Emergency resource allocation method and system for narrow non-exposed space based on consequence research

Abstract

The application provides a kind of narrow non-exposed space emergency resource allocation method and system based on consequence research, belongs to emergency rescue technical field, and includes the following steps: constructing various weight influence factors of secondary disaster under narrow non-exposed area emergency;Ideal point is constructed into progress model, and the probability of secondary disaster in the narrow non-exposed space of disaster is estimated;Determine the resource demand of disaster point and distribute materials by improved bee colony algorithm, satisfy the highest time satisfaction, the most fair distribution and the minimum total cost.The application can timely and efficiently allocate emergency resources for multiple narrow non-exposed space disasters, and the allocation result is good and timely.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of emergency rescue, in particular to a narrow non-exposed space emergency resource allocation method and system based on consequence analysis. BACKGROUND

[0002] The narrow non-exposed space emergency has the characteristics of easy occurrence of secondary disasters, and the traditional resource allocation method according to the scale of the rescue center, the distance from the disaster point and the disaster type does not consider the evolution process of the disaster situation, which is easy to cause the phenomenon of uneven allocation of resources. SUMMARY

[0003] The present application aims to provide a narrow non-exposed space emergency resource allocation method and system based on consequence analysis, to solve at least one of the technical problems existing in the background art.

[0004] In order to achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0005] On the one hand, the present application provides a narrow non-exposed space emergency resource allocation method based on consequence analysis, comprising:

[0006] determining the comprehensive weight of the influence factors affecting the secondary damage of the narrow non-exposed space;

[0007] constructing a positive ideal point matrix and an inverse ideal point matrix for the narrow non-exposed space where the accident occurs, and calculating the ideal point sticking degree of each influence factor, combining the comprehensive weight of each influence factor, and calculating the possibility of secondary disaster of the narrow non-exposed space where the accident occurs;

[0008] Based on the possibility of secondary damage of the narrow non-exposed area, the required materials of each affected narrow non-exposed area are calculated, and a multi-objective function with the highest time satisfaction, the most fair allocation and the lowest rescue cost is constructed, and an improved bee colony algorithm is used to solve the resource allocation method; wherein,

[0009] The improved bee colony algorithm is to consider that the disaster conditions of each disaster point are different when generating the initial solution, and to initialize the population according to the resource demand ratio when generating the initial solution:

[0010] X ijmax '=X i ' j +β·{r imax ∈R i ,d min}

[0011] Wherein, X ijmaxβ represents the initial resource allocation coefficient of the required resources corresponding to the disaster point with the largest required amount of rescue materials, r imax R represents the maximum type of rescue materials required by the disaster point i i d represents the total amount of rescue materials required by the disaster point i min X represents that the initial resource allocation of rescue materials is from the rescue center closest to the disaster point i ' j The initial resource allocation is randomly generated.

[0012] Preferably, the subjective weight and objective weight of the influencing factors affecting secondary injury in narrow non-exposed space are determined, and the comprehensive weight is calculated by the way of combination weighting of game theory.

[0013] Preferably, the index sample data is sorted, and the Gini coefficient G of each index x i is calculated. i :

[0014]

[0015] Wherein: m represents the sample number; y ip , y iq are the p, q sample values corresponding to the index x i , and Y i represents the sum of the sample data of the index x i .

[0016] The least important index x i is selected from the index set X = {x t}(i = 1, 2,..., n), and the relative importance r t of each index to x i is calculated.

[0017]

[0018] Wherein, G t represents the Gini coefficient of the least important index x t .

[0019] The subjective weight of each evaluation index x i is calculated as:

[0020]

[0021] The information content C of the index x i is calculated by using range standardization on the index sample data. i :

[0022]

[0023] wherein, is the index standard deviation coefficient; r ij is the correlation coefficient between indexes;

[0024] The evaluation index x i The objective weight is:

[0025]

[0026] Preferably, the combination coefficient κ k The optimization obtains a countermeasure model:

[0027]

[0028] The condition for converting into the optimal first derivative is:

[0029]

[0030] The combination coefficient κ k , k = 1, 2,..., L, and normalizing it:

[0031]

[0032] The evaluation index x i The optimal comprehensive weight is:

[0033]

[0034] Preferably, the risk evaluation index of a specific secondary accident in a narrow non-exposed space is λ ij , i = 1, 2,..., m; j = 1, 2,..., n, wherein m is the number of regions to be evaluated, n is the number of evaluation indexes, and the evaluation matrix is divided into three aspects of factors, which are environmental instability factors, factors affecting rescue timeliness, and inducing factors.

[0035]

[0036] Positive indexes and inverse indexes are used to evaluate the possibility of secondary disasters in narrow non-exposed areas; wherein, the positive indexes are positively correlated with the possibility of secondary accidents, and the inverse indexes are the opposite;

[0037] Assuming that the possibility of secondary accidents changes in a monotonic trend, the positive ideal point and the negative ideal point can be determined.

[0038] When the evaluation index is a positive index, the positive ideal point and the negative ideal point vectors are:

[0039]

[0040] When the evaluation index is the inverse index, the positive ideal point and the negative ideal point vector are:

[0041]

[0042] wherein f i (p) and f i (n) are the positive ideal point vector and the negative ideal point vector of the i-th index of the risk of the specific secondary accident in the narrow non-exposed space, respectively, λ i is the weight of the i-th index, and p r is the risk evaluation index value of the specific secondary accident in the narrow non-exposed space.

[0043] The ideal point evaluation function is the distance between the index and the ideal point and the negative ideal point, and the relative distance between the index and the positive ideal point and the negative ideal point is expressed by the Euclidean distance.

[0044] The distance to the positive ideal point is:

[0045]

[0046] The distance to the negative ideal point is:

[0047]

[0048] The ideal point sticking degree is calculated as:

[0049]

[0050] The probability of the injury occurring in a certain narrow non-exposed space region that may trigger a secondary accident is defined as:

[0051]

[0052] Preferably, according to the size of the determined possibility of triggering a secondary accident, it is judged whether the impact of the secondary accident is considered when resources are allocated to the narrow non-exposed region i:

[0053]

[0054] wherein E r is the judgment of whether a certain narrow non-exposed region i needs to consider the secondary accident, 1 means considering the secondary accident, and 0 means not considering the secondary accident, p i represents the possibility of the narrow non-exposed region i being affected by the secondary accident, and α is the threshold value for determining whether the secondary accident needs to be considered.

[0055] Then the material demand of the disaster point is:

[0056] R i = r i1 · I ti + r i2·E ri ·I ti

[0057] where, R i represents the required resources of disaster point i, r i1 represents the average amount of resources required by each person in a first-incident, r i2 represents the average amount of resources required by each person in a second-incident, E ri represents the second-incident coefficient of disaster point i, I ti represents the number of trapped people in disaster point i at time t.

[0058] Assuming that the current decision-making time is t, there are n disaster points and m rescue centers. Due to the different number of trapped people in each disaster point, in order to solve this type of resource allocation problem and make resources as fairly distributed as possible to each disaster point, an emergency weight coefficient w i is assigned to each disaster point, which represents the emergency degree of disaster point i, and the calculation formula is

[0059]

[0060] Based on the emergency degree, the value function represented by the rescue time deviation degree is used to measure the satisfaction of trapped personnel to the rescue time:

[0061]

[0062] where, T i represents the rescue event deviation degree of disaster point i; t ij represents the generalized event distance between rescue center j and disaster point i; J represents the rescue center set; X ij represents a 0 / 1 variable, which is 1 if rescue is performed, and 0 otherwise;

[0063] The time satisfaction function is:

[0064]

[0065] In the formula, since the slope of the value function changes, it is a concave function in the gain interval, a is less than or equal to 1, and a convex function in the loss interval, b is less than or equal to 1; according to the risk aversion principle, the function in the loss interval is steeper than that in the gain interval,

[0066]

[0067] where, F1 represents the time satisfaction of one of the objective functions, and the greater the time satisfaction is, the better;

[0068] In order to make the allocation more fair and reasonable, the variance of the resource satisfaction degree of all disaster points is used to measure the fairness of rescue; then:

[0069]

[0070]

[0071]

[0072] wherein, n ij represents the amount of supplies sent by the rescue center j to the disaster point i; P i represents the degree of satisfaction of the supplies at the disaster point i, represents the average degree of satisfaction of the supplies at all disaster points, F2 is one of the objective functions, and the smaller the variance, the smaller the difference, i.e., the more equitable the distribution;

[0073] The cost function is defined as follows:

[0074]

[0075] wherein, F3 represents one of the objective functions, and the smaller the cost, the better, C ij represents the cost of sending a unit of supplies from the rescue center j to the disaster point i;

[0076] In the allocation of rescue resources, the supplies are allocated according to the following objective functions:

[0077]

[0078]

[0079]

[0080]

[0081]

[0082] X ij = 0 / 1

[0083] if X ij = 0 → n ij = 0

[0084] if X ij = 1 → 0 < n ij < min{R i , G j}

[0085] wherein, G j represents the amount of supplies in the rescue center, and the allocation of supplies in each rescue center will not exceed the demand of the disaster point, and similarly, the demand of each disaster point will not exceed the maximum storage capacity of the rescue center.

[0086] In a second aspect, the present application provides a narrow non-exposed space emergency resource allocation system based on consequence research and judgment, comprising:

[0087] A determining module is configured to determine the comprehensive weight of the influence factors affecting secondary damage in the narrow non-exposed space.

[0088] A first calculating module is configured to construct a positive ideal point matrix and an inverse ideal point matrix for the narrow non-exposed space where the accident occurs, and calculate the ideal point sticking degree of each influence factor, and combine the comprehensive weight of each influence factor to calculate the possibility of secondary disaster in the narrow non-exposed space where the accident occurs.

[0089] A second calculating module is configured to calculate the required resources of each affected narrow non-exposed area based on the possibility of secondary damage in the narrow non-exposed area, and construct a multi-objective function with the highest time satisfaction, the most fair distribution and the lowest rescue cost, and use an improved bee colony algorithm to solve the resource allocation mode.

[0090] In the improved bee colony algorithm,

[0091] In the improved bee colony algorithm, when randomly generating initial foraging bees, the disaster conditions of each disaster point are considered, and the initial population is initialized according to the resource demand ratio when generating the initial solution.

[0092] X ijmax '=X i ' j +β·{r imax ∈R i ,d min}

[0093] In the improved bee colony algorithm, X ijmax ' represents a newly generated initial foraging bee, β represents an initial resource allocation coefficient of the required resource corresponding to the disaster point with the largest required rescue resource, r imax represents the largest type of rescue resource required by the disaster point i, R i represents the total amount of rescue resources required by the disaster point i, d min represents that the rescue resource is allocated from the nearest rescue center to the disaster point after initial resource allocation, and X i ' j represents a foraging bee randomly generated after initial resource allocation.

[0094] In a third aspect, the present application provides a non-transitory computer readable storage medium for storing computer instructions, wherein the computer instructions are executed by a processor to implement the narrow non-exposed space emergency resource allocation method based on consequence research and judgment as described above.

[0095] In a fourth aspect, the present application provides a computer program product comprising a computer program for implementing the method for allocating emergency resources in a narrow non-exposed space based on consequence judgment as described above when run on one or more processors.

[0096] In a fifth aspect, the present application provides an electronic device comprising a processor, a memory and a computer program; wherein the processor is connected with the memory, and the computer program is stored in the memory; when the electronic device is running, the processor executes the computer program stored in the memory to make the electronic device execute the instructions for implementing the method for allocating emergency resources in a narrow non-exposed space based on consequence judgment as described above.

[0097] The present application has the following advantages: the method of using game theory combination weighting comprehensively considers the influence of factors affecting secondary accidents caused by narrow non-exposed spaces, and comprehensively weights subjective factors and objective factors. The ideal point and inverse ideal point set of the factors affecting the secondary accidents in the narrow non-exposed space are constructed, the ideal point is calculated, and the possibility of the secondary accidents in the narrow non-exposed space is calculated in combination with the weight of each influencing factor. Under the condition of considering the influence of secondary accidents, a multi-objective function is constructed with the highest time satisfaction, the most fair distribution and the minimum cost, an improved bee colony algorithm is used for solving, and the final distribution scheme is obtained by comparing with other solving algorithms. The bee colony algorithm is more optimal in time satisfaction, more fair in distribution and lower in cost, and is suitable for actual rescue environment; and the improved bee colony algorithm greatly improves the timeliness of the algorithm, and the operation time under the same condition is almost one tenth of the original, which can effectively respond to emergencies and provide protection for emergency rescue after disaster.

[0098] The advantages of the additional aspects of the present application will be more apparent from the following description part or understood through the practice of the present application. BRIEF DESCRIPTION OF DRAWINGS

[0099] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0100] Figure 1 The flow chart of the improved bee colony algorithm described in the embodiments of the present application.

[0101] Figure 2 The comparison diagram of the distribution time of the improved bee colony algorithm and the traditional bee colony algorithm under different threshold values alpha. DETAILED DESCRIPTION

[0102] Embodiments of the present application will be described in detail below with reference to the attached drawings, which are given by way of illustration and thus do not limit the present application. In the drawings, like reference numerals refer to like elements throughout.

[0103] As used herein, the terms "have," "has," "have," "having," "include," "includes," "includes," "including," "indude," "indudes," "induding," "indudeing," or the like are used inclusively and therefore will be followed by other elements.

[0104] It is also to be understood that terminology from common dictionaries and the like are to be taken in a manner that is consistent with their usage in the relevant art and that they are not to be interpreted in an idealized or overly formal sense unless expressly so defined herein.

[0105] As used herein, the singular forms "a," "an," and "the" include plural referents unless the context clearly dictates otherwise.

[0106] In the description of the present application, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" and the like means that the particular feature, structure, material or characteristic being described is included in at least one embodiment or example of the present application. Moreover, the description of such a particular feature, structure, material or characteristic as being included in one embodiment or example is not intended to be taken as an exclusion of such feature, structure, material or characteristic from other embodiments or examples of the present application. That is, the particular features, structures, materials, or characteristics described are not exclusive to any one or more embodiments or examples of the present application and can be combined with or added to other embodiments or examples of the present application.

[0107] In order to facilitate the understanding of the present application, the present application will be further explained and described in detail below with reference to the accompanying drawings in specific embodiments, and the specific embodiments do not constitute a limitation on the embodiments of the present application.

[0108] Those skilled in the art should understand that the drawings are only schematic illustrations of the embodiments, and the components in the drawings are not necessarily necessary for the implementation of the present application.

[0109] Embodiment 1

[0110] The present embodiment 1 provides a narrow non-exposed space emergency resource allocation system based on consequence judgment, comprising:

[0111] The determining module is configured to determine a comprehensive weight of an influence factor affecting secondary damage in the narrow non-exposed space;

[0112] The first calculating module is configured to construct a positive ideal point matrix and an inverse ideal point matrix for the narrow non-exposed space where the accident occurs, calculate an ideal point fitting degree of each influence factor, combine the comprehensive weight of each influence factor, and calculate a possibility of secondary disaster in the narrow non-exposed space where the accident occurs;

[0113] The second calculating module is configured to calculate required resources of each affected narrow non-exposed area based on the possibility of secondary damage in the narrow non-exposed area, construct a multi-objective function with the highest time satisfaction, the most fair distribution and the lowest rescue cost, and solve a resource allocation mode by using an improved bee colony algorithm.

[0114] In the embodiment 1, the system is used to realize an emergency resource allocation method for narrow non-exposed space based on consequence research and judgment, which includes the following steps:

[0115] The determining module is configured to determine a comprehensive weight of an influence factor affecting secondary damage in the narrow non-exposed space;

[0116] The first calculating module is configured to construct a positive ideal point matrix and an inverse ideal point matrix for the narrow non-exposed space where the accident occurs, calculate an ideal point fitting degree of each influence factor, combine the comprehensive weight of each influence factor, and calculate a possibility of secondary disaster in the narrow non-exposed space where the accident occurs;

[0117] The second calculating module is configured to calculate required resources of each affected narrow non-exposed area based on the possibility of secondary damage in the narrow non-exposed area, construct a multi-objective function with the highest time satisfaction, the most fair distribution and the lowest rescue cost, and solve a resource allocation mode by using an improved bee colony algorithm.

[0118] The improved bee colony algorithm is to consider different disaster conditions of each affected point when randomly generating initial foraging bees, and initialize the population according to the resource demand ratio when generating initial solutions:

[0119] X ijmax '=X’ ij +β·{r imax ∈R i ,d min}

[0120] Wherein, X ijmax ' represents a newly generated initial foraging bee, β represents an initial resource allocation coefficient of the required resource corresponding to the affected point with the largest required rescue resource amount, r imax represents the largest type of rescue resource required by the affected point i, R i represents the total amount of rescue resources required by the affected point i, and dmin X' represents that the rescue materials are allocated from the rescue center closest to the disaster point for initial resource allocation. ij X represents that the honey bees are randomly generated after initial resource allocation.

[0121] The subjective weight and the objective weight of the influence factors affecting secondary injury in narrow non-exposed space are determined, and the comprehensive weight is calculated by the way of combination weighting of game theory.

[0122] Specifically, when calculating the subjective weight and the objective weight:

[0123] The index sample data is sorted, and the Gini coefficient G of each index x is calculated. i i

[0124]

[0125] Wherein: m represents the sample number; y ip and y iq are the pth and qth sample values corresponding to the index x i , and Y i represents the sum of the sample data of the index x i .

[0126] The least important index x i is selected from the index set X={x t}(i=1, 2,..., n), and the relative importance r i of each index to x t is calculated.

[0127]

[0128] Wherein, G t represents the Gini coefficient of the least important index x t .

[0129] The subjective weight of each evaluation index x i is calculated as:

[0130]

[0131] The index sample data is standardized by using range transformation, and the information content C of the index x i is calculated. i

[0132]

[0133] Wherein, is the index standard deviation coefficient; r ij is the correlation coefficient between indexes; ​​​

[0134] The evaluation index x is calculated i The objective weight is:

[0135]

[0136] The combination coefficient κ is calculated k The optimization strategy model is obtained:

[0137]

[0138] The condition for converting to the optimal first derivative is:

[0139]

[0140] The combination coefficient κ is calculated k , k = 1, 2,..., L, and normalized:

[0141]

[0142] The evaluation index x is calculated i The optimal comprehensive weight is:

[0143]

[0144] Assume that the risk evaluation index of a specific secondary accident in a narrow non-exposed space is λ ij , i = 1, 2,..., m; j = 1, 2,..., n, where m is the number of regions to be evaluated, n is the number of evaluation indexes, and the evaluation matrix is divided into three aspects of factors, which are environmental instability factors, factors affecting rescue timeliness, and inducing factors:

[0145]

[0146] Positive indexes and negative indexes are used to evaluate the possibility of secondary disasters in narrow non-exposed areas; among them, the positive indexes are positively correlated with the possibility of secondary accidents, and the negative indexes are the opposite;

[0147] Assuming that the possibility of secondary accidents changes in a monotonic trend, the positive ideal point and the negative ideal point can be determined;

[0148] When the evaluation index is a positive index, the positive ideal point and the negative ideal point vectors are:

[0149]

[0150] When the evaluation index is a negative index, the positive ideal point and the negative ideal point vectors are:

[0151]

[0152] wherein f i (p), f i (n) are the positive ideal point vector and the negative ideal point vector of the i-th index of the specific secondary accident risk in the narrow non-exposed space, respectively, λ i The risk evaluation index value of the specific secondary accident in the narrow non-exposed space;

[0153] The ideal point evaluation function is the distance between the index and the ideal point and the negative ideal point, and the relative distance between the index and the positive ideal point and the negative ideal point is expressed by using the Euclidean distance;

[0154] The distance to the positive ideal point:

[0155]

[0156] The distance to the negative ideal point:

[0157]

[0158] Calculate the ideal point sticking degree:

[0159]

[0160] The probability of a secondary accident caused by a harm occurring in a certain narrow non-exposed space area is defined as:

[0161]

[0162] According to the size of the determined possibility of causing a secondary accident, it is judged whether the influence of the secondary accident is considered when resources are allocated to the narrow non-exposed area i:

[0163]

[0164] wherein E r is used to judge whether a certain narrow non-exposed area i needs to consider the secondary accident, 1 means considering the secondary accident, 0 means not considering the secondary accident, p i represents the possibility of the narrow non-exposed area i being affected by the secondary accident, and α is the threshold value for determining whether the secondary accident needs to be considered;

[0165] Then the material demand of the disaster point is:

[0166] R i = r i1 · I ti + r i2 · E ri · I ti

[0167] wherein R i represents the material needed by the disaster point i, r i1r represents the average amount of supplies needed per person in a first-aid case, r i2 E represents the average amount of supplies needed per person in a second-aid case, E ri I represents a second-aid coefficient of the disaster point i, I ti N represents the number of trapped people at the disaster point i at time t;

[0168] Assuming that the current decision time is t, there are n disaster points and m rescue centers, due to the different number of trapped people in each disaster point, in order to solve the material allocation problem, the materials are as evenly distributed as possible to each disaster point, and an emergency weight coefficient w i is given to each disaster point, which represents the emergency degree of the disaster point i, and the calculation formula is

[0169]

[0170] Based on the emergency degree, the value function represented by the rescue time deviation degree is used to measure the satisfaction of trapped personnel to the rescue time:

[0171]

[0172] Wherein, T i represents the rescue event deviation degree of the disaster point i; t ij represents the generalized event distance between the rescue center j and the disaster point i; J represents the rescue center set; X ij represents a 0 / 1 variable, which is 1 if rescue is performed, otherwise it is 0;

[0173] The time satisfaction function is:

[0174]

[0175] In the formula, since the slope of the value function changes, it is a concave function in the gain interval, a is less than or equal to 1, and it is a convex function in the loss interval, b is less than or equal to 1; according to the risk aversion principle, the function in the loss interval is steeper than that in the gain interval,

[0176]

[0177] Wherein, F1 represents one of the target functions of time satisfaction, and the greater the time satisfaction is, the better;

[0178] In order to make the allocation more fair and reasonable, the variance of the material satisfaction degree of all disaster points is used to measure the fairness of the rescue; then:

[0179]

[0180]

[0181]

[0182] wherein, n ij represents the amount of supplies sent by the rescue center j to the disaster point i; P i represents the degree of satisfaction of the disaster point i with supplies, F1 is one of the objective functions, and the smaller the value, the better the allocation. represents the average degree of satisfaction of all disaster points with supplies, and F2 is one of the objective functions, and the smaller the variance, the smaller the difference, i.e., the more equitable the allocation.

[0183] The cost function is defined as follows:

[0184]

[0185] wherein, F3 represents one of the objective functions, and the smaller the cost, the better, and C ij represents the cost of sending a unit of supplies from the rescue center j to the disaster point i.

[0186] In the allocation of rescue resources, the allocation of supplies is performed according to the following objective functions:

[0187]

[0188]

[0189]

[0190]

[0191]

[0192] X ij = 0 / 1

[0193] if X ij = 0 → n ij = 0

[0194] if X ij = 1 → 0 < n ij < min{R i , G j}

[0195] wherein, G j represents the amount of supplies in the rescue center, and the allocation of supplies to each rescue center will not exceed the demand of the disaster point, and similarly, the demand of each disaster point will not exceed the maximum storage capacity of the rescue center.

[0196] Example 2

[0197] In this embodiment 2, a method for allocating emergency resources in narrow non-exposed space based on consequence analysis is provided, which determines the weight of influencing factors affecting the possibility of secondary disasters in narrow non-exposed space by using the method of combination weighting of game theory and performs probability estimation, calculates the resources required by each narrow non-exposed space, and solves by comparing the improved bee colony algorithm with other heuristic algorithms to form the optimal resource allocation scheme, the method comprises the following steps:

[0198] Step one, determine the subjective weight and objective weight of the influencing factors affecting the secondary damage in narrow non-exposed space, and calculate the weight by using the method of combination weighting of game theory.

[0199] Step two, construct the positive ideal point matrix and inverse ideal point matrix for the narrow non-exposed space where the accident occurs, and calculate the ideal point sticking degree of each influencing factor, and calculate the possibility of secondary disasters in the narrow non-exposed space where the accident occurs by collecting the weight of each influencing factor calculated in step one.

[0200] Step three, based on the possibility of secondary damage in narrow non-exposed area in step two, calculate the materials required by each affected narrow non-exposed area, and construct a multi-objective function with the highest time satisfaction, the most fair distribution and the lowest rescue cost, and use the improved bee colony algorithm to solve the resource allocation method.

[0201] The specific form of the first step is:

[0202] 2.1 Sort the index sample data, calculate the Gini coefficient G of each index x i : i

[0203]

[0204] Where: m represents the number of samples; y ip , y iq are the p, q sample values corresponding to the index x i , and Y i represents the sum of the sample data of the index x i .

[0205] Select the least important index x i from the index set X={x t}(i=1,2,...,n), calculate the relative importance r i of each index to x t :

[0206]

[0207] Where, G t represents the Gini coefficient of the least important index x t .​

[0208] Calculate each evaluation index x i Subjective weighting:

[0209]

[0210] 2.2 Standardize the indicator sample data using range scaling and calculate the indicator x. i Information content C i :

[0211]

[0212] in, r is the coefficient of standard deviation of the indicator. ij This represents the correlation coefficient between the indicators.

[0213] Calculate the evaluation index x i Objective weighting:

[0214]

[0215] 2.3 Combine the coefficients κ k The optimized strategy model is as follows:

[0216]

[0217] The condition for transforming into the optimal first derivative is:

[0218]

[0219] κ is calculated using the above formula. k k = 1, 2, ..., L, and then normalize them:

[0220]

[0221] Calculate the evaluation index x i Optimal overall weight:

[0222]

[0223] The specific steps of step two are as follows:

[0224] 3.1 Assume that the risk assessment index for a specific secondary accident occurring in a confined, non-exposed space is λ. ij Let i = 1, 2, ..., m; j = 1, 2, ..., n, where m is the number of areas to be evaluated and n is the number of indicators to be evaluated. The evaluation is divided into three aspects: unstable environmental factors, factors affecting rescue timeliness, and inducing factors. The evaluation matrix is ​​as follows:

[0225]

[0226] In this embodiment, we use positive indicators and negative indicators to evaluate the possibility of secondary disasters in narrow non-exposed areas. Among them, the positive indicators are positively correlated with the possibility of secondary accidents, and the negative indicators are the opposite. Assuming that the possibility of secondary accidents changes monotonically, the positive ideal point and the negative ideal point can be determined.

[0227] When the evaluation index is a positive index, the positive ideal point and the negative ideal point vector are:

[0228]

[0229] When the evaluation index is a negative index, the positive ideal point and the negative ideal point vector are:

[0230]

[0231] Among them, f i (p), f i (n) are the positive ideal point vector and the negative ideal point vector of the i-th index of the risk of secondary accidents in narrow non-exposed spaces, respectively, λ i The risk evaluation index value of the secondary accident in the narrow non-exposed space.

[0232] 3.2 The ideal point evaluation function is the distance from the index to the ideal point and the negative ideal point. In this embodiment, we use the Euclidean distance to represent the relative distance between the index and the positive ideal point and the negative ideal point.

[0233] Distance to positive ideal point:

[0234]

[0235] Distance to negative ideal point:

[0236]

[0237] Calculate the ideal point sticking degree:

[0238]

[0239] 3.3 In combination with various factors that affect the possibility of secondary accidents in narrow non-exposed areas, in this embodiment, we define the probability of secondary accidents caused by injuries occurring in a certain narrow non-exposed space area as:

[0240]

[0241] The specific steps of step three are:

[0242] 4.1 After the occurrence of a damage accident in a narrow non-exposed area, it is necessary to quickly determine the possibility of a secondary accident caused by a primary damage accident, and determine the amount of resources needed to be dispatched to each disaster point according to the size of the possibility and the trapped situation of the personnel. In this embodiment, we determine whether to consider the impact of secondary accidents when allocating resources to narrow non-exposed area i according to the size of the possibility of causing secondary accidents determined before:

[0243]

[0244] wherein E r In order to determine whether a narrow non-exposed area i needs to consider secondary accidents, 1 means considering secondary accidents, 0 means not considering secondary accidents, p i represents the possibility of narrow non-exposed area i being affected by secondary accidents, and a is the threshold value for determining whether to consider secondary accidents.

[0245] The material demand of the disaster point is:

[0246] R i = r i1 · I ti + r i2 · E ri · I ti

[0247] wherein R i represents the material needed by disaster point i, r i1 represents the average amount of material needed by each person in a primary damage, r i2 represents the average amount of material needed by each person in a secondary accident, E ri represents the secondary accident coefficient of disaster point i.

[0248] 4.2 Assuming that the current decision-making time is t, there are n disaster points and m rescue centers. Due to the different number of trapped personnel in each disaster point, in order to solve this kind of material allocation problem and make the material as fair as possible to each disaster point, an emergency weight coefficient w i is given to each disaster point, which represents the emergency degree of disaster point i, and the calculation formula is

[0249]

[0250] Based on the emergency degree, the value function represented by the rescue time deviation degree is used to measure the satisfaction of trapped personnel to the rescue time.

[0251]

[0252] wherein T i represents the rescue time deviation degree of disaster point i; t ijDij represents the generalized event distance between the rescue center j and the disaster point i; J represents the rescue center set; X ij Dij represents the generalized event distance between the rescue center j and the disaster point i; J represents the rescue center set; X

[0253] The time satisfaction function is as follows:

[0254]

[0255] In the formula, since the slope of the value function changes, it is a concave function in the income interval, a is less than or equal to 1, and it is a convex function in the loss interval, b is less than or equal to 1. According to the risk aversion principle, the function in the loss interval is steeper than that in the income interval,

[0256]

[0257] In the formula, F1 represents the time satisfaction of one of the target functions, and the greater the time satisfaction is, the better.

[0258] In order to make the allocation more fair and reasonable, the variance of the material satisfaction degree of all disaster points is used to measure the fairness of the rescue.

[0259]

[0260]

[0261]

[0262] In the formula, n ij Dij represents the generalized event distance between the rescue center j and the disaster point i; J represents the rescue center set; X i Dij represents the generalized event distance between the rescue center j and the disaster point i; J represents the rescue center set; X Dij represents the generalized event distance between the rescue center j and the disaster point i; J represents the rescue center set; X

[0263] When the rescue resource allocation is performed, the cost cannot be completely ignored, and in the present application, the cost function is defined as follows.

[0264]

[0265] In the formula, F3 represents one of the target functions, and the smaller the cost is, the better; C ij Dij represents the generalized event distance between the rescue center j and the disaster point i; J represents the rescue center set; X

[0266] When the rescue resource allocation is performed, the cost cannot be completely ignored, and in the present application, the cost function is defined as follows.

[0267]

[0268]

[0269]

[0270]

[0271]

[0272] X ij = 0 / 1

[0273] if X ij = 0 -> n ij = 0

[0274] if X ij = 1 -> 0 < n ij < min{R i , G j}

[0275] where G j denotes the reserve of the relief center, the allocation of the relief center will not exceed the demand of the disaster point, similarly, the demand of the disaster point will not exceed the maximum storage capacity of the relief center.

[0276] 4.3 The standard ABC algorithm divides the artificial bee colony into three categories: employed bees, onlooker bees and scout bees by simulating the foraging mechanism of real bees. The goal of the whole colony is to find the flower with the largest amount of nectar. In the standard ABC algorithm, the employed bees use the previous information of the flower to find new flowers and share the information with onlooker bees; the onlooker bees wait in the hive and find new flowers according to the information shared by the employed bees; the task of the scout bees is to find a new valuable flower, they randomly search for flowers near the hive. Therefore, the algorithm is divided into three parts.

[0277] Assume that the solution space of the problem is D-dimensional, the number of employed bees and onlooker bees is S, the number of employed bees or onlooker bees is equal to the number of flowers. The standard ABC algorithm regards the solving process of the optimization problem as searching in the D-dimensional search space.

[0278] The position of each flower represents a possible solution of the problem, and the amount of nectar of the flower corresponds to the fitness of the corresponding solution. An employed bee corresponds to a flower. The original bee colony algorithm uses a completely random method when forming the initial employed bees, in this embodiment, when generating the initial employed bees, we consider that the disaster conditions of each disaster point are different, and we perform solution pre-allocation when generating the initial solution, which can significantly improve the search ability of the bee colony algorithm.

[0279] X ijmax = X i ' j + β · {rimax ∈R imax ,d min}

[0280] wherein, X ijmax ' represents newly generated initial scout bees, β represents a pre-allocated coefficient of the required material corresponding to the disaster point with the largest required amount of rescue material, r imax represents the largest type of rescue material required by the disaster point i, R i represents the total amount of rescue material required by the disaster point i, d min represents that the rescue material is pre-allocated from the rescue center closest to the disaster point, X' ij represents the scout bees randomly generated after pre-allocation.

[0281] The model in step two is solved by the improved bee colony algorithm to obtain the resource allocation scheme of each rescue center.

[0282] Embodiment 3

[0283] In this embodiment 3, a secondary collapse accident in the process of railway tunnel construction is taken as an example to calculate the combined weight proportion of various subjective factors and objective factors. The tunnel is about 2 km long, with a single slope in the tunnel hole, the maximum burial depth is about 105 m, the landform is hilly, the terrain is large, there are many "U" and "V" shaped gullies between hills, and there are many streams in the valleys, the hilly surface is covered with silty clay, the vegetation grows luxuriantly, the slope is generally gentle, the average is 20 degrees, and there are a few steep places reaching more than 30 degrees, the altitude is 185 to 290 meters.

[0284] The surrounding rock grade F1, joint surface spacing F2, bias angle F3, maximum water inflow per unit length F4, tunnel span F5, tunnel burial depth F6, geological survey depth F7, disturbance of construction to surrounding rock F8, construction technology level F9, construction management level F 10 These 10 factors are used as evaluation factors for the risk evaluation of secondary collapse of the tunnel section. The tunnel entrance section, exit section, and middle section are selected as sampling points, and the actual data of the sample evaluation indexes are shown in Table 1.

[0285] Table 1

[0286]

[0287] Based on the above data, the G2 method improved by Gini coefficient is used to calculate the subjective weight of the evaluation index, the CRITIC method improved by standard deviation coefficient is used to calculate the objective weight of the evaluation index, and the game theory combined weighting method is used to obtain the comprehensive weight of each index. The calculation results are shown in Table 2.

[0288] Table 2

[0289]

[0290] According to the engineering data and relevant engineering experience, the classification standards of the selected 10 secondary collapse influencing factors are shown in Table 3, and the reference standards of surrounding rock grade F1, construction disturbance to surrounding rock F8, construction technical level F9, construction management level F 10 The reference standards are shown in Table 3.

[0291] Table 3

[0292]

[0293]

[0294] The classification method of some factors is shown in Table 4:

[0295] Table 4

[0296]

[0297] The positive ideal point matrix and the inverse ideal point matrix are constructed as follows:

[0298]

[0299]

[0300] Taking a serious injury accident in the construction process of a railway tunnel as an example, this injury accident caused 5 collapses, and the tunnel risk index value and the initial number of trapped people are shown in Table 5.

[0301] Table 5

[0302]

[0303]

[0304] The ideal point sticking degree of the secondary collapse risk grade of each tunnel section and the probability p of secondary injury are calculated i as shown in Table 6.

[0305] Table 6

[0306]

[0307] According to the possibility of secondary injury in each tunnel interval calculated, the threshold value α of judging secondary injury is compared, here we let α equal to 0.520, and determine the required quantity of various rescue materials of each disaster point, as shown in Table 7.

[0308] Table 7

[0309]

[0310] After determining the initial required rescue materials, the traditional bee colony algorithm, genetic algorithm, particle swarm algorithm and improved bee colony algorithm are used to compare the allocation results of the rescue materials, and the flowchart of the improved bee colony algorithm is as shown in Figure 1

[0311] The target function is normalized to obtain the results shown in Table 8:

[0312] Table 8

[0313]

[0314]

[0315] In this embodiment, by analyzing the sudden disaster of narrow non-exposed space, the method of game theory combination weighting is used to weight the subjective and objective factors affecting the secondary accidents in narrow non-exposed space, and the positive ideal point and inverse ideal point set are constructed according to these factors, the ideal point sticking degree is calculated, and the possibility of secondary injury in narrow non-exposed space is calculated combined with the weight of each influencing factor. Under the condition of considering the influence of secondary injury, the multi-objective function is constructed with the highest time satisfaction, the most fair distribution and the minimum cost, and the improved bee colony algorithm is used for solving, and compared with other solving algorithms, the final allocation scheme is obtained. At the same time, the decision maker can adjust the threshold of secondary injury according to the actual situation, the improved bee colony algorithm is not only optimal in the allocation scheme, but also improves the time efficiency. Figure 2 It can be found (in the figure, the lower column represents the improved bee colony algorithm) that the time efficiency of the algorithm is also rapidly improved.

[0316] Embodiment 4

[0317] Embodiment 4 of the present application provides a non-transitory computer readable storage medium for storing computer instructions, which, when executed by a processor, implements a narrow non-exposed space emergency resource allocation method based on consequence research and judgment.

[0318] Embodiment 5

[0319] Embodiment 5 of the present application provides a computer program (product) comprising a computer program for implementing a narrow non-exposed space emergency resource allocation method based on consequence research and judgment when running on one or more processors.

[0320] Embodiment 6

[0321] ​Embodiment 6 of the present application provides an electronic device, comprising: a processor, a memory and a computer program; wherein the processor is connected with the memory, and the computer program is stored in the memory; when the electronic device is running, the processor executes the computer program stored in the memory, so that the electronic device executes instructions for realizing the method for allocating emergency resources in a narrow non-exposed space based on consequence judgment.

[0322] Those skilled in the art will understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0323] The present application is described with reference to flowcharts and / or block diagrams of the method, device (system), and computer program product according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus generate a means for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 means for performing the functions specified in the flowchart

[0324] These computer program instructions can also be stored in a computer-readable memory capable of guiding a computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer-readable memory produce a product including instruction means, which implements the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 means for performing the functions specified in the flowchart

[0325] These computer program instructions can also be loaded into a computer or other programmable data processing apparatus to execute a series of operation steps to produce a computer-implemented process, so that the instructions executed on the computer or other programmable data processing apparatus provide a means for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 means for performing the functions specified in the flowchart

[0326] The above describes the specific embodiments of the present application in combination with the drawings, but is not a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications or variations made by those skilled in the art on the basis of the disclosed technical solutions of the present application, without creative labor, should be covered in the protection scope of the present application.

Claims

1. A method for allocating emergency resources in a narrow non-exposed space based on consequence analysis, characterized in that, The method comprises the following steps: Determine the comprehensive weight of the influencing factors of secondary damage in the narrow non-exposed space; Construct the positive ideal point matrix and the inverse ideal point matrix of the narrow non-exposed space where the accident occurs, and calculate the ideal point sticking degree of each influencing factor, combine the comprehensive weight of each influencing factor, and calculate the possibility of secondary disaster in the narrow non-exposed space where the accident occurs; Based on the possibility of secondary damage in the narrow non-exposed area, calculate the material demand of each disaster narrow non-exposed area, and construct a multi-objective function with the highest time satisfaction, the most fair distribution and the lowest rescue cost, and use the improved bee colony algorithm to solve the resource allocation mode; wherein, The improved bee colony algorithm is to consider that the disaster conditions of each disaster point are different when generating the initial solution, and to initialize the population according to the resource demand ratio when generating the initial solution: X ijmax ' = X' ij + β · {r imax ∈ R i ,d min} wherein, X ijmax represents newly generated initial scout bees, β represents an initial resource allocation coefficient of the required resources corresponding to the disaster point with the largest required amount of rescue materials, r imax represents the largest type of rescue materials required by the disaster point i, R i represents the total amount of rescue materials required by the disaster point i, d min represents that the rescue materials are allocated from the rescue center closest to the disaster point for initial resource allocation, X′ ij represents randomly generated scout bees after initial resource allocation.

2. The method of claim 1, wherein, Determine the subjective weight and objective weight of the influencing factors of secondary damage in the narrow non-exposed space, and calculate the comprehensive weight by using the game theory combination weighting method.

3. The method of claim 2, wherein, Sort the index sample data, calculate the Gini coefficient G of each index x i i :​ where: m represents the number of samples; y ip , y iq is the p, qth sample value of the index x i ; Y i represents the sum of the sample data of the index x i ; From the index set X = {x i Select the least important index x from} (i = 1, 2, ..., n) t Calculate the relationship between each index and x. t The relative importance of r i : where G t represents the Gini coefficient of the least important indicator x t . The subjective weight of each evaluation index x i is calculated as follows: The index sample data is standardized by using the range and the index x is calculated i The information content C i : wherein is the index standard deviation coefficient; r ij is the inter-index correlation coefficient; The evaluation index x is calculated i The objective weight is:

4. The method of claim 3, wherein, The combined coefficient K is calculated as follows: k The optimization obtains a countermeasure model: The condition for converting to the optimal first derivative is: Combining the above formula, we get κ k k = 1, 2,..., L, and normalize it: The evaluation index x is calculated i The optimal comprehensive weight is:

5. The method of claim 4, wherein, Assume that the risk evaluation index of a certain narrow non-exposed space for a specific secondary accident is λ ij i = 1, 2,..., m; j = 1, 2,.., n, where m is the number of areas to be evaluated, n is the number of indexes to be evaluated, which is divided into three aspects of factors, which are environmental instability factors, factors affecting rescue timeliness and inducing factors, and the evaluation matrix is: The positive index and the inverse index are used to evaluate the possibility of secondary disaster in the narrow non-exposed area; wherein, the positive index is positively correlated with the possibility of secondary accident, and the inverse index is opposite; Assuming that the possibility of secondary accident changes in a monotonous trend, the positive ideal point and the inverse ideal point can be determined; When the evaluation index is a positive index, the positive ideal point and the inverse ideal point vector are: When the evaluation index is an inverse index, the positive ideal point and the inverse ideal point vector are: wherein f i (p), f i (n) are the positive ideal point vector and the negative ideal point vector of the i-th index of the specific secondary accident risk in the narrow non-exposed space, respectively, λ i the risk evaluation index value of the specific secondary accident in the narrow non-exposed space; The ideal point evaluation function is the distance between the index and the ideal point and the inverse ideal point, and the Euclidean distance is used to represent the relative distance between the index and the positive ideal point and the inverse ideal point; The distance to the positive ideal point is: The distance to the inverse ideal point is: Calculate the ideal point sticking degree: The probability of damage occurring in a narrow non-exposed space area triggering a secondary accident is defined as:

6. The method of claim 5, wherein, According to the size of the determined possibility of triggering a secondary accident, it is judged whether the influence of the secondary accident is considered when the resources of the narrow non-exposed area i are allocated: wherein E r is a judgment of whether a secondary accident needs to be considered for a narrow non-exposed area i, 1 means that a secondary accident needs to be considered, 0 means that a secondary accident does not need to be considered, p i represents the possibility of a secondary accident for the narrow non-exposed area i, and a is a threshold value for judging whether a secondary accident needs to be considered. The material demand of the disaster point is: R i = r i1 • I ti + r i2 • E ri • I ti where R i represents the amount of resources needed at disaster point i, r i1 represents the average amount of resources needed per person in a first-incident, r i2 represents the average amount of resources needed per person in a second-incident, E ri represents the second-incident coefficient of disaster point i, I ti represents the number of trapped people at disaster point i at time t; Assuming the current decision-making time is t, there are n disaster sites and m rescue centers. Since the number of people trapped varies at each disaster site, to solve the resource allocation problem and ensure fair distribution of resources to each disaster site, an emergency weight coefficient w is assigned to each disaster site. i , representing the urgency of disaster point i, is calculated using the following formula: Based on the emergency degree, the value function represented by the rescue time deviation degree is used to measure the satisfaction degree of the trapped personnel to the rescue time: where T i represents the rescue event deviation degree of disaster point i; t ij represents the generalized event distance between rescue center j and disaster point i; J represents a rescue center set; X ij represents a 0 / 1 variable, which is 1 if rescue is performed and 0 otherwise; The time satisfaction function is: In the formula, since the slope of the value function changes, in the gain interval it is a concave function, a is less than or equal to 1, in the loss interval it is a convex function, b is less than or equal to 1; according to the risk-averse principle, the function in the loss interval is steeper than that in the gain interval, Wherein, F1 represents the time satisfaction of one of the target functions, and the larger the time satisfaction is, the better it is; In order to make the allocation more fair and reasonable, the rescue fairness is measured by the variance of the material satisfaction degree of all disaster points; then: Wherein, n ij represents the amount of supplies sent by the rescue center j to the disaster point i; P i represents the degree of satisfaction of the supplies at the disaster point i, represents the average degree of satisfaction of the supplies at all disaster points, and F2 is the variance of one of the objective functions, and the smaller the variance, the smaller the difference, that is, the more equitable the distribution. The cost function is defined as follows: where F3 represents the cost of one of the objective functions, the smaller the better, C ij represents the cost of transporting a unit of goods from the rescue center j to the disaster point i; When the rescue resources are allocated, the materials are allocated according to the following target function; X ij ＝0 / 1 if X ij = 0 → n ij = 0 if X ij = 1 -> 0 < n ij < min{R i , G j} where G j represents the amount of supplies in the rescue center, the amount of supplies allocated to each rescue center will not exceed the demand of the disaster point, and similarly, the demand of each disaster point will not exceed the maximum storage capacity of the rescue center.

7. A consequence-based judgment narrow non-exposed space emergency resource allocation system, characterized in that, The method comprises the following steps: The determination module is used to determine the comprehensive weight of the influencing factors of secondary damage in the narrow non-exposed space; The first calculation module is used to construct the positive ideal point matrix and the inverse ideal point matrix of the narrow non-exposed space where the accident occurs, and calculate the ideal point sticking degree of each influencing factor, combine the comprehensive weight of each influencing factor, and calculate the possibility of secondary disaster in the narrow non-exposed space where the accident occurs; The second calculation module is configured to calculate the required resources of each affected narrow non-exposed area based on the possibility of secondary injury in the narrow non-exposed area, and construct a multi-objective function with the highest time satisfaction, the most fair distribution and the lowest rescue cost, and solve the resource allocation mode by using an improved bee colony algorithm. In the improved bee colony algorithm, when randomly generating initial foraging bees, the disaster conditions of each disaster point are considered, and the initial population is initialized according to the resource demand ratio when generating an initial solution. The non-transitory computer readable storage medium is configured to store computer instructions, and the computer instructions are executed by the processor to implement the method for allocating emergency resources in narrow non-exposed space based on consequence analysis according to any one of claims 1-6. X ijmax ' = X' ij + β · {r imax ∈ R i , d min} wherein X ijmax represents newly generated initial scout bees, β represents an initial resource allocation coefficient for the required resources corresponding to the disaster point requiring the largest amount of rescue materials, r imax represents the largest type of rescue materials required by the disaster point i, R i represents the total amount of rescue materials required by the disaster point i, d min represents that the rescue materials are allocated from the rescue center closest to the disaster point after initial resource allocation, X′ ij represents randomly generated scout bees after initial resource allocation.

8. A non-transitory computer-readable storage medium, comprising: The computer program is configured to implement the method for allocating emergency resources in narrow non-exposed space based on consequence analysis according to any one of claims 1-6 when running on one or more processors.

9. A computer program product, characterised in that, The computer program is configured to implement the method for allocating emergency resources in narrow non-exposed space based on consequence analysis according to any one of claims 1-6 when running on one or more processors.

10. An electronic device, comprising: The computer program is configured to implement the method for allocating emergency resources in narrow non-exposed space based on consequence analysis according to any one of claims 1-6 when running on one or more processors. The computer program is configured to implement the method for allocating emergency resources in narrow non-exposed space based on consequence analysis according to any one of claims 1-6 when running on one or more processors.

Images