CN101763467A - Method for optimizing protection capability in safety protection system - Google Patents

Abstract

The invention relates to the technical field of safety protection, in particular to a method for optimizing a protection capability in a safety protection system. An optimized model of the protection capability of a safety protection system is established in an abstract graph of the safety protection system, and a protection capability value of the node is improved or reduced by adjusting protecting facilities of all nodes, thereby obtaining a maximal attack path in the safety protection system and an optimal value of the protection capability. The cost on leading an attacker to attack a protected object along the path is maximal, and an average risk value of the safety protection system is reduced to the minimum. The invention can more scientifically and effectively guide the reasonable distribution and dispatch of the protecting facilities, enable the protection capability of the whole safety protection system to be fully utilized and has great practical significance for planning and distributing the safety protection facilities.

Description

The optimization method of protective capacities in the safety-protection system

Technical field

The present invention relates to the security precautions technology field, relate in particular to the optimization method of protective capacities in a kind of safety-protection system.

Background technology

Safety precaution is the part of social public security, and with regard to risk prevention instruments, safety precaution is divided into manpower and takes precautions against (people's air defense), the in kind strick precaution (thing is anti-) and technological prevention (technical precaution).Wherein manpower strick precaution and in kind the strick precaution are traditional risk prevention instrumentses.Along with continuous progress in science and technology, technological prevention forms a kind of novel risk prevention instruments gradually.

Safety-protection system normally is made of jointly people, material object or technical precaution equipment, can be divided into a plurality of nodes.So-called node is meant the protective unit of people, protective equipment or technical precaution equipment with safe precaution ability.Security system is obtaining positive effect as the important technological platform of public safety system aspect the social security prevention and control, but has also exposed some problems: be lower than the desired protection threshold value in this zone as some node owing to protective capacities; Cause this regional risk to increase; Also some node causes this regional protective equipment a large amount of redundancies to occur because protective capacities is higher than the desired protection threshold value in this zone, has caused the serious waste of resource; In order can the safety-protection system that node is formed reasonably to be distributed and scheduling resource, need the protective capacities of safety-protection system be optimized, set up corresponding protective capacities Optimization Model, instruct the distribution and the scheduling of resource theoretically, make the average risk value of whole safety-protection system drop to minimum.

At present, research aspect the protective capacities optimization of security system both at home and abroad also lacks very much, mainly concentrate on the layout of optimization, emergency resources of optimization, the city fire resistance property of the logistics capability in the economics and optimization etc. for ability optimization aspect, proposed corresponding Optimization Model.But the discussion to optimisation strategy is less.Therefore, need the optimization method of invention security system protective capacities urgently, set up the model that the safety-protection system protective capacities is optimized.

Summary of the invention

The optimization method that the purpose of this invention is to provide protective capacities in a kind of safety-protection system, optimize the reasonable distribution and the scheduling of protective equipment by science and effective method, it is minimum that the value-at-risk of whole safety-protection system is dropped to, to improve and optimizate the protective capacities of safety-protection system.

For achieving the above object, the present invention adopts following technical scheme:

Set up the abstract graph of security system;

Set up the Optimization Model of safety-protection system protective capacities;

The Optimization Model of safety-protection system protective capacities is found the solution in calculating, and this step further comprises following substep:

1. according to the security system abstract graph,, obtain the set M of all attack paths from the point of attack to the protected object by enumeration methodology;

2. respectively below set M, utilize the Optimization Model of protective capacities, adjust the protective capacities value of node, protective capacities to safety-protection system is optimized, draw under different attack paths the maximal value of safety-protection system protective capacities, and the optimized protection ability value of each node;

3. the protective capacities value that compares the safety-protection system under all attack paths respectively, obtain the protective capacities maximal value of safety-protection system, simultaneously this path is designated as maximum attack path, when the assailant prevents protected object along this path attack, the cost maximum that need pay;

4. the weight of considering every paths equates, the objective function of different paths correspondence is weighted on average, and multi-objective optimization question is become the single goal optimization problem, and this objective function is optimized, and obtains overall optimum solution.

The described Optimization Model step of setting up the safety-protection system protective capacities comprises:

Regard whole abstract security protection network as a network chart, be designated as: D=(V, A), V represents the set of all security protection nodes in the safety-protection system, and A represents the set on all limits in the safety-protection system, for each bar limit (vi, vj) ∈ A supposes that the protective capacities value of each node is respectively: x ₁, x ₂... ..x _n, the weight on every limit is designated as c (v _i, v _j) 〉=0, then the Optimization Model of safety-protection system protective capacities is:

$f=\max \underset{\underset{n\∈V}{(v_i,v_j)\∈A}}{\mathrm{\Σ}}{c}_{\mathrm{ij}}{x}_n$

$\left\{\begin{array}{c}{x}_1+x_2+...+x_n=S\\ 0\≤x_1\≤m_1\\ 0\≤x_2\≤m_2\\ ..........\\ 0\≤x_n\≤m_n\\ 0\≤c_{\mathrm{ij}}\≤1\end{array}\right.$

Wherein S is known, expression whole node protective capacities sum; m ₁, m ₂... ..m _nBe known, represent the threshold value of the protective capacities of each node.

The present invention has the following advantages and good effect:

1) science and effectively instruct the reasonable distribution and the scheduling of protective equipment more is fully utilized the protective capacities of whole safety-protection system;

2) improved and optimizated the protective capacities of safety-protection system, the planning and the layout of security protection facility had great practice significance.

Description of drawings

Fig. 1 is the abstract graph of security system among the present invention.

Embodiment

In the Combinatorial Optimization theory, the least cost flow model is a model the most basic in the all-network flow problem.This problem can be explained like this: we will find a minimum traffic program of cost that certain goods is sent to the top out by a network from the supply place, and make it to satisfy the demand at these places, summit.In security system, exist a similar Optimization Model equally, this Optimization Model problem to be solved can be expressed as: we are by the scheduling of resource between the node, to in safety-protection system, find the attack path of a protective capacities maximum, and need make the average risk value of whole safety-protection system drop to minimum.

The optimization method of protective capacities in the safety-protection system that the present invention proposes, concrete technical scheme is for may further comprise the steps:

1, sets up the abstract graph of security system.

This method is abstracted into a security protection network chart with security system, referring to the abstract graph of security system shown in Figure 1.

Suppose in the design: the assailant is target from outside the venue with the protection object, has a paths from the position of assault to the protection object at least; All nodes all have protective capacities on the path, and the protective capacities value is to calculate by risk entropy; Every there is certain protection weight on the limit, by given in advance.Weight is big more, and representative is by the difficult more attack protected object of this paths.The assailant needs to pay certain cost by limit on the network and node.

2, set up the Optimization Model of safety-protection system protective capacities, concrete operations are as follows:

Regard whole abstract security protection network as a network chart, be designated as: D=(V, A).V represents the set of all security protection nodes in the safety-protection system.A represents the set on all limits in the safety-protection system.(vi, vj) ∈ A suppose that the protective capacities value of each node is respectively: x for each bar limit ₁, x ₂... ..x _nThe weight on every limit is designated as c (v _i, v _j(brief note is c in) 〉=0 _Ij).Then the Optimization Model of safety-protection system protective capacities is:

$f=\max \underset{\underset{n\∈V}{(v_i,v_j)}\∈A}{\mathrm{\Σ}}{c}_{\mathrm{ij}}{x}_n$

$\left\{\begin{array}{c}{x}_1+x_2+...+x_n=S\\ 0\≤x_1\≤m_1\\ 0\≤x_2\≤m_2\\ ..........\\ 0\≤x_n\≤m_n\\ 0\≤c_{\mathrm{ij}}\≤1\end{array}\right.$

Wherein S is known, expression whole node protective capacities sum; m ₁, m ₂... ..m _nBe known, represent the threshold value of the protective capacities of each node.

3, calculate the Optimization Model of finding the solution the safety-protection system protective capacities, concrete substep is as follows:

1. according to the security system abstract graph,, obtain the set M of all attack paths from the point of attack to the protected object by enumeration methodology.Be designated as M={ph ₁, ph ₂... .ph _n.Ph wherein ₁, ph ₂... .ph _nThe expression attack path.

2. respectively below set M, utilize the Optimization Model of protective capacities, adjust the protective capacities value of node, protective capacities to safety-protection system is optimized, draw under different attack paths the maximal value of safety-protection system protective capacities, and the optimized protection ability value of each node.Because corresponding each bar attack path, the Optimization Model of protective capacities is a typical linear programming model, and finding the solution of Optimization Model can adopt simplicial method to calculate.

3. the protective capacities value that compares the safety-protection system under all attack paths respectively; obtain the protective capacities maximal value of safety-protection system; simultaneously this path is designated as maximum attack path, when the assailant prevents protected object along this path attack, the cost maximum that need pay.4. the weight of considering every paths equates, objective function to different paths correspondence is weighted on average, and multi-objective optimization question is become the single goal optimization problem, and this objective function is optimized, obtain the optimum solution of the overall situation, be the protective capacities optimal value of whole safety-protection system.Under this condition, the average risk value of safety-protection system drops to minimum.

The invention will be further described in conjunction with the accompanying drawings with specific embodiment below:

1, sets up the security system abstract graph.Referring to accompanying drawing 1, we suppose that the security system abstract graph is made up of 5 nodes, are designated as V={A respectively, B, and C, D, E}, corresponding protective capacities value calculates by the formula of risk entropy, is respectively: x ₁=3.1, x ₂=2.7, x ₃=4.2, x ₄=1.2, x ₅=1.8.The weighted value on every limit provides on figure.

2, set up an Optimization Model that the safety-protection system protective capacities of 5 nodes is arranged.

$f=\max \underset{\underset{n\∈A}{(v_i,v_j)\∈A}}{\mathrm{\Σ}}{c}_{\mathrm{ij}}{x}_n$

$\left\{\begin{array}{c}{x}_1+x_2+...+x_5=13\\ 0\≤x_1\≤3.5\\ 0\≤x_2\≤2.9\\ 0\≤x_3\≤3.2\\ 0\≤x_4\≤2.3\\ 0\≤x_5\≤2.7\\ 0\≤c_{\mathrm{ij}}\≤1\end{array}\right.$

3, calculate the Optimization Model of finding the solution the safety-protection system protective capacities.

1. according to security protection network system abstract graph,, obtain the set M of all attack paths from the point of attack to the protected object by enumeration methodology.M＝{ph ₁，ph ₂，.......ph ₅}。Wherein: ph ₁Be S-A-C-E-T; Ph ₂Be S-B-D-E-T; Ph ₃Be S-A-D-E-T; Ph ₄Be S-A-D-C-E-T; Ph ₅Be S-B-D-C-E-T;

2. respectively below set M, utilize the Optimization Model of protective capacities, adjust the protective capacities value of node, protective capacities to safety-protection system is optimized, draw under different attack paths the maximal value of safety-protection system protective capacities, and the optimized protection ability value of each node.Calculate the maximal value of the safety-protection system protective capacities under above-mentioned 5 paths below respectively.

(1) corresponding to ph ₁The Optimization Model in path

f ₁＝Max(0.7x ₁+0.4x ₃+0.3x ₅)

$\left\{\begin{array}{c}{x}_1+x_2+...+x_5=13\\ 0\≤x_1\≤3.5\\ 0\≤x_2\≤2.9\\ 0\≤x_3\≤3.2\\ 0\≤x_4\≤2.3\\ 0\≤x_5\≤2.7\\ 0\≤c_{\mathrm{ij}}\≤1\end{array}\right.$

Above-mentioned Optimization Model belongs to typical linear programming model, utilizes simple method to find the solution and can obtain following optimum solution.Be respectively: f _1max=4.54, x ₁=3.5, x ₂=1.3, x ₃=3.2, x ₄=2.3, x ₅=2.7.

(2) corresponding to ph ₂The Optimization Model in path

f ₂＝Max(0.5x ₂+0.4x ₄+0.6x ₅)

$\left\{\begin{array}{c}{x}_1+x_2+...+x_5=13\\ 0\≤x_1\≤3.5\\ 0\≤x_2\≤2.9\\ 0\≤x_3\≤3.2\\ 0\≤x_4\≤2.3\\ 0\≤x_5\≤2.7\\ 0\≤c_{\mathrm{ij}}\≤1\end{array}\right.$

Above-mentioned Optimization Model belongs to typical linear programming model, utilizes simple method to find the solution and can obtain following optimum solution.Be respectively: f _2max=3.99, x ₁=3.5, x ₂=2.9, x ₃=1.6, x ₄=2.3, x ₅=2.7,

(3) corresponding to ph ₃The Optimization Model in path

f ₃＝Max(0.7x ₁+0.5x ₄+0.6x ₅)

$\left\{\begin{array}{c}{x}_1+x_2+...+x_5=13\\ 0\≤x_1\≤3.5\\ 0\≤x_2\≤2.9\\ 0\≤x_3\≤3.2\\ 0\≤x_4\≤2.3\\ 0\≤x_5\≤2.7\\ 0\≤c_{\mathrm{ij}}\≤1\end{array}\right.$

Above-mentioned Optimization Model belongs to typical linear programming model, utilizes simple method to find the solution and can obtain following optimum solution.Be respectively: f _3max=5.22, x ₁=3.5, x ₂=2.9, x ₃=1.6, x ₄=2.3, x ₅=2.7,

(4) corresponding to ph ₄The Optimization Model in path

f ₄＝Max(0.7x ₁+0.5x ₄+0.2x ₃+0.3x ₅)

$\left\{\begin{array}{c}{x}_1+x_2+...+x_5=13\\ 0\≤x_1\≤3.5\\ 0\≤x_2\≤2.9\\ 0\≤x_3\≤3.2\\ 0\≤x_4\≤2.3\\ 0\≤x_5\≤2.7\\ 0\≤c_{\mathrm{ij}}\≤1\end{array}\right.$

Above-mentioned Optimization Model belongs to typical linear programming model, utilizes simple method to find the solution and can obtain following optimum solution.Be respectively: f _4max=5.05, x ₁=3.5, x ₂=1.3, x ₃=3.2, x ₄=2.3, x ₅=2.7

(5) corresponding to ph ₅The Optimization Model in path

f ₅＝Max(0.5x ₂+0.4x ₄+0.2x ₃+0.3x ₅)

$\left\{\begin{array}{c}{x}_1+x_2+...+x_5=13\\ 0\≤x_1\≤3.5\\ 0\≤x_2\≤2.9\\ 0\≤x_3\≤3.2\\ 0\≤x_4\≤2.3\\ 0\≤x_5\≤2.7\\ 0\≤c_{\mathrm{ij}}\≤1\end{array}\right.$

Above-mentioned Optimization Model belongs to typical linear programming model, utilizes simple method to find the solution and can obtain following optimum solution.Be respectively: f _5max=3.82, x ₁=1.9, x ₂=2.9, x ₃=3.2, x ₄=2.3, x ₅=2.7

3. respectively relatively at ph ₁, ph ₂... .ph ₅The protective capacities value of safety-protection system under the path, note f _Max=Max{f _1max, f _2max, f _3max, f _4max, f _5maxBeing the maximal value of security protection network protection ability, pairing that paths of this maximal value is maximum attack path, obtains f according to top computational analysis _Max=5.22.

This moment, corresponding maximum attack path was ph ₃

4. the weight of supposing every paths equates that after above-mentioned a plurality of target weighted means, it is as follows to change into the single goal Optimization Model:

f _avg＝Max(0.42x ₁+0.2x ₂+0.16x ₃+0.32x ₄+0.42x ₅)

$\left\{\begin{array}{c}{x}_1+x_2+...+x_5=13\\ 0\≤x_1\≤3.5\\ 0\≤x_2\≤2.9\\ 0\≤x_3\≤3.2\\ 0\≤x_4\≤2.3\\ 0\≤x_5\≤2.7\\ 0\≤c_{\mathrm{ij}}\≤1\end{array}\right.$

Above-mentioned Optimization Model belongs to typical linear programming model, utilizes simple method to find the solution and can obtain following optimum solution.Be respectively: f _Avg=4.176, x ₁=3.5, x ₂=2.9, x ₃=1.6, x ₄=2.3, x ₅=2.7.This moment, the protective capacities optimal value of safety-protection system was 4.176, and the average risk value of safety-protection system drops to minimum.

Claims (2)

1. the optimization method of protective capacities in the safety-protection system is characterized in that, may further comprise the steps:

Set up the abstract graph of security system;

Set up the Optimization Model of safety-protection system protective capacities;

The Optimization Model of safety-protection system protective capacities is found the solution in calculating, and this step further comprises following substep:

1. according to the security system abstract graph,, obtain the set M of all attack paths from the point of attack to the protected object by enumeration methodology;

2. respectively below set M, utilize the Optimization Model of protective capacities, adjust the protective capacities value of node, protective capacities to safety-protection system is optimized, draw under different attack paths the maximal value of safety-protection system protective capacities, and the optimized protection ability value of each node;

3. the protective capacities value that compares the safety-protection system under all attack paths respectively, obtain the protective capacities maximal value of safety-protection system, simultaneously this path is designated as maximum attack path, when the assailant prevents protected object along this path attack, the cost maximum that need pay;

4. the weight of considering every paths equates, the objective function of different paths correspondence is weighted on average, and multi-objective optimization question is become the single goal optimization problem, and this objective function is optimized, and obtains overall optimum solution.

2. the optimization method of protective capacities in the safety-protection system according to claim 1 is characterized in that:

The described Optimization Model step of setting up the safety-protection system protective capacities comprises:

Regard whole abstract security protection network as a network chart, be designated as: D=(V, A), V represents the set of all security protection nodes in the safety-protection system, and A represents the set on all limits in the safety-protection system, for each bar limit (vi, vj) ∈ A supposes that the protective capacities value of each node is respectively: x ₁, x ₂... ..x _n, the weight on every limit is designated as c (v _i, v _j) 〉=0, then the Optimization Model of safety-protection system protective capacities is:

$f=\max \underset{\underset{n\∈V}{(v_i,v_j)\∈A}}{\mathrm{\Σ}}{c}_{\mathrm{ij}}{x}_n$

$\left\{\begin{array}{c}{x}_1+x_2+...+x_n=S\\ 0\≤x_1\≤m_1\\ 0\≤x_2\≤m_2\\ ..........\\ 0\≤x_n\≤m_n\\ 0\≤c_{\mathrm{ij}}\≤1\end{array}\right.$

Wherein S is known, expression whole node protective capacities sum; m ₁, m ₂... ..m _nBe known, represent the threshold value of the protective capacities of each node.

Images