CN103246818A - TOPSIS-method multi-target threat ordering method based on information entropy - Google Patents

Abstract

The invention provides a TOPSIS-method multi-target threat ordering method based on information entropy. The method includes first building a threat assessment threat factor model, building a target attribute decision matrix according to the threat factor model, adopting an entropy weight method to calculate a target attribute weight vector according to the decision matrix, adopting a TOPSIS method to calculate relative close degree of target schemes according to the weight vector calculated through the entropy weight method and conducting threat assessment ordering on the schemes according to the close degree, namely the threat degree. By means of the method, shortcomings of the traditional ideal solution approaching ordering method in determination of weight coefficients are effectively overcome, and the target threat degree prediction accuracy is improved.

Description

TOPSIS fado method for sorting target threat based on information entropy

Technical field

The invention belongs to the data fusion technical field, particularly a kind of multiple goal threatens sort method, can be used for airborne task processor, perhaps aerial/ground control command system.

Background technology

Data fusion is an informix treatment technology, threat assessment is important component part in the information fusion model that proposes of laboratory director joint council of the U.S. army, it is the important component part of setting up advanced command automation system, its research emphasis is how the target threat degree to be carried out assessment and the ordering of reasonable science, and guide one's own side further to take action, have uncertainty usually.All bringing into play important effect no matter be at military, civil area.Under the attack of terrorism background that is becoming increasingly rampant in the world, civil aviation because have the controllability height, influence is big, present stage becomes the preferred object that the terrorist implements terrorist activity, this has just proposed acid test to enterprise of civil aviaton, the civil aviation mangement department of China.Development and large-scale application along with electronic information technology; quantity of information is increasing; also become increasingly complex; correctly realize the target threat assessment fast; help the tower man of civil aviaton, civilian pilot's rapid reaction; take measures necessary protection crew's safety, improve survival probability, avoid unnecessary life and property loss.

Have a lot of methods to be applied to target threat assessment at present, as maximum weighted degree of membership method, minimum weight degree of membership method etc., it is only consideration and ideal solution or poor with negative ideal solution all.The TOPSIS method takes into full account both, not only near ideal solution but also away from negative ideal solution, is that a kind of to find the solution multiobjectives decision very effectively also be the method for using always at present.But traditional TOPSIS method is relatively more difficult to the objective attribute target attribute Weight Determination, often needs the people for providing, and very big subjectivity and blindness are arranged.

Summary of the invention

In order to overcome the deficiencies in the prior art, the invention provides a kind of TOPSIS fado method for sorting target threat based on information entropy, use entropy theory to determine the objective attribute target attribute weight, and will calculate gained objective attribute target attribute weight and be applied to and carry out target threat ordering in the TOPSIS method and resolve, improve target threat degree prediction accuracy.

The technical solution adopted for the present invention to solve the technical problems may further comprise the steps:

Step 1: structure objective attribute target attribute decision matrix $H={\left(h_{\mathrm{ij}}\right)}_{m\×n}=\left[\begin{array}{cccc}{h}_{11}& h_{12}& ...& h_{1n}\\ h_{21}& h_{22}& ...& h_{2n}\\ .& .& & .\\ .& .& ...& .\\ .& .& & .\\ h_{m1}& h_{m2}& ...& h_{\mathrm{mn}}\end{array}\right],$ In the formula, h _IjJ attribute representing i frame enemy plane, the attribute of enemy plane comprises following content:

(1) angle threatens the factor $h_i$ $1_{}=\left(\right|q_i^B|+|q_i^R\left|\right)/360,$ In the formula,

Be the target angle of lead,

Be bogey heading and score angle;

(2) distance threatens the factor $h_{i2}=\left\{\begin{array}{cc}0.5& r_i\&le;\mathrm{rm},r_i\&le;\mathrm{rm}{t}_i\\ 0.5-\frac{0.2(r_i-\mathrm{rm}{t}_i)}{(\mathrm{rm}-\mathrm{rm}{t}_i)}& \mathrm{rm}{t}_i<r_i<\mathrm{rm}\\ 1.0& \mathrm{rm}{t}_i>r_i>\mathrm{rm}\\ 0.8& \max (\mathrm{rm},\mathrm{rm}{t}_i)<r_i<\mathrm{rr}\end{array}\right.,$ In the formula, r _iBe target range, rmt _iBe the attack distance of the entrained guided missile of enemy plane, rm is my air defence missile maximum range, and rr is the maximum tracking range of our radar;

(3) speed threatens the factor $h_{i3}=\left\{\begin{array}{cc}0.1& v_i<0.6v_z\\ -0.5+v_i/v_z& 0.6v_z\&le;v_i\&le;1.5v_z\\ 1.0& v_i>1.5v_z\end{array}\right.,$ In the formula, v _zBe my motor speed, v _iBe target velocity;

Step 2: with the normalization of objective attribute target attribute decision matrix H row, obtain

Step 3: calculate the objective attribute target attribute entropy

J=1,2,, n, when

The time, regulation $\ln {\stackrel{.}{h}}_{\mathrm{ij}}=0;$

Step 4: calculate the objective attribute target attribute weight vectors

In the formula, 0≤ω _j≤ 1,

Step 5: substitution objective attribute target attribute weight, calculate the weighting standard matrix

Step 6: determine ideal solution $V^{*}=\{\left(\underset{1\&le;i\&le;m}{\mathrm{<figure-callout\; id="1"\; label="max"\; filenames="HDA00003193826100011.png"\; state="\{\{state\}\}">max</figure-callout>}}{v}_{\mathrm{ij}}\right|j\&Element;J^{+}),\left(\underset{1\&le;i\&le;m}{\mathrm{<figure-callout\; id="1"\; label="min"\; filenames="HDA00003193826100011.png"\; state="\{\{state\}\}">min</figure-callout>}}{v}_{\mathrm{ij}}\right|j\&Element;J^{-})\}=\{v_1^{*},v_2^{*},...,v_n^{*}\}$ With negative ideal solution $V^{-}=\{\left(\underset{1\&le;i\&le;m}{\mathrm{<figure-callout\; id="1"\; label="min"\; filenames="HDA00003193826100011.png"\; state="\{\{state\}\}">min</figure-callout>}}{v}_{\mathrm{ij}}\right|j\&Element;J^{+}),\left(\underset{1\&le;i\&le;m}{\max }{v}_{\mathrm{ij}}\right|j\&Element;J^{-})\}=\{v_1^{-},v_2^{-},...,v_n^{-}\},$ In the formula, J ⁺={ benefit type index set }, J ^-={ cost type index set };

Step 7: the distance that calculates ideal solution

With the distance to negative ideal solution $S_i^{-}=\sqrt{\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}{(v_{\mathrm{ij}}-v_j^{-})}^2};$

Step 8: the relative approach degree that calculates each scheme

Step 9: according to the i.e. size of degree of threat of the size of relative approach degree, to the assessment ordering that impends of each scheme.

The invention has the beneficial effects as follows: the present invention weighs method and TOPSIS method with entropy and combines the target threat degree is assessed, namely use entropy theory to handle the objective information that obtains, determine each attribute weight of target, weight is applied in the TOPSIS method, calculate multiobject threaten degree.The present invention can effectively avoid tradition to approach defective and the deficiency of the ranking method of ideal solution (TOPSIS) on definite weight coefficient, improves target threat degree prediction accuracy.

Description of drawings

Fig. 1 is both sides' aircraft flight situation synoptic diagram in the intimidation estimating method of the present invention;

Fig. 2 is intimidation estimating method schematic flow sheet of the present invention.

Embodiment

The present invention is further described below in conjunction with drawings and Examples.

The present invention includes following steps:

(1) sets up threat assessment and threaten factor model;

(2) according to threatening factor model, set up the objective attribute target attribute decision matrix;

(3) according to decision matrix, adopt entropy power method to calculate the objective attribute target attribute weight vectors;

(4) weight vectors that calculates according to entropy power method adopts the TOPSIS method to calculate the relative approach degree of each target protocol, according to the i.e. size of degree of threat of the size of relative approach degree, to the assessment ordering that impends of each scheme.

For instance, both sides' aircraft flight situation as shown in Figure 1, calculation procedure of the present invention is as follows:

Step 1: structure objective attribute target attribute decision matrix H=(h _Ij) _{M * n}

$H={\left(h_{\mathrm{ij}}\right)}_{m\&times;n}=\left[\begin{array}{cccc}{h}_{11}& h_{12}& ...& h_{1n}\\ h_{21}& h_{22}& ...& h_{2n}\\ .& .& & .\\ .& .& ...& .\\ .& .& & .\\ h_{m1}& h_{m2}& ...& h_{\mathrm{mn}}\end{array}\right]$

In the formula, h _IjJ attribute representing i frame enemy plane.

(1) angle threatens the factor

$h_{i1}=\left(\right|q_i^B|+|q_i^R\left|\right)/360$

In the formula,

Be the target angle of lead,

Be bogey heading and score angle (clockwise for just).

(2) distance threatens the factor

$h_{i2}=\left\{\begin{array}{cc}0.5& r_i\&le;\mathrm{rm},r_i\&le;\mathrm{rm}{t}_i\\ 0.5-\frac{0.2(r_i-\mathrm{rm}{t}_i)}{(\mathrm{rm}-\mathrm{rm}{t}_i)}& \mathrm{rm}{t}_i<r_i<\mathrm{rm}\\ 1.0& \mathrm{rm}{t}_i>r_i>\mathrm{rm}\\ 0.8& \max (\mathrm{rm},\mathrm{rm}{t}_i)<r_i<\mathrm{rr}\end{array}\right.$

In the formula, r _iBe target range, rmt _iBe the attack distance of the entrained guided missile of enemy plane, rm is our air defence missile maximum range, and rr is the maximum tracking range of our radar.

(3) speed threatens the factor

$h_{i3}=\left\{\begin{array}{cc}0.1& v_i<0.6v_z\\ -0.5+v_i/v_z& 0.6v_z\&le;v_i\&le;1.5v_z\\ 1.0& v_i>1.5v_z\end{array}\right.$

In the formula, v _zBe my motor speed, v _iBe target velocity.

Step 2: with the normalization of objective attribute target attribute decision matrix H row

$\stackrel{.}{H}={\left({\stackrel{.}{h}}_{\mathrm{ij}}\right)}_{m\&times;n}={(h_{\mathrm{ij}}/\underset{i=i}{\overset{m}{\mathrm{\&Sigma;}}}{h}_{\mathrm{ij}})}_{m\&times;n}$

Step 3: calculate the objective attribute target attribute entropy

$E_j=-\frac{1}{\ln n}\underset{i=1}{\overset{m}{\mathrm{\&Sigma;}}}{\stackrel{.}{h}}_{\mathrm{ij}}\ln {\stackrel{.}{h}}_{\mathrm{ij}},(j=\mathrm{1,2},...,n)$ (when ${\stackrel{.}{h}}_{\mathrm{ij}}=0$ The time, regulation $\ln {\stackrel{.}{h}}_{\mathrm{ij}}=0$ ）；

Step 4: calculate the objective attribute target attribute weight vectors

${\mathrm{\&omega;}}_j=(1-E_j)/\underset{k=1}{\overset{n}{\mathrm{\&Sigma;}}}(1-E_k)$ In the formula, 0≤ω _j≤ 1, $\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{\&omega;}}_j=1.$

Step 5: substitution objective attribute target attribute weight, calculate the weighting standard matrix

$V={\left(v_{\mathrm{ij}}\right)}_{m\&times;n}={\left({\mathrm{\&omega;}}_j{\stackrel{.}{h}}_{\mathrm{ij}}\right)}_{m\&times;n}$

Step 6: determine ideal solution and negative ideal solution

Ideal solution is got under this attribute solution most threatening between the different target machine for each index attribute.Negative ideal solution is the solution of threat value minimum.At benefit type index, ideal solution is got the index maximal value; At cost type index, ideal solution is got the index minimum value.Otherwise can draw negative ideal solution.Its computing formula is as follows,

Ideal solution

$V^{*}=\{\left(\underset{1\&le;i\&le;m}{\mathrm{<figure-callout\; id="1"\; label="max"\; filenames="HDA00003193826100011.png"\; state="\{\{state\}\}">max</figure-callout>}}{v}_{\mathrm{ij}}\right|j\&Element;J^{+}),\left(\underset{1\&le;i\&le;m}{\mathrm{<figure-callout\; id="1"\; label="min"\; filenames="HDA00003193826100011.png"\; state="\{\{state\}\}">min</figure-callout>}}{v}_{\mathrm{ij}}\right|j\&Element;J^{-})\}=\{{\mathrm{<figure-callout\; id="1"\; label="v"\; filenames="HDA00003193826100011.png"\; state="\{\{state\}\}">v</figure-callout>}}_1^{*},v_2^{*},...,v_n^{*}\}$

Negative ideal solution

$V^{-}=\{\left(\underset{1\&le;i\&le;m}{\mathrm{<figure-callout\; id="1"\; label="min"\; filenames="HDA00003193826100011.png"\; state="\{\{state\}\}">min</figure-callout>}}{v}_{\mathrm{ij}}\right|j\&Element;J^{+}),\left(\underset{1\&le;i\&le;m}{\mathrm{<figure-callout\; id="1"\; label="max"\; filenames="HDA00003193826100011.png"\; state="\{\{state\}\}">max</figure-callout>}}{v}_{\mathrm{ij}}\right|j\&Element;J^{-})\}=\{v_1^{-},v_2^{-},...,v_n^{-}\}$

In the formula, J ⁺={ benefit type index set }, J ^-={ cost type index set };

Step 7: the distance that calculates ideal solution and negative ideal solution

Distance to ideal solution

$S_i^{*}=\sqrt{\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}{(v_{\mathrm{ij}}-v_j^{*})}^2}$

Distance to negative ideal solution

$S_i^{-}=\sqrt{\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}{(v_{\mathrm{ij}}-v_j^{-})}^2}$

Step 8: the relative approach degree that calculates each scheme

$C_i=S_i^{-}/(S_i^{-}+S_i^{*})$

Step 9: according to the i.e. size of degree of threat of the size of relative approach degree, to the assessment ordering that impends of each scheme.

Whole threat assessment process as shown in Figure 2.

The present invention is directed to traditional TOPSIS method and improve, introduce the concept of entropy, adopt entropy power method to calculate the objective attribute target attribute weight, and the weight that this resolves is brought in the TOPSIS method, avoided traditional TOPSIS method to determine subjectivity and the blindness of weight.Use method of the present invention and solve aerial target threat assessment problem, fast than the additive method computing velocity, be easy to realize, improve the accuracy of threat assessment.

Claims (1)

1. the TOPSIS fado method for sorting target threat based on information entropy is characterized in that comprising the steps:

Step 1: structure objective attribute target attribute decision matrix $H={\left(h_{\mathrm{ij}}\right)}_{m\&times;n}=\left[\begin{array}{cccc}{h}_{11}& h_{12}& ...& h_{1n}\\ h_{21}& h_{22}& ...& h_{2n}\\ .& .& & .\\ .& .& ...& .\\ .& .& & .\\ h_{m1}& h_{m2}& ...& h_{\mathrm{mn}}\end{array}\right],$ In the formula, h _IjJ attribute representing i frame enemy plane, the attribute of enemy plane comprises following content:

(1) angle threatens the factor $h_{i1}=\left(\right|q_i^B|+|q_i^R\left|\right)/360,$ In the formula,

Be the target angle of lead,

Be bogey heading and score angle;

(2) distance threatens the factor $h_{i2}=\left\{\begin{array}{cc}0.5& r_i\&le;\mathrm{rm},r_i\&le;\mathrm{rm}{t}_i\\ 0.5-\frac{0.2(r_i-\mathrm{rm}{t}_i)}{(\mathrm{rm}-\mathrm{rm}{t}_i)}& \mathrm{rm}{t}_i<r_i<\mathrm{rm}\\ 1.0& \mathrm{rm}{t}_i>r_i>\mathrm{rm}\\ 0.8& \max (\mathrm{rm},\mathrm{rm}{t}_i)<r_i<\mathrm{rr}\end{array}\right.,$ In the formula, r _iBe target range, rmt _iBe the attack distance of the entrained guided missile of enemy plane, rm is my air defence missile maximum range, and rr is the maximum tracking range of our radar;

(3) speed threatens the factor $h_{i3}=\left\{\begin{array}{cc}0.1& v_i<0.6v_z\\ -0.5+v_i/v_z& 0.6v_z\&le;v_i\&le;1.5v_z\\ 1.0& v_i>1.5v_z\end{array}\right.,$ In the formula, v _zBe my motor speed, v _iBe target velocity;

Step 2: with the normalization of objective attribute target attribute decision matrix H row, obtain

Step 3: calculate the objective attribute target attribute entropy

J=1,2,, n, when

The time, regulation $\ln {\stackrel{.}{h}}_{\mathrm{ij}}=0;$

Step 4: calculate the objective attribute target attribute weight vectors

In the formula, 0≤ω _j≤ 1,

Step 5: substitution objective attribute target attribute weight, calculate the weighting standard matrix

Step 6: determine ideal solution $V^{*}=\{\left(\underset{1\&le;i\&le;m}{\max }{v}_{\mathrm{ij}}\right|j\&Element;J^{+}),\left(\underset{1\&le;i\&le;m}{\min }{v}_{\mathrm{ij}}\right|j\&Element;J^{-})\}=\{v_1^{*},v_2^{*},...,v_n^{*}\}$ With negative ideal solution $V^{-}=\{\left(\underset{1\&le;i\&le;m}{\min }{v}_{\mathrm{ij}}\right|j\&Element;J^{+}),\left(\underset{1\&le;i\&le;m}{\max }{v}_{\mathrm{ij}}\right|j\&Element;J^{-})\}=\{v_1^{-},v_2^{-},...,v_n^{-}\},$ In the formula, J ⁺={ benefit type index set }, J ^-={ cost type index set };

Step 7: the distance that calculates ideal solution

With the distance to negative ideal solution $S_i^{-}=\sqrt{\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}{(v_{\mathrm{ij}}-v_j^{-})}^2};$

Step 8: the relative approach degree that calculates each scheme

Step 9: according to the i.e. size of degree of threat of the size of relative approach degree, to the assessment ordering that impends of each scheme.

Images