Evaluation method for emergency decision-making effect of dam break emergencies based on multi-element coupling

Abstract

The invention discloses a multi-element coupled dam break emergency emergency decision effect evaluation method. According to the method, through organization-personnel-technology multi-element coupling, a mapping and evolution mechanism of emergency decision-making elements and disaster-causing consequences of dam break emergencies is represented. The post-disaster emergency decision-making effect evaluation method based on the mathematical model is provided, the emergency decision-making process of a dam break emergency event is described under data-mechanism driving, and the emergency decision-making effect is quantified. According to the method, the emergency decision-making effect of the dam break emergency can be reasonably and reliably evaluated, disaster-causing consequences are reduced, and a scientific basis is provided for timely and effectively realizing emergency disposal decision-making.

Description

Technical Field

[0001] The present invention relates to a multi-factor coupled dam-break emergency decision-making effect evaluation method, specifically refers to forming a scientific and quantitative basis for judging the disaster emergency decision-making effect by establishing a multi-factor coupled relationship of organization-personnel-technology. Background Art

[0002] Emergency decision-making for dam-break incidents is a key means to reduce the consequences of disasters. It refers to the dynamic process in which the decision-making subject collects information and data from all parties under the subjective and objective environment at the time, analyzes and judges the situation based on prior knowledge, technology, and experience, and formulates emergency response plans in a short period of time to control the development of the disaster until the incident is eliminated. With the development of the economy and society, the objects of protection downstream of reservoir dams in my country have undergone profound changes (increased population, increased infrastructure, and expanded urban scale, etc.), and reservoir dam breaches will face unprecedented losses of life and property. Effective emergency decision-making for dam-break incidents is a key means to reduce the consequences of disasters, especially to ensure that "no casualties" occur.

[0003] The emergency decision-making of dam-break incidents involves the coupling of multiple dynamic factors, including people (emergency organizations, downstream people), water (dam-break floods), and objects (rescue and emergency materials), and is characterized by urgency, spread, and comprehensive complexity. The existing emergency decision-making methods only consider the impact of one or several factors, lack the consideration of comprehensive factors of emergency decision-making, and have unclear understanding of the coupling relationship between factors. The critical warning decision-making time is vaguely defined, and the effect of emergency decision-making cannot be measured. It is difficult to form a scientific and timely emergency response plan, and it cannot be effectively applied in the emergency process. In dam-break incidents, wrong decision-making plans may even lead to an exponential increase in the consequences and losses of the disaster. Summary of the invention

[0004] Purpose of the invention: In view of the problems and shortcomings in the prior art, the present invention provides a multi-factor coupled dam break emergency decision-making effect evaluation method, which can extract and learn the quantitative relationship between decision factors and decision effects, and predict the emergency decision-making results of dam break emergencies through machine learning simulation.

[0005] Technical solution: A multi-factor coupled emergency decision-making effect evaluation method for dam break emergencies. By establishing a multi-factor coupled relationship between organization, personnel and technology, a model is built to characterize the emergency decision-making effect. Combined with the dam engineering elements, the risk level perceived by the emergency organization, the emergency material allocation behavior, the rescue material follow-up rate, the number of people at risk, the evacuation rate, the average flooding degree of the disaster area, and the number of people who lost their lives are quantified to form a basis for judging the effectiveness of disaster emergency decision-making. A neural network model is established - a learning model for the emergency decision-making effect of dam break emergencies based on a shared learning architecture to estimate the dam risk level and the number of people who lost their lives, so as to quantify the effectiveness of post-disaster emergency decision-making.

[0006] A learning model for the effect of emergency decision-making in dam break emergencies based on a shared learning architecture is constructed; the input of the model is the emergency decision-making elements of dam break emergencies, including: (1) the risk level perceived by the emergency organization, (2) the allocation behavior of emergency materials, (3) the follow-up rate of rescue materials, (4) the number of people at risk, (5) the evacuation rate, and (6) the average flooding degree of the disaster area; the output of the model is the effect of post-disaster emergency decision-making, including: (1) the risk level of the dam, and (2) the number of people whose lives were lost; the learning model for the effect of emergency decision-making in dam break emergencies is optimized by driving historical data; the emergency decision-making elements of dam break emergencies at the current moment are input, and the emergency decision-making effect at the current moment is output.

[0007] The inputs are the risk level perceived by the emergency organization, the deployment of emergency materials, the follow-up rate of rescue materials, the number of people at risk, the evacuation rate, and the average flooding degree of the disaster area. The expression is as follows:

[0008] X={a,b,c,d,e,f}

[0009] Where a is the risk level perceived by the emergency organization, b is the deployment behavior of emergency materials, c is the follow-up rate of rescue materials, d is the evacuation rate, e is the number of people at risk, and f is the average flooding degree of the disaster area; a and b are auxiliary variables, c and d are speed variables, and e and f are state variables.

[0010] The output of the dam-break emergency decision-making effect learning model based on the shared learning architecture is used to obtain the dam risk level and the number of lives lost, which can be expressed as follows:

[0011] y={θ,λ}

[0012] Where θ represents the dam risk level; λ represents the number of lives lost.

[0013] The model architecture is:

[0014] The first layer is the input layer: the attributes of the input data divide the input into three groups, corresponding to auxiliary variables, speed variables and state variables, which fully describe the organizational, personnel and technical elements;

[0015] The second layer is the organization-personnel-technology coupling relationship calculation layer, which is used to calculate the quantitative value of the coupling relationship between the three;

[0016] The third layer is the coupling weighted module layer, which uses the coupling quantization value as the weight to perform weighted calculation and splicing on the vectors composed of auxiliary variables, speed variables and state variables, respectively, to form three-channel features of three-channel auxiliary variables, speed variables and state variables;

[0017] The fourth layer is the attention weighting module layer, which performs attention weighting on the three-channel features;

[0018] The fifth layer is the concatenation layer, which concatenates the attention-weighted features to form multi-factor comprehensive features;

[0019] The sixth to eighth layers are two-stream recognition layers, including two-stream convolutional layers, pooling layers, and Softmax layers, which are used to output the dam risk level and the number of lives lost, representing the effectiveness of emergency decision-making at the current moment.

[0020] Calculate the coupling relationship between the three elements of organization, personnel and technology;

[0021] Among them, the organizational, personnel and technical elements are:

[0022] x1=[a,b], x2=[c,d], x3=[e,f]

[0023] Calculate the coupling relationship weights among the three elements of organization, personnel and technology:

[0024] Q 11 =<x1,x1> , Q 12 =<x1,x2> , Q 13 =<x1,x3>

[0025] Q 21 =<x2,x1> , Q 22 =<x2,x2> , Q 23 =<x2,x3>

[0026] Q 31 =<x3,x1> , Q 32 = <x 31 ,x2>,Q 33 =<x3,x3>

[0027] Among them, <> is the dot product operation.<x1,x1> ,

[0028] Based on the quantitative results of the element coupling relationship, the vectors composed of auxiliary variables, speed variables and state variables are weighted respectively:

[0029] F1=concat(Q 11 x1 T ,Q 12 x1 T ,Q 13 x1 T )

[0030] F2=concat(Q 21 x2 T ,Q 22 x2 T ,Q 23 x2T )

[0031] F3=concat(Q 31 x3 T ,Q 32 x3 T ,Q 33 x3 T )

[0032] Among them, concat() is a concatenation calculation, which performs attention weighted concatenation on F1, F2, and F3 to obtain comprehensive features:

[0033] F=con(W1F1,W2F2,W3F3)

[0034] Among them, W1, W2, and W3 are the attention weights corresponding to the characteristics of organizational, personnel, and technical elements;

[0035] The concatenated comprehensive feature F is input into the pooling layer and the softmax layer for calculation to obtain the dam risk level θ and the number of lives lost λ.

[0036] The emergency decision-making effect learning model of the dam-break incident is optimized by driving historical data;

[0037] First, a loss function is constructed to optimize the dam risk level and the number of lives lost. The specific expression is as follows:

[0038] L=ɑ1L1+α2L2

[0039] Where L1 represents the loss function for optimizing the dam risk level, and L2 represents the loss function for optimizing the number of lives lost.

[0040] Beneficial effects: The multi-factor coupled dam-break emergency decision-making effect evaluation method provided by the present invention takes into account the coupling between multiple dynamic factors of organization, personnel, and technology involved in the emergency decision-making of dam-break emergencies. It breaks through the error of a single factor in predicting the effect of emergency decision-making, accurately evaluates the effect of emergency decision-making, and improves the effectiveness of emergency decision-making for disaster prevention and mitigation. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a model structure diagram of an embodiment of the present invention. DETAILED DESCRIPTION

[0042] The present invention is further explained below in conjunction with specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention. After reading the present invention, various equivalent forms of modifications to the present invention by those skilled in the art all fall within the scope defined by the claims attached to this application.

[0043] A multi-factor coupled emergency decision-making effect evaluation method for dam break emergencies, which establishes a multi-factor coupled relationship between organization, personnel and technology to model and characterize the emergency decision-making effect; combined with dam engineering elements, quantify the risk level perceived by the emergency organization, the deployment behavior of emergency materials, the follow-up rate of rescue materials, the number of people at risk, the evacuation rate, the average degree of flooding in the disaster area, and the number of people who lost their lives, to form a basis for judging the effectiveness of disaster emergency decision-making; establish a neural network model - a learning model for emergency decision-making effects of dam break emergencies based on a shared learning architecture, estimate the dam risk level and the number of people who lost their lives, so as to quantify the effectiveness of post-disaster emergency decision-making.

[0044] A learning model for the effect of emergency decision-making in dam break emergencies based on a shared learning architecture is constructed; the input of the model is the emergency decision-making elements of dam break emergencies, including: (1) the risk level perceived by the emergency organization, (2) the allocation behavior of emergency materials, (3) the follow-up rate of rescue materials, (4) the number of people at risk, (5) the evacuation rate, and (6) the average flooding degree of the disaster area; the output of the model is the effect of post-disaster emergency decision-making, including: (1) the risk level of the dam, and (2) the number of people whose lives were lost; the learning model for the effect of emergency decision-making in dam break emergencies is optimized by driving historical data; the emergency decision-making elements of dam break emergencies at the current moment are input, and the emergency decision-making effect at the current moment is output.

[0045] The model input expression is as follows:

[0046] X={a,b,c,d,e,f}

[0047] In the formula, a is the risk level perceived by the emergency organization, b is the deployment behavior of emergency materials, c is the follow-up rate of rescue materials, d is the evacuation rate, e is the number of people at risk, and f is the average flooding degree of the disaster area; a and b are auxiliary variables, c and d are speed variables, and e and f are state variables;

[0048] The input is divided into three groups according to the attributes of the input data.

[0049] x1=[a,b], x2=[c,d], x3=[e,f]

[0050] At this point, the input of the emergency decision-making learning model for dam break emergencies based on the shared learning architecture is completed.

[0051] The learning module consists of 2 parallel streams, each of which receives 3 sets of inputs at the same time and inputs them into the shared layer. The specific expression is as follows:

[0052]

[0053] In the formula, is the shared weight, x′1, x′2, and x′3 are the eigenvalues ​​after shared learning, which are input into the convolution layer with a convolution kernel size of 1×5 for convolution calculation, and the outputs are x″1, x″2, and x″3, which are input into the concat (function) layer for splicing. The specific expression is as follows:

[0054] X″=[x″1x″2x″3]

[0055] The concatenated feature X″ is input into the pooling layer and the softmax layer for calculation to obtain the dam risk level θ and the number of lives lost λ. The feature sequence obtained after shared learning is further input into the convolution layer with a convolution kernel size of 1×5 for convolution calculation, and further input into the concat layer for concatenation. The concatenated features are input into the pooling layer and the softmax layer for calculation to obtain the calculation results of the dam risk level and the number of lives lost.

[0056] At this point, the output of the emergency decision-making effect learning model for dam break emergencies based on the shared learning architecture has been completed.

[0057] In the process of model optimization, the historical database is used to drive model optimization:

[0058] A loss function is constructed to optimize the dam risk level and the number of lives lost. The specific expression is as follows:

[0059] L=ɑ1L1+ɑ2L2

[0060] Where L1 represents the loss function that optimizes the dam risk level, and L2 represents the loss function that optimizes the number of lives lost;

[0061] Driven by the historical database, the shared layer parameters are optimized through the back propagation algorithm. The specific expression is as follows:

[0062]

[0063] Then the weight parameters are updated through back propagation, as follows:

[0064]

[0065] The following expression is obtained by updating the weight parameters:

[0066]

[0067] Where ε represents the learning rate of multi-task learning.

[0068] At this point, the optimization of the emergency decision-making effect learning model for dam break emergencies based on the shared learning architecture has been completed.

[0069] The output expression of the emergency decision-making effect learning model for dam break emergencies based on the shared learning architecture is as follows:

[0070] y={θ,λ}

[0071] Where θ represents the dam risk level; λ represents the number of lives lost.

[0072] like Figure 1 As shown, the model architecture is:

[0073] The first layer is the input layer: the attributes of the input data divide the input into three groups, corresponding to auxiliary variables, speed variables and state variables, which fully describe the organizational, personnel and technical elements;

[0074] The second layer is the organization-personnel-technology coupling relationship calculation layer, which is used to calculate the quantitative value of the coupling relationship between the three;

[0075] The third layer is the coupling weighted module layer, which uses the coupling quantization value as the weight to perform weighted calculation and splicing on the vectors composed of auxiliary variables, speed variables and state variables, respectively, to form three-channel features of three-channel auxiliary variables, speed variables and state variables;

[0076] The fourth layer is the attention weighting module layer, which performs attention weighting on the three-channel features;

[0077] The fifth layer is the concatenation layer, which concatenates the attention-weighted features to form multi-factor comprehensive features;

[0078] The sixth to eighth layers are two-stream recognition layers, including two-stream convolutional layers, pooling layers, and Softmax layers, which are used to output the dam risk level and the number of lives lost, representing the effectiveness of emergency decision-making at the current moment.

[0079] Calculate the coupling relationship weights among the three elements of organization, personnel and technology:

[0080] Q 11 =<x1,x1> , Q 12 =<x1,x2> , Q 13 =<x1,x3>

[0081] Q 21 =<x2,x1> , Q 22 =<x2,x2> , Q 23 =<x2,x3>

[0082] Q 31 =<x3,x1> , Q 32 = <x 31 ,x2>,Q 33 =<x3,x3>

[0083] Among them, <> is the dot product operation,<x1,x1> For example,

[0084] In the quantitative results based on the element coupling relationship, the vectors composed of auxiliary variables, speed variables and state variables are weighted respectively:

[0085] F1=concat(Q 11 x1 T ,Q 12 x1 T ,Q 13 x1 T )

[0086] F2=concat(Q 21 x2 T ,Q 22 x2 T ,Q 23 x2 T )

[0087] F3=concat(Q 31 x3 T ,Q 32 x3 T ,Q 33 x3 T )

[0088] Among them, concat() is a concatenation calculation, which performs attention weighted concatenation on F1, F2, and F3 to obtain comprehensive features:

[0089] F=concat(W1F1,W2F2,W3F3)

[0090] Among them, W1, W2, and W3 are the attention weights corresponding to the characteristics of organizational, personnel, and technical elements;

[0091] The concatenated comprehensive feature F is input into the pooling layer and the softmax layer for calculation to obtain the dam risk level θ and the number of lives lost λ.

Claims

1. A multi-factor coupled dam-break emergency decision-making effect evaluation method, characterized in that: By establishing the coupling relationship of multiple factors such as organization, personnel and technology, a model is built to characterize the effect of emergency decision-making. By combining the dam engineering elements, the risk level perceived by the emergency organization, the behavior of emergency material allocation, the follow-up rate of rescue materials, the number of people at risk, the evacuation rate, the average degree of flooding in the disaster area, and the number of people who lost their lives are quantified to form a basis for judging the effectiveness of disaster emergency decision-making. A neural network model is established to estimate the dam risk level and the number of people who lost their lives in order to quantify the effect of post-disaster emergency decision-making.

2. The multi-factor coupled dam-break emergency decision-making effect evaluation method according to claim 1 is characterized in that: Construct a learning model for emergency decision-making effects of dam-break emergencies based on a shared learning architecture; The model input is the emergency decision-making elements of dam break emergencies, including: (1) the risk level perceived by the emergency organization, (2) the allocation of emergency materials, (3) the follow-up rate of rescue materials, (4) the number of people at risk, (5) the evacuation rate, and (6) the average flooding degree of the disaster area; the model output is the post-disaster emergency decision-making effect, including: (1) the dam risk level, and (2) the number of people whose lives were lost; the emergency decision-making effect learning model of dam break emergencies is optimized by driving historical data; the emergency decision-making elements of dam break emergencies at the current moment are input, and the emergency decision-making effect at the current moment is output.

3. The multi-factor coupled dam-break emergency decision-making effect evaluation method according to claim 2 is characterized in that: The inputs are the risk level perceived by the emergency organization, the deployment of emergency materials, the follow-up rate of rescue materials, the number of people at risk, the evacuation rate, and the average flooding degree of the disaster area. The expression is as follows: X={a,b,c,d,e,f} In the formula, a is the risk level perceived by the emergency organization, b is the deployment behavior of emergency materials, c is the follow-up rate of rescue materials, d is the evacuation rate, e is the number of people at risk, and f is the average flooding degree of the disaster area; a and b are auxiliary variables, c and d are speed variables, and e and f are state variables; The output of the dam-break emergency decision-making effect learning model based on the shared learning architecture is used to obtain the dam risk level and the number of lives lost, which can be expressed as follows: y={θ,λ} Where θ represents the dam risk level; λ represents the number of lives lost.

4. The multi-factor coupled dam-break emergency decision-making effect evaluation method according to claim 2 is characterized in that: Construct a learning model for emergency decision-making effect of dam break incidents based on shared learning architecture. The model architecture is as follows: The first layer is the input layer: the attributes of the input data divide the input into three groups, corresponding to auxiliary variables, speed variables and state variables, which fully describe the organizational, personnel and technical elements; The second layer is the organization-personnel-technology coupling relationship calculation layer, which is used to calculate the quantitative value of the coupling relationship between the three; The third layer is the coupling weighted module layer, which uses the coupling quantization value as the weight to perform weighted calculation and splicing on the vectors composed of auxiliary variables, speed variables and state variables, respectively, to form three-channel features of three-channel auxiliary variables, speed variables and state variables; The fourth layer is the attention weighting module layer, which performs attention weighting on the three-channel features; The fifth layer is the concatenation layer, which concatenates the attention-weighted features to form multi-factor comprehensive features; The sixth to eighth layers are two-stream recognition layers, including two-stream convolutional layers, pooling layers, and Softmax layers, which are used to output the dam risk level and the number of lives lost, representing the effectiveness of emergency decision-making at the current moment.

5. The multi-factor coupled dam-break emergency decision-making effect evaluation method according to claim 4 is characterized in that: Calculate the coupling relationship between the three elements of organization, personnel and technology; Among them, the organizational, personnel and technical elements are: x1=[a,b], x2=[c,d], x3=[e,f] Calculate the coupling relationship weights among the three elements of organization, personnel and technology: Q 11 =<x1,x1>,Q 12 =<x1,x2>,Q 13 =<x1,x3> Q 21 =<x2,x1>,Q 22 =<x2,x2>,Q 23 =<x2,x3> Q 31 =<x3,x1> ,Q 32 = <x 31 ,x2>,Q 33 =<x3,x3> Among them, <> is the dot product operation.<x1,x1> , 6. The multi-factor coupled dam-break emergency decision-making effect evaluation method according to claim 4 is characterized in that: Based on the quantitative results of the element coupling relationship, the vectors composed of auxiliary variables, speed variables and state variables are weighted respectively: F1=concat(Q 11 x1 T ,Q 12 x1 T ,Q 13 x1 T ) F2=concat(Q 21 x2 T ,Q 22 x2 T ,Q 23 x2 T ) F3=concat(Q 31 x3 T ,Q 32 x3 T ,Q 33 x3 T ) Among them, concat() is a concatenation calculation, which performs attention weighted concatenation on F1, F2, and F3 to obtain comprehensive features: F=concat(W1F1,W2F2,W3F3) Among them, W1, W2, and W3 are the attention weights corresponding to the characteristics of organizational, personnel, and technical elements; The concatenated comprehensive feature F is input into the pooling layer and the softmax layer for calculation to obtain the dam risk level θ and the number of lives lost λ.

7. The multi-factor coupled dam-break emergency decision-making effect evaluation method according to claim 4 is characterized in that: The emergency decision-making effect learning model of the dam-break incident is optimized by driving historical data; First, a loss function is constructed to optimize the dam risk level and the number of lives lost. The specific expression is as follows: L=α1L1+α2L2 Where L1 represents the loss function for optimizing the dam risk level, and L2 represents the loss function for optimizing the number of lives lost.

Images