipt of warning SMS in real time. For those responsible for not signing for the warning information, the staff can check their information one by one, and promptly notify them again by phone or other means to remind them to take corresponding actions to ensure that the warning "can be sent, received, and called". In the flash flood disaster warning SMS call and response system, the issuance of each warning message (call and response) will trigger a waiting for confirmation state. Only when the recipient confirms that the information has been received and understood (signed), this state will be closed, forming a closed loop. However, the feedback information can only indicate that the person in charge has received the warning message, but it cannot accurately confirm whether the person in charge understands the warning message. Summary of the invention

[0004] In order to solve the problem of being unable to accurately confirm whether the person in charge understands the warning message, the present invention provides a flash flood disaster warning text message accurate response and receipt feedback system.

[0005] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0006] A flash flood disaster warning SMS accurate response and receipt feedback system, comprising:

[0007] The warning information generation and release module is used to send warning information to each receiving node R i Send warning information M;

[0008] Receiving Node R i , used to receive the warning information M, and perform local confirmation according to the indicators to be fed back contained in the warning information M, and send feedback information F to the warning information generation and release module i , where the feedback information F i Contains the receiving node R i Feedback indicators;

[0009] The comparison and correction module compares the difference between the to-be-feedback indicator and the feedback indicator to obtain the receiving node R i The instant deviation B i ; The instant deviation B i Added to the previous cumulative deviation value E i,prev , forming the cumulative deviation E i ; According to the cumulative deviation E i and delay deviation T i Dynamically adjust the weight of feedback indicators ω i ; Through the deviation accumulation E i and delay deviation T i and the weight of the feedback indicator ω i Calculate the global deviation sum D i , judgement |D i |Whether it is lower than the set threshold, it decides to close the feedback loop or initiate information retransmission.

[0010] Further: in the comparison and correction module, the difference between the to-be-feedback indicator and the feedback indicator is compared to obtain the receiving node R i The instant deviation B i include:

[0011] For the indicator j in the feedback indicator, it is recorded as M j ;

[0012] For the indicator j in the feedback indicator, it is recorded as F i,j ;

[0013] Define deviation measurement methods according to the type of indicators, including numerical indicators, categorical indicators, and regional indicators;

[0014] For numerical indicators, the instantaneous deviation is calculated by the difference B i,j =F i,j -M j ;

[0015] For the classification indicators, the instant deviation is calculated by discrete distance, and the weights B between different classifications are assigned. i,j = distance(F i,j ,M j );

[0016] For regional indicators, the instant deviation is calculated by matching accuracy, which is defined as follows:

[0017]

[0018] Logarithmic instant deviation B i,j , with the maximum possible deviation B i,j,max Normalization

[0019] Bisection type instant deviation B i,j , by category total C max Normalization

[0020] Regional instant deviation B i,j ,make

[0021] Construct the instant deviation matrix B:

[0022]

[0023] Where: N is the number of receiving nodes, and M is the number of indicators of warning information.

[0024] Further: In the comparison and correction module, the instant deviation B i Added to the previous cumulative deviation value E i,prev , forming the cumulative deviation E i include:

[0025] For each receiving node R i Initialize the cumulative deviation value with index j;

[0026] Get the receiving node R from the instant deviation matrix i The current instantaneous deviation from indicator j

[0027] Calculate the weighted cumulative deviation:

[0028]

[0029] Where: E i,j Represents the receiving node R i The current cumulative deviation of indicator j; E i,prev Represents the receiving node R i The previous cumulative deviation of indicator j; Indicates that the node receives the node R i The current instant deviation of indicator j; α represents the attenuation factor of historical deviation, 0≤α≤1;

[0030] If α = 0, the cumulative deviation depends entirely on the latest deviation;

[0031] If α = 1, the cumulative deviation depends entirely on historical data;

[0032] For the receiving node R i , summarizing the accumulated deviations of all indicators to provide a basis for weight adjustment and global deviation calculation:

[0033]

[0034] Where: E i is the receiving node R i The total cumulative deviation; M is the number of indicators included in the early warning information.

[0035] Further: In the comparison and correction module, the cumulative amount of deviation E i and delay deviation T i Dynamically adjust the weight of feedback indicators ω i include:

[0036] For each receiving node R i Assign initial weight Indicates that all receiving nodes are initially equally important; calculate the deviation factor: F E,i =exp(-β·E i )

[0037] When the deviation E i When it is larger, the deviation factor F E,i reduce;

[0038] Calculate the delay factor: F T,i =exp(-γ·T i )

[0039] When the delay T i When the delay factor F is larger, T,i reduce;

[0040] Calculate each receiving node R based on the dynamic adjustment factor i The unnormalized weights of :

[0041]

[0042] Represents the receiving node R i The unnormalized weights of

[0043] The unnormalized weights of all nodes Normalized to a weight distribution that sums to 1;

[0044]

[0045] make sure:

[0046] The newly calculated weight Replace old weights

[0047] Further: In the comparison and correction module, the accumulated deviation E i and delay deviation T i and the weight of the feedback indicator ω iCalculate the global deviation sum D i , judgement |D i |Whether the value is lower than the set threshold, the decision to close the feedback loop or initiate information retransmission includes:

[0048] For each receiving node R i , calculate the total global deviation:

[0049]

[0050] Among them, ε is the weight coefficient of feedback delay;

[0051] The global deviation sum of the nodes D i With the threshold value D threshold For comparison:

[0052] If |D i |≤D threshold , receiving node R i The feedback can be closed loop;

[0053] If |D i |≥D threshold , receiving node R i The feedback does not meet the requirements and the information needs to be resent.

[0054] Compared with the prior art, the present invention has the following technical advances:

[0055] The present invention realizes accurate verification and response to early warning information through a closed-loop feedback mechanism, thereby effectively ensuring the accurate communication and execution of early warning information at the receiving point. By comparing the difference between the feedback of the receiving point and the original early warning index, and assigning dynamic weights to different indicators, the feedback quality of each receiving point can be accurately evaluated according to the calculation result of the global deviation sum, so as to determine whether it is necessary to resend the information separately. This precise feedback mechanism not only reduces the risk of response errors caused by mistransmission or deviation of information, but also significantly improves the reliability and efficiency of early warning information transmission. At the same time, the triggering conditions for retransmitting information are limited to receiving points whose deviations exceed the threshold, avoiding unnecessary retransmission of information to all receiving points, reducing system load, enhancing the pertinence and flexibility of resource utilization, and ensuring efficient response and accurate execution in emergency scenarios of disaster warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0057] In the attached picture:

[0058] Figure 1 is a system structure diagram of the present invention;

[0059] Figure 2 These are steps implemented by the comparison and correction module of the present invention. DETAILED DESCRIPTION

[0060] The following specific embodiments may be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments. The embodiments of the present invention will be described below in conjunction with the accompanying drawings.

[0061] like Figure 1 As shown, the present invention discloses a flash flood disaster warning SMS accurate response and receipt feedback system, including: a warning information generation and release module, which is used to send a warning message to each receiving node R i Send warning information M;

[0062] Receiving Node R i , used to receive the warning information M, and perform local confirmation according to the indicators to be fed back contained in the warning information M, and send feedback information F to the warning information generation and release module i , where the feedback information F i Contains the receiving node R i Feedback indicators;

[0063] Comparison and correction modules, such as Figure 2 As shown, the implementation steps are as follows: Step 1: Compare the difference between the to-be-feedback indicator and the feedback indicator to obtain the receiving node R i The instant deviation B i ; Step 2: Obtain the instant deviation B i Added to the previous cumulative deviation value F i,prev , forming the cumulative deviation E i ; Step 3: According to the cumulative deviation E i and delay deviation T i Dynamically adjust the weight of the feedback indicator; Step 4: Through the deviation accumulation E i and delay deviation T i The total global deviation D is calculated by the weight of the feedback index i , judgement |D i |Whether it is lower than the set threshold, it decides to close the feedback loop or initiate information retransmission.

[0064] The warning information generation and release module sends the same warning information M to multiple receiving nodes. This warning information M contains several indicators, such as: indicator 1 warning level (blue warning, yellow warning, orange warning, red warning), indicator 2 precipitation, indicator 3 disaster area, indicator 4 warning response (increase vigilance, prepare to transfer, transfer immediately), etc. The receiving point Ri must feedback the same indicators to the warning information generation and release module, so that the receiving point can form a closed feedback loop.

[0065] For example, the warning information generation and release module sends a warning message M to the receiving points R1 and R2. This warning message M contains three indicators: indicator 1 warning level: orange warning; indicator 2 precipitation: ≥70mm per hour; indicator 3 disaster area: area A; indicator 4 warning response: increase alert. Assuming that all the indicators fed back by the receiving point R1 are the same as those sent by the warning information generation and release module, it means that the receiving point R1 has accurately fed back, and the receiving point R1 forms a closed feedback loop.

[0066] However, the indicators fed back by the receiving point R2 may have one or more errors, such as indicator 1 warning level feedback error, indicator 2 precipitation feedback error, indicator 3 disaster area error, indicator 4 warning response error. Therefore, it is necessary to assign different weights to the various feedback indicators of the receiving point R2, and finally obtain the global deviation sum D through calculation, and judge whether D is lower than the set threshold, so as to determine whether the receiving point R2 can form a closed feedback loop. If a closed feedback loop cannot be formed, then the information retransmission is initiated to the receiving point R2 alone.

[0067] Specifically, the difference between the to-be-feedback indicator and the feedback indicator is compared to obtain the receiving node R i The instant deviation B i include:

[0068] Instant deviation B i,j is a measure of the receiving node R i The key parameters of the difference between the feedback and the original warning information on indicator j. Extract the data required for the immediate deviation calculation from the system, including:

[0069] 1. Original indicator value at the sender:

[0070] For the indicator j in the indicators to be fed back, it is recorded as M j , provided by the early warning information generation and release module.

[0071] 2. Feedback indicator value at the receiving end:

[0072] In the feedback index, for index j, it is recorded as F i,j , by the receiving node R i Feedback information provided.

[0073] Choose the appropriate deviation measurement method and define it according to the type of indicator (numerical, categorical or regional):

[0074] For numerical indicators (such as precipitation), the instantaneous deviation is calculated by difference B i,j =F i,j -M j ;

[0075] Classification indicators (such as warning level, warning response), instant deviation is calculated by discrete distance, and weights B are assigned to different categories i,j = distance(F i,j ,M j ):

[0076] Warning level mapping: blue = 1, yellow = 2, orange = 3, red = 4.

[0077] Warning response mapping: Increase alert = 1, prepare to evacuate = 2, evacuate immediately = 3.

[0078] Deviation calculation is based on discrete distances, such as B i,j =|F i,j -M j |.

[0079] For regional indicators (such as disaster areas), the instant deviation is calculated by matching accuracy and is defined as follows:

[0080]

[0081] To quantify the deviations of different types of indicators to the same range, ensure that the metrics are unified to normalize the instant deviations

[0082] Logarithmic Deviation B i,j , with the maximum possible deviation B i,j,max Normalization

[0083] Bisection type bias B i,j Total by Category C max Normalization

[0084] Regional instant deviation direct use Because its value is already in the range [0,1].

[0085] In order to allocate all receiving nodes R i The instant deviation is organized into a matrix for subsequent processing, and the instant deviation matrix B is constructed:

[0086]

[0087] Where: N is the number of receiving nodes, and M is the number of indicators of warning information.

[0088] In one embodiment, assume that:

[0089] The warning information M contains 4 indicators:

[0090] Indicator 1: Warning level M1 = orange warning; Indicator 2: Precipitation M2 = 70 mm / h; Indicator 3: Disaster area M3 = Area A; Indicator 4: Warning response M4 = prepare for evacuation.

[0091] Feedback from receiving node R1:

[0092] F 1,1 =Red Alert; F 1,2 =80 = mm / h; F 1,3 =Area A; F 1,4 =Transfer immediately.

[0093] Instant deviation calculation:

[0094] 1. Type indicators (warning level):

[0095] Mapping: blue = 1, yellow = 2, orange = 3, red = 4;

[0096] Deviation: B 1,1 =|F 1,1 -M1|=|4-3|=1;

[0097] Normalization:

[0098] 2. Numerical indicators (precipitation):

[0099] Deviation: B 1,2 =F 1,2 -M2=80-70=10;

[0100] Normalization: Assumption but:

[0101]

[0102] 3. Regional indicators (disaster areas):

[0103] Deviation: B 1,3 =0 because the feedback value is consistent with the sent value);

[0104] Normalization:

[0105] 4. Type indicators (early warning response):

[0106] Mapping: Raise alert = 1, prepare to move = 2, move immediately = 3;

[0107] Deviation: B 1,4 =|F 1,4 -M4|=|3-2|=1;

[0108] Normalization:

[0109] For the instantaneous deviation of the receiving node R1:

[0110] B1=[0.25 0.1 0 0.33]

[0111] Specifically, the instant deviation B i Added to the previous cumulative deviation value E i,prev , forming the cumulative deviation E i include:

[0112] Cumulative deviation update, the cumulative deviation update is through the dynamic accumulation of instant deviation, recording each receiving node R i The long-term cumulative error trend on each indicator j.

[0113] For each receiving node R i Initialize the cumulative deviation value with index j:

[0114] Take advantage of the current immediate bias Update the cumulative deviation, taking into account the importance of historical cumulative deviations and dynamically responding to new deviations. Before starting this step, it is necessary to ensure that each receiving node has recorded the initial cumulative deviation when it first receives the warning information. Usually, the initial value is zero because there is no historical deviation record. i The system will send node R i The currently calculated instantaneous deviation is combined with its previous cumulative deviation value to form a new cumulative deviation.

[0115] Get node R from the instant deviation matrix i The current instantaneous deviation from indicator j The weighted cumulative deviation is calculated using the following formula:

[0116]

[0117] Where: E i,j Represents the receiving node R i The current cumulative deviation of indicator j; E i,prev Represents the receiving node R i The previous cumulative deviation of indicator j; Indicates that the node receives the node R i The current instant deviation of indicator j; α represents the attenuation factor of historical deviation, 0≤α≤1;

[0118] If α = 0, the cumulative deviation depends entirely on the latest deviation;

[0119] If α=1, the cumulative deviation depends entirely on historical data.

[0120] After updating, the new cumulative deviation E i,j Store for later use.

[0121] For node R i , summarizing the accumulated deviations of all indicators to provide a basis for weight adjustment and global deviation calculation:

[0122]

[0123] Where: E i is the receiving node R i The total cumulative deviation; M is the number of indicators included in the early warning information.

[0124] In one embodiment, assume that:

[0125] Instant deviation matrix B:

[0126]

[0127] The instantaneous deviation of R1 is B = [0.25, 0.1, 0, 0.33]

[0128] The instant deviation of R2 is B = [0.5, 0.2, 0.1, 0.67]

[0129] Historical cumulative deviation:

[0130] R1: E 1,prev =[0.2, 0.05, 0, 0.2]

[0131] R2: E 2,.prec =[0.4, 0.1, 00.5, 0.5]

[0132] Attenuation factor: α = 0.8

[0133] Cumulative deviation update for node R1:

[0134] 1. Indicator 1: E 1,1 =0.8·0.2+(1-0.8)·0.25=0.16+0.05=0.21

[0135] 2. Indicator 2: E 1,2 =0.8·0.05+(1-0.8)·0.1=0.04+0.02=0.06

[0136] 3. Indicator 3: E 1,3=0.8·0+(1-0.8)·0=0

[0137] 4. Indicator 4: E 1,4 =0.8·0.2+(1-0.8)·0.33=0.16+0.066=0.226

[0138] Cumulative deviation vector: E1 = [0.21, 0.06, 0, 0.226]

[0139] Cumulative deviation update for node R2:

[0140] 1. Indicator 1: E 2,1 =0.8·0.4+(1-0.8)·0.5=0.32+0.1=0.42

[0141] 2. Indicator 2: E 2,2 =0.8·0.1+(1-0.8)·0.2=0.08+0.04=0.12

[0142] 3. Indicator 3: E 2,3 =0.8·0.05+(1-0.8)·0.1=0.04+0.02=0.06

[0143] 4. Indicator 4: E 2,4 =0.8·0.5+(1-0.8)·0.67=0.4+0.134=0.534

[0144] Cumulative deviation vector: E1 = [0.42, 0.12, 0.06, 0.534

[0145] Total cumulative deviation:

[0146] 1. Node R1:

[0147] 2. Node R2:

[0148] Specifically, according to the cumulative deviation E i and delay deviation T i The weights of the dynamic adjustment feedback indicators include:

[0149] Dynamic weight adjustment is designed to adjust the weight according to the accumulated deviation E of the node. i , feedback delay T i and historical weight ω i ,The node weights are optimized in real time, so that the system assigns reasonable weights after comprehensively evaluating the importance and reliability of each node, which further affects the global deviation evaluation.

[0150] For each node R i Assign initial weight It means that all nodes are equally important initially, and a dynamic factor for adjusting weights is generated by combining the node cumulative deviation and feedback delay.

[0151] 1. Calculate the deviation factor: F E,i =exp(-β·E i )

[0152] When the deviation E i When it is larger, the deviation factor F E,i Reduce.

[0153] 2. Calculate the delay factor: F T,i =exp(-β·T i )

[0154] When the delay T i When the delay factor F is larger, T,i Reduce.

[0155] Calculate each receiving node R based on the dynamic adjustment factor i The unnormalized weights of :

[0156]

[0157] Represents node R i The unnormalized weights of .

[0158] The unnormalized weights of all nodes Normalized to a weight distribution that sums to 1:

[0159]

[0160] make sure:

[0161]

[0162] The newly calculated weight Replace old weights For subsequent use.

[0163] The weight adjustment process is run regularly within the preset feedback cycle to ensure that the system dynamically responds to the latest node status. In one embodiment, assume that:

[0164] 1. There are 3 receiving nodes: R1, R2, R3.

[0165] 2. Cumulative deviation E i : E1=0.2, E2=0.5, E3=0.1.

[0166] 3. Feedback delay T i : T1=1, T2=3, T3=0.5.

[0167] 4. Parameters: β=1, γ=1.

[0168] Step 1 Dynamic adjustment factor calculation:

[0169] 1. Node R1:

[0170] Deviation factor: F E,1 =exp(-2·0.2)=exp(-0.4)≈0.67032

[0171] Delay factor: F T,2 =exp(-1·1)=exp(-1)≈0.36788

[0172] Unnormalized weights:

[0173] 2. Node R2:

[0174] Deviation factor: F E,2 =exp(-2·0.5)=exp(-1)≈0.36788

[0175] Delay factor: F T,2 =exp(-1·3)=exp(-3)≈0.04979

[0176] Unnormalized weights:

[0177] 3. Node R3:

[0178] Deviation factor: F E,3 =exp(-2·0.1)=exp(-0.2)≈0.81873

[0179] Delay factor: F T,3 =exp(-1·0.5)=exp(-0.5)≈0.60653

[0180] Unnormalized weights: Step 2: Weight normalization:

[0181] 1. Calculate the sum of unnormalized weights: 2. Normalized weights:

[0182]

[0183]

[0184] Final weights:

[0185] Node R1:

[0186] Node R2:

[0187] Node R3:

[0188] Through dynamic adjustment, node R2 obtains a higher weight due to its smaller deviation and delay; the weight of node R2 is significantly reduced due to severe delay.

[0189] Specifically, through the deviation accumulation E i and delay deviation T i The total global deviation D is calculated by the weight of the feedback index i , judgement |D i |Whether the value is lower than the set threshold, the decision to close the feedback loop or initiate information retransmission includes:

[0190] Get each node R from the previous module i : Cumulative deviation E i , Dynamic Weight Feedback delay T i , for each receiving node R i , calculate the total global deviation:

[0191]

[0192] Wherein, ε is the weight coefficient of feedback delay.

[0193] The global deviation sum of the nodes D i With the threshold value D threshold For comparison:

[0194] The global deviation sum of the nodes D i With the threshold value D threshold For comparison:

[0195] If |D i |≤D threshold , receiving node R i The feedback can be closed loop;

[0196] If |D i |≥D threshold , receiving node R i The feedback does not meet the requirements and the information needs to be resent.

[0197] In one embodiment, assume that:

[0198] 1. Number of nodes: 3 receiving nodes R1, R2, R3.

[0199] 2. Dynamic weight: ω1 = 0.4, ω2 = 0.3, ω3 = 0.3

[0200] 3. Cumulative deviation: E1 = 0.2, E2 = 0.5, E3 = 0.1

[0201] 4. Feedback delay: T1 = 1, T2 = 2, T3 = 0.1

[0202] 5. Delay weight coefficient: ε=0.8.

[0203] 6.Threshold: D threshold =0.5.

[0204] Calculate the total global deviation of each node

[0205] Node R1: D1 = 0.4·(0.2+0.8·1) = 0.4·1.0 = 0.4

[0206] |D i |=0.4≤0.5, closing the feedback loop.

[0207] Node R2: S2 = 0.3·(0.5+0.8·2) = 0.3·2.1 = 0.63

[0208] |D i |=0.63≥0.5, it is necessary to initiate information retransmission.

[0209] Node R3: D3 = 0.3·(0.1+0.8·0.5) = 0.3·0.5 = 0.15

[0210] |D i |=0.15≤0.5, closing the feedback loop.

[0211] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein with equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the scope of protection of the claims of the present invention.

Claims

1. A flash flood disaster warning SMS accurate response and receipt feedback system, characterized in that: include: The warning information generation and release module is used to send warning information to each receiving node R i Send warning information M; Receiving Node R i , used to receive the warning information M, and perform local confirmation according to the indicators to be fed back contained in the warning information M, and send feedback information F to the warning information generation and release module i , where the feedback information F i Contains the receiving node R i Feedback indicators; The comparison and correction module compares the difference between the to-be-feedback indicator and the feedback indicator to obtain the receiving node R i The instant deviation B i ; The instant deviation B i Added to the previous cumulative deviation value E i,prev , forming the cumulative deviation E i ; According to the cumulative deviation E i and delay deviation T i Dynamically adjust the weight of feedback indicators; By the deviation accumulator E i and delay deviation T i The total global deviation D is calculated by the weight of the feedback index i , judgement |D i |Whether it is lower than the set threshold, it decides to close the feedback loop or initiate information retransmission.

2. According to claim 1, a flash flood disaster warning SMS accurate response and receipt feedback system is characterized in that: In the comparison and correction module, the difference between the indicator to be fed back and the feedback indicator is compared to obtain the receiving node R i The instant deviation B i include: For the indicator j in the indicators to be fed back, it is recorded as M j ; For the indicator j in the feedback indicator, it is recorded as F i,j ; Define deviation measurement methods according to the type of indicators, including numerical indicators, categorical indicators, and regional indicators; For numerical indicators, the instantaneous deviation is calculated by the difference B i,j =F i,j -M j ; For the classification indicators, the instant deviation is calculated by discrete distance, and the weights B between different classifications are assigned. i,j = distance(F i,j ,M j ); For regional indicators, the instant deviation is calculated by matching accuracy, which is defined as follows: Logarithmic instant deviation B i,j , with the maximum possible deviation B i,j,max Normalization Bisection type instant deviation B i,j , by category total C max Normalization Regional instant deviation B i,j ,make Construct the instant deviation matrix B: Where: N is the number of receiving nodes, and M is the number of indicators of warning information.

3. A flash flood disaster warning SMS accurate response and receipt feedback system according to claim 2, characterized in that: In the comparison and correction module, the instant deviation B i Added to the previous cumulative deviation value E i,prev , forming the cumulative deviation E i include: For each receiving node R i Initialize the cumulative deviation value with index j; Get the receiving node R from the instant deviation matrix i The current instantaneous deviation from indicator j Calculate the weighted cumulative deviation: Where: E i,j Represents the receiving node R i The current cumulative deviation of indicator j; E i,prev Represents the receiving node R i The previous cumulative deviation of indicator j; Indicates that the node receives the node R i The current instant deviation of indicator j; α represents the attenuation factor of historical deviation, 0≤α≤1; If α = 0, the cumulative deviation depends entirely on the latest deviation; If α = 1, the cumulative deviation depends entirely on historical data; For the receiving node R i , summarizing the accumulated deviations of all indicators to provide a basis for weight adjustment and global deviation calculation: Where: E i is the receiving node R i The total cumulative deviation; M is the number of indicators included in the early warning information.

4. According to claim 3, a flash flood disaster warning SMS accurate response and receipt feedback system is characterized in that: In the comparison and correction module, the cumulative amount of deviation E i and delay deviation T i The weights of the dynamic adjustment feedback indicators include: For each receiving node R i Assign initial weight Indicates that all receiving nodes are initially equally important; Calculate the deviation factor: F E,i =exp(-β·E i ) When the deviation E i When it is larger, the deviation factor F E,i reduce; Calculate the delay factor: F T,i =exp(-γ·T i ) When the delay T i When the delay factor F is larger, T,i reduce; Calculate each receiving node R based on the dynamic adjustment factor i The unnormalized weights of : Represents the receiving node R i The unnormalized weights of The unnormalized weights of all nodes Normalized to a weight distribution that sums to 1; make sure: The newly calculated weight Replace old weights 5. According to claim 3, a flash flood disaster warning SMS accurate response and receipt feedback system is characterized in that: In the comparison and correction module, the accumulated deviation E i and delay deviation T i The total global deviation D is calculated by the weight of the feedback index i , judgement |D i |Whether the value is lower than the set threshold, the decision to close the feedback loop or initiate information retransmission includes: For each receiving node R i , calculate the total global deviation: Among them, ε is the weight coefficient of feedback delay; The global deviation sum of the nodes D i With the threshold value D threshold For comparison: If |D i |≤D threshold , receiving node R i The feedback can be closed loop; If |D i |≥D threshold , receiving node R i The feedback does not meet the requirements and the information needs to be resent.

Images