Ground target threat degree evaluation method based on intuitionistic fuzzy TOPSIS and adaptive weight

Abstract

The invention discloses a ground time-sensitive target threat degree evaluation method based on intuitionistic fuzzy TOPSIS and adaptive weight, which comprises the following steps: constructing target dynamic and static attribute evaluation indexes, and selecting the dynamic attribute evaluation index and the static attribute evaluation index, constructing a quantitative processing method of dynamic and static attribute indexes, and converting the dynamic and static attribute evaluation indexes of the target into intuitionistic fuzzy numbers; calculating the weight of the static attribute index and the constant weight of the dynamic attribute index by adopting a nonlinear programming model based on entropy minimization; solving an adaptive weight vector of the dynamic attribute index by adopting a variable weight theory, taking the adaptive weight vector as a weight of the dynamic attribute index, and constructing a combined weight of the dynamic and static attribute indexes according to the weight of the dynamic attribute index; and based on the intuitionistic fuzzy values of the dynamic and static attribute indexes and the weights thereof, solving the comprehensive threat degree of the ground time-sensitive target by adopting an intuitionistic fuzzy TOPSIS method based on the adaptive weight, and sorting the comprehensive threat degree. According to the invention, the target threat degree prediction precision is improved.

Description

Technical Field

[0001] The technical field of the present invention belongs to accurate assessment of the threat level of multiple ground targets, and specifically, is a method for assessing the threat level of time-sensitive ground targets based on intuitive fuzzy TOPSIS and adaptive weights. Background Art

[0002] As an important part of fire control, the accuracy of target threat assessment will directly affect the subsequent fire allocation and dispatch of command and control personnel, and further affect the security status of our positions. In this context, comprehensive analysis of various information of incoming targets and accurate and reasonable threat assessment of incoming targets are of great significance for optimizing command and control.

[0003] At present, the main model for ground target threat assessment is the intuitive fuzzy TOPSIS method. However, the traditional intuitive fuzzy TOPSIS method has many problems, which always affect the accuracy of the prediction results. When dealing with dynamic attribute indicators such as target speed and target distance, the traditional intuitive fuzzy TOPSIS method often converts their weights into static constant weights for analysis, ignoring the time-sensitive characteristics of dynamic indicators and reducing the accuracy of threat assessment; in addition, the traditional intuitive fuzzy TOPSIS method ignores the internal connection between the indicators, simply analyzes each attribute indicator separately, and lacks comprehensive consideration of multiple indicators. At present, the existing intuitive fuzzy TOPSIS method cannot effectively deal with these problems. Summary of the invention

[0004] The purpose of the present invention is to provide a ground target threat assessment method based on intuitionistic fuzzy TOPSIS and adaptive weights to improve the prediction accuracy of the traditional intuitionistic fuzzy TOPSIS method.

[0005] The implementation steps of the present invention are as follows:

[0006] A ground target threat assessment method based on intuitionistic fuzzy TOPSIS and adaptive weights includes:

[0007] Step 1: construct the target dynamic and static attribute evaluation indicators, select dynamic attribute indicators and static attribute indicators, and quantify the dynamic and static attribute indicators and convert them into intuitive fuzzy numbers; among them, the static attribute indicators are converted into intuitive fuzzy numbers using fuzzy evaluation language; the dynamic attribute indicators are calculated by the membership function value as a real value, and then the real number is converted into an intuitive fuzzy number;

[0008] Step 2, the static attribute index and the dynamic attribute index are jointly calculated, and the weight of the static attribute index and the constant weight of the dynamic attribute index are obtained by using a nonlinear programming model based on entropy minimization, wherein the constant weight of the dynamic attribute index is used to further calculate its adaptive weight vector;

[0009] Step 3: Use variable weight theory to solve the state variable weight vector of the dynamic attribute index, so as to obtain the adaptive weight vector of each target dynamic attribute index as the weight of the dynamic attribute index, which together with the static attribute weight constitutes the combined weight of the dynamic and static attribute indicators;

[0010] Step 4: Based on the intuitive fuzzy values ​​and weights of dynamic and static attribute indicators, the intuitive fuzzy TOPSIS method is used to calculate the positive and negative ideal solutions of the intuitive fuzzy values ​​of the target set, and the similarity between each target and the positive and negative ideal solutions is solved to obtain the comprehensive threat degree of each time-sensitive target on the ground and its ranking.

[0011] Compared with the traditional intuitive fuzzy TOPSIS method, the advantages of the present invention are:

[0012] (1) The present invention uses a nonlinear programming model based on entropy minimization to solve the weights of static attribute indicators and the constant weights of dynamic attribute indicators. It can well consider the weight relationship between the attribute indicators, and divide the constant weights of the dynamic attribute indicators and the weights of the static attribute indicators together, ensuring that the weights of each attribute indicator are maintained within a reasonable range, making the evaluation results more accurate and reliable.

[0013] (2) Based on the intuitive fuzzy TOPSIS method, the present invention introduces the variable weight theory, and uses the dynamic attribute index in the form of an adaptive weight vector to participate in the subsequent calculation of the comprehensive threat level. This makes the dynamic attribute index weight of the target have adaptive characteristics and can be adjusted at any time according to the time-sensitive state of the target, making the threat level assessment result more objective and comprehensive. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a flow chart of the ground target threat assessment method based on intuitionistic fuzzy TOPSIS and adaptive weights.

[0015] Figure 2 This is a diagram of the ground target threat assessment index system. DETAILED DESCRIPTION

[0016] The specific implementation of this scheme will be described below to facilitate the relevant technical personnel in this field to understand the content of the present invention. When the design details and known knowledge may dilute the main content of the present invention, these descriptions will be skipped here.

[0017] The following will be combined Figure 1 , Figure 2 , describes the specific implementation of the ground target threat assessment method based on intuitionistic fuzzy TOPSIS and adaptive weight. The specific implementation steps are as follows:

[0018] Step 1, construct the dynamic and static attribute evaluation indicators of the target, select the three factors of target distance, speed and attack angle as the dynamic attribute evaluation indicators, select the target type (including infantry fighting vehicles, tanks, troop carriers, 122 vehicle-mounted howitzers, 155 vehicle-mounted howitzers, 122 rocket launchers, command vehicles, air defense missile vehicles, single soldiers, etc., the target types can be added according to the specific battlefield environment, and fuzzy evaluation language is given according to their threat level), command capability (that is, the ability of the target to command and coordinate the operations of other targets, and fuzzy evaluation language is given according to the target type), reconnaissance capability (that is, the ability of the target to reconnaissance and detect our units, and fuzzy evaluation language is given according to the target type), mobility (that is, the ability of the target to change the state of motion per unit time, and fuzzy evaluation language is given according to the target type), damage capability (that is, the ability of the target to damage our positions, and fuzzy evaluation language is given according to the target type and equipment situation), We measure 10 factors as static attribute evaluation indicators, including anti-interference capability (i.e., the ability of the target to operate stably in an electronic interference environment, and fuzzy evaluation language is assigned according to the target type), counterattack capability (i.e., the ability of the target to counterattack after being hit, and fuzzy evaluation language is assigned according to the target type), combat intention (including attack, march, reconnaissance, transportation, etc., and fuzzy evaluation language is assigned according to the threat level), number of weapons carried (dividing the number of weapons carried into gradients, such as 0, 1-3, 4-6, etc., and fuzzy evaluation language is assigned according to the threat level of the number gradient), and type of weapons carried (including portable light weapons such as firearms and grenades, rockets, small and medium caliber artillery, large caliber artillery, tank guns, anti-aircraft missiles, etc., and the types of weapons can be added according to the specific battlefield environment, and fuzzy evaluation language is assigned according to the threat level). The dynamic and static attribute evaluation indicators are quantified and converted into intuitive fuzzy numbers. Among them, the static attribute evaluation indicators can be converted into intuitive fuzzy numbers through fuzzy evaluation language; the dynamic attribute evaluation indicators can be processed by calculating the membership function value as its real value, and then converting the real number into intuitive fuzzy number.

[0019] The processing of static attribute indicators in step 1 specifically includes the following steps:

[0020] (1) First, the 10 static attribute indicators of target type, command capability, reconnaissance capability, mobility, damage capability, our anti-interference capability, counterattack capability, combat intention, number of weapons carried, and type of weapons carried are expressed in fuzzy evaluation language. The fuzzy evaluation language is divided into 11 scales, with a scale value of α, and values ​​of 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, and 0, corresponding to extremely large, very large, large, relatively large, slightly large, medium, slightly small, relatively small, small, very small, and extremely small, respectively. The threat level of the described indicators gradually decreases. Based on expert experience, the fuzzy evaluation language is assigned to the static attribute indicators of different targets, and the initial intuitive fuzzy number values ​​μ and v are obtained by looking up the table based on this. μ and v are the quantitative values ​​of the fuzzy evaluation language. The larger μ and the smaller v, the higher the threat level. The corresponding values ​​of the fuzzy evaluation language converted into the initial intuitive fuzzy number are shown in Table 1 below:

[0021] Table 1 Conversion of fuzzy evaluation language into initial intuitive fuzzy numbers

[0022]

[0023]

[0024] (2) Secondly, obtain the interval value of the degree of certainty of the target fuzzy evaluation language. The degree of certainty represents the degree of certainty of the fuzzy evaluation language. According to the degree of certainty of the fuzzy evaluation language, the table is looked up to obtain the interval number corresponding to the degree of certainty. Here, 5 gradients of degree of certainty c1~c5 are constructed, which respectively represent the degree of certainty of the fuzzy evaluation language. The interval values ​​corresponding to various degrees of certainty are shown in Table 2 below:

[0025] Table 2 Interval values ​​corresponding to the degree of certainty

[0026]

[0027] In the table, C L is the lower limit of the interval, C U is the upper limit of the interval.

[0028] (3) Finally, the intuitive fuzzy number of each static attribute index of the target is solved. According to the obtained initial intuitive fuzzy number and the interval value of the degree of certainty, the intuitive fuzzy number τ of each static attribute index is calculated. ik ,ζ ik , the specific calculation formula is as follows:

[0029]

[0030] In the formula, τ ik The positive intuitive fuzzy number representing the kth static attribute index of the ith target, ζ ik Represents the negative intuitive fuzzy number of the kth static attribute index of the i-th target, τik ,ζ ik The intuitive fuzzy value of the static attribute index can be directly substituted into the subsequent threat degree calculation. m ,v m is the initial intuitive fuzzy number corresponding to the intermediate scale α = (L-1) / 2, which can be obtained by looking up the table in Table 1, and L = 11 is the number of divided fuzzy evaluation language scales. It indicates the degree of certainty of the decision maker's fuzzy evaluation language of the target, which can generally be taken as is the lower limit of the certainty interval of the fuzzy evaluation language of the kth static attribute index of the ith target, is the upper limit of the certainty interval of the fuzzy evaluation language of the kth static attribute index of the i-th target. <μ m ,v m The effect of > is that the smaller the decision maker's degree of certainty, the closer the decision result will be to the middle decision value, making the judgment result more realistic.

[0031] The processing of dynamic attribute indicators in step 1 specifically includes the following steps:

[0032] (1) First, the membership functions of the three dynamic attribute indicators, namely target speed, target distance, and attack angle, are constructed and solved. The membership function values ​​of the dynamic attribute indicators are solved by the following formula:

[0033] ① Membership function of target speed

[0034] The faster the target moves, the more difficult it is for us to attack it, and the greater the threat. Therefore, the membership function of the target speed threat can be processed according to the benefit-type index. The membership function of the target speed threat is

[0035]

[0036] Where, T i1 is the membership function of the velocity threat of the i-th target. β1,β2,β3,β4∈[0,1] are the threat factors of armored vehicle targets (including armored vehicle targets such as infantry fighting vehicles, tanks, and personnel carriers), suppression weapon targets (including various howitzers, howitzers, mortars, rocket launchers, etc. that provide fire suppression), high-value targets (including command vehicles, radar vehicles, air defense missile vehicles, anti-tank missile vehicles, etc., which have high technical content and cost), and individual targets. The setting of threat factors can be selected according to the actual scenario. Generally, the higher the threat level, the larger the value; v maxe v is the maximum speed that the e-th target can reach among the above four types of targets, e = 1, 2, 3, 4; e is the current speed of the e-th target.

[0037] ② Membership function of target distance

[0038] On the basis of considering the target distance factor, it is necessary to comprehensively consider the reconnaissance distance and strike distance of our weapon platform and the target, and on this basis construct the membership function of the target distance threat degree. a ,r ta are the maximum attack distances between our weapon platform and the target, r r ,r tr They are respectively the maximum reconnaissance distances between our weapon platform and the target.

[0039] Based on the relationship between the four distance parameters, the membership function of the target distance can be divided into the following six cases:

[0040] 1) When r tr >r ta >r r >r a hour

[0041]

[0042] 2) When r tr ≥r r >r ta ≥r a hour

[0043]

[0044] 3) When r tr ≥r r >r a >r ta hour

[0045]

[0046] 4) When r r ≥r tr >r ta ≥r a hour

[0047]

[0048] 5) When r r >r tr ≥r a ≥r ta hour

[0049]

[0050] 6) When r r >r a ≥r tr >r ta hour

[0051]

[0052] In the formula, r i is the distance between the ith target and our weapon platform, T i2 is the distance membership function of the i-th target. ab ∈[0,1], is the bth quantized coefficient in the ath case among the above 6 cases. ab The value of can be determined according to different target characteristics and combat missions.

[0053] ③ Membership function of target attack angle

[0054] When the target attacks us, the angle between its direction and our visual position will inevitably increase. The closer the angle is to 180°, the greater the threat to us. Therefore, the membership function value of the target attack angle threat is positively correlated with the size of the angle. The membership function of the target attack angle is as follows:

[0055]

[0056] Where, T i3 is the membership function of the attack angle of the ith target, θ∈[0,180°] is the target’s entry angle, a=180° is the upper limit of θ, and k θ Indicates the deflection angle influence factor, take k θ =0.0005.

[0057] (2) Secondly, the intuitive fuzzy value of each dynamic attribute index is calculated. The dynamic attribute index is processed by converting real numbers into intuitive fuzzy numbers. The membership function value of the dynamic attribute index is used as its real value (membership value) to calculate the intuitive fuzzy value of each dynamic attribute index. The specific calculation formula for converting real values ​​into intuitive fuzzy values ​​is as follows:

[0058]

[0059] In the formula, τ il The positive intuitive fuzzy number representing the lth dynamic attribute index of the ith target, ζ il The negative intuitive fuzzy number representing the lth dynamic attribute index of the i-th target, l = 1, 2, 3, representing the sequence number of the dynamic attribute index, corresponding to target speed, target distance, and attack angle, respectively. g ∈[0,1] is an additional fuzzy factor that prevents the deterministic characteristics of real numbers from affecting other uncertainty indicators. g The value of T is selected according to the specific situation. il is the membership function value of the lth dynamic attribute index of the ith target, corresponding to T i1 ,T i2 ,Ti3 , m is the number of incoming targets.

[0060] The intuitive fuzzy values ​​of the dynamic and static attribute indicators of the target are solved in the order of target type, command capability, reconnaissance capability, mobility, damage capability, our anti-interference capability, counterattack capability, combat intention, number of weapons carried, type of weapons carried, target speed, target distance, and attack angle to form an intuitive fuzzy value matrix M, where the first 20 columns of the matrix are the intuitive fuzzy values ​​of the static attribute indicators, and the last 6 columns are the intuitive fuzzy values ​​of the dynamic attribute indicators. M is as follows:

[0061]

[0062] In the formula, τ ij is the positive intuitive fuzzy number of the jth attribute index of the ith target, ζ ij is the negative intuitionistic fuzzy number of the jth attribute index of the ith target, n is the sum of the number of dynamic and static attribute indexes, and n=13.

[0063] Step 2: Calculate the static attribute index and the dynamic attribute index together, and use a nonlinear programming model based on entropy minimization to obtain the weights of the static attribute index and the constant weights of the dynamic attribute index, where the constant weights of the dynamic attribute index are used to further calculate its adaptive weight vector. The specific steps are as follows:

[0064] (1) First, the dynamic and static attribute indicators of the target are converted into weighted analysis values ​​(IFE). The IFE calculation formulas for the dynamic and static attributes of the target are as follows:

[0065]

[0066] In the formula, E j is the IFE of the jth attribute index, τ ij is the positive intuitive fuzzy number of the jth attribute index of the ith target, ζ ij is the negative intuitionistic fuzzy number of the jth attribute index of the i-th target.

[0067] (2) Then, a quadratic nonlinear programming model based on IFE is established for the weights of each target attribute. The calculation formula is as follows:

[0068]

[0069] In the formula, ω j ≥0, indicating the static objective weight or dynamic constant weight of the jth attribute index, satisfying the constraint condition

[0070] Solving the quadratic nonlinear programming model, we can obtain the weights of the static attribute index and the constant weights of the dynamic attribute index.j The row matrix W is composed in the order of target type, command capability, reconnaissance capability, mobility, damage capability, our anti-interference capability, anti-strike capability, combat intention, number of weapons carried, type of weapons carried, target speed, target distance, and attack angle. q , called the weight matrix.

[0071] Step 3: Use variable weight theory to solve the state variable weight vector of the dynamic attribute index, so as to obtain the adaptive weight vector of each target dynamic attribute index as the weight of the dynamic attribute index, which together with the static attribute weight constitutes the combined weight of the dynamic and static attribute indexes. The specific steps are as follows:

[0072] (1) First, based on the dynamic attribute constant weight (constant weight) solved in step 2, the variable weight theory is used to calculate the state variable weight vector of each target dynamic attribute. The calculation formula is as follows:

[0073]

[0074] Where U il is the state variable weight vector of the lth dynamic attribute of the ith target. N is the penalty excitation amplitude ratio, N=1.8, k' is the penalty threshold coefficient, k'=0.7. is the average threat quantification value of the lth dynamic attribute indicator; T il is the membership function value of the lth dynamic attribute index of the ith target. ω 10+l is the constant weight of the lth dynamic attribute index, corresponding to the weight matrix W q The last 3 columns of elements.

[0075] (2) Secondly, the Hadamard product is performed on the dynamic attribute constant weight vector and the state variable weight vector to calculate the adaptive weight vector of the dynamic attribute index. The calculation formula is as follows:

[0076]

[0077] In the formula, is the adaptive weight vector of the lth dynamic attribute index of the ith target. d The total weight of the added dynamic indicators is used to incentivize and punish based on the total weight of the dynamic indicators, so that the sum of the dynamic indicator weights remains unchanged and the total value of the dynamic and static indicator weights is 1. The calculation formula is as follows:

[0078]

[0079] (3) Finally, the one-dimensional weight matrix W obtained in step 2 q The first 10 columns (i.e., the objective weights of static attribute indicators) are expanded into two dimensions according to the target number m as the number of rows, so that each row element is consistent with Wq The first 10 columns are consistent, and the constructed matrix is ​​called the static attribute objective weight matrix, denoted by W s The static attribute objective weight matrix and the dynamic attribute index adaptive weight are combined to form the combined weight matrix W of dynamic and static attribute indicators. The constituent elements are shown in the following formula:

[0080]

[0081] Where W ij is the element in the i-th row and j-th column of the matrix W, that is, the weight value of the j-th attribute index of the i-th target; is the matrix W s The element in the i-th row and j-th column is the objective weight value of the j-th static attribute indicator of the i-th target; is the adaptive weight of the j-10th dynamic attribute index of the ith target.

[0082] Step 4: Based on the intuitive fuzzy values ​​and weights of the dynamic and static attribute indicators, the intuitive fuzzy TOPSIS method is used to calculate the positive and negative ideal solutions of the intuitive fuzzy values ​​of the target set, and the similarity between each target and the positive and negative ideal solutions is solved to obtain the comprehensive threat level and ranking of each time-sensitive target on the ground. The specific steps are as follows:

[0083] (1) First, select the optimal solution A for each attribute index + and the worst solution A - The expressions of the optimal solution and the worst solution are as follows:

[0084]

[0085] In the formula, A + is the optimal solution matrix for each attribute index, A - is the worst solution matrix for each attribute index, They respectively represent the positive ideal solution (i.e., maximum value) of the positive intuitive fuzzy number, the positive ideal solution (i.e., minimum value) of the negative intuitive fuzzy number, the negative ideal solution (i.e., minimum value) of the positive intuitive fuzzy number, and the negative ideal solution (i.e., maximum value) of the negative intuitive fuzzy number of the j-th attribute index.

[0086] (2) Secondly, calculate the similarity between each target and the positive ideal solution and the negative ideal solution The calculation formula is as follows:

[0087]

[0088] In the formula, is the similarity between the ith target and the positive ideal solution, is the similarity between the i-th target and the negative ideal solution, W ijis the weight of the jth attribute index of the ith target. is the similarity between the intuitionistic fuzzy number and the positive ideal solution and the negative ideal solution. The calculation formula is as follows:

[0089]

[0090] (3) Finally, calculate the relative closeness p of each target i , thereby obtaining the comprehensive threat degree Th of the target i , the calculation formula is as follows:

[0091]

[0092] In the formula, p i is the relative closeness of the i-th target, Th i is the quantitative value of the threat level of the i-th target, Th i The larger the value, the greater the threat level of the target. According to the solved target threat level quantification value, the targets are sorted in ascending order to obtain the order of ground target threat level.

[0093] Implementation Cases

[0094] Assume that the ground reconnaissance system has discovered 9 incoming ground targets. The detected target characteristic parameters are shown in Table 3 below, where the target numbers are represented by M1 to M9.

[0095] Table 3 Target raw data collection

[0096]

[0097] The implementation steps for evaluating the threat level of ground targets based on the target characteristic parameters detected by the ground reconnaissance system are as follows:

[0098] Step 1: Construct the dynamic and static attribute evaluation indicators of the target, select three factors including target distance, speed and attack angle as dynamic attribute evaluation indicators, select 10 factors including target type, command capability, reconnaissance capability, mobility and damage capability as static attribute evaluation indicators, and quantify the dynamic and static attribute evaluation indicators into intuitive fuzzy numbers. The specific steps are as follows:

[0099] (1) First, the 10 static attribute indicators of the nine targets, including target type (F1), command capability (F2), reconnaissance capability (F3), mobility (F4), damage capability (F5), our anti-interference capability (F6), counterattack capability (F7), combat intention (F8), number of weapons carried (F9), and type of weapons carried (F10), are assigned different fuzzy evaluation languages ​​and certainty levels based on expert experience to form a quantitative table. The quantification of target F1 to F7 indicators is shown in Table 4. The left side of the data x / cy in the table represents the fuzzy evaluation language scale value, and the value of cy on the right side represents the certainty level.

[0100] Table 4 Quantification table of static attribute indicators of targets F1~F7

[0101]

[0102]

[0103] The quantification of the target combat intention (F8), the number of weapons carried (F9), and the type of weapons carried (F10) are shown in Tables 5 to 7. The meaning of the data x / cy in the table is the same as that in Table 4.

[0104] Table 5 Quantitative table of target combat intention indicators

[0105]

[0106] Table 6 Quantitative table of indicators of the number of weapons carried by targets

[0107]

[0108] Table 7 Quantitative table of indicators of target-carried weapon types

[0109]

[0110] (2) Secondly, the 10 static attribute indicators of the nine targets, including target type (F1), command capability (F2), reconnaissance capability (F3), mobility (F4), damage capability (F5), our anti-interference capability (F6), counterattack capability (F7), combat intention (F8), number of weapons carried (F9), and type of weapons carried (F10), are quantified by converting fuzzy evaluation language into intuitionistic fuzzy numbers. The intuitionistic fuzzy numbers of the target static attribute evaluation indicators are obtained by solving the obtained static attribute indicator quantization table as shown in Table 8.

[0111]

[0112]

[0113] Table 8 Intuitive fuzzy numbers of target static attribute indicators

[0114]

[0115] (3) Finally, the three dynamic attribute indicators of the nine targets, namely, target distance (F11), target speed (F12), and attack angle (F13), are processed according to the method of converting real numbers into intuitive fuzzy numbers. The membership function values ​​of the dynamic attribute indicators are solved according to the target speed, distance, and attack angle parameters obtained by the reconnaissance equipment. The membership function values ​​are taken as their real values ​​(membership values), and the intuitive fuzzy values ​​of each dynamic attribute indicator are obtained according to the formula.

[0116] The membership values ​​of the dynamic attribute indicators of the target are shown in Table 9.

[0117] Table 9 Dynamic attribute index membership function values ​​of the target

[0118]

[0119] The intuitive fuzzy values ​​of each dynamic attribute index are calculated according to the obtained target dynamic attribute index membership values. The calculated intuitive fuzzy values ​​are shown in Table 10.

[0120] Table 10 Intuitive fuzzy numbers of target dynamic attribute indicators

[0121]

[0122] Step 2: Calculate the static attribute index and the dynamic attribute index together, and use a nonlinear programming model based on entropy minimization to obtain the weights of the static attribute index and the constant weights of the dynamic attribute index, where the constant weights of the dynamic attribute index are used to further calculate its adaptive weight vector. The specific steps are as follows:

[0123] (1) First, substitute the intuitive fuzzy number obtained in step 1 into the formula to obtain the weighted analysis value E of each dynamic and static attribute of the target: j , the weight analysis value matrix E is as follows:

[0124] E=[0.1849 0.1956 0.1891 0.2722 0.1770 0.2634 0.1975 0.1502 0.11250.18360.24280.28140.2430]

[0126] (2) Secondly, according to the IFE of each dynamic and static attribute, the static index weight and the dynamic index constant weight are calculated to form the weight matrix W q As shown below, the first 10 elements of the matrix are the objective weights of static attribute indicators, and the last 3 are the constant weights of dynamic attribute indicators.

[0127] W q =[0.0810 0.0766 0.0792 0.0550 0.0846 0.0569 0.0758 0.0997 0.1331 0.0816 0.0617 0.0532 0.0616]

[0129] Step 3: Use variable weight theory to solve the state variable weight vector of the dynamic attribute index, so as to obtain the adaptive weight vector of each target dynamic attribute index as the weight of the dynamic attribute index, which together with the static attribute weight constitutes the combined weight of the dynamic and static attribute indexes. The specific steps are as follows:

[0130] (1) First, according to the dynamic attribute constant weights solved in step 2, the state variable weight vector of each target dynamic attribute is calculated using the variable weight theory to obtain the state variable weight vector U il , the state variable weight vector matrix U is as follows:

[0131]

[0132] (2) Secondly, based on the calculated dynamic attribute constant weight and state variable weight vector, the dynamic attribute adaptive weight vector of each incoming target is obtained through Hadamard product Composed of dynamic attribute adaptive weight matrix W d as follows:

[0133]

[0134] (3) Finally, the one-dimensional static indicator weight matrix obtained in step 2 is expanded into two dimensions and combined with the two-dimensional dynamic indicator adaptive weight matrix. Finally, the combined weight matrix W of the nine objectives can be obtained. The calculated combined weight matrix is ​​as follows:

[0135]

[0136] Among them, the first 10 columns of the matrix are static indicator weights, and the last 3 columns are adaptive weights of dynamic indicators.

[0137] Step 4: Based on the intuitive fuzzy values ​​and weights of the dynamic and static attribute indicators, the intuitive fuzzy TOPSIS method is used to calculate the positive and negative ideal solutions of the intuitive fuzzy values ​​of the target set, and the similarity between each target and the positive and negative ideal solutions is calculated to obtain the comprehensive threat level of each time-sensitive target on the ground. The specific steps are as follows:

[0138] (1) First, the optimal and worst solutions of each dynamic and static attribute index are selected, and the positive ideal solution of the positive intuitive fuzzy number, the negative ideal solution of the positive intuitive fuzzy number, the positive ideal solution of the negative intuitive fuzzy number, and the negative ideal solution matrix τ of the 13 attribute indexes are obtained. + ,τ - ,ζ + ,ζ - as follows:

[0139] τ + =[0.8906 1 0.4000 0.4000 0.8906 0.5757 0.8906 0.8906 1 0.89060.8000 0.8000 0.8000]

[0140] τ - =[0.1181 0.1868 0.1181 0.1868 0.0252 0.2555 0.1181 0.4000 0.02520.1199 0.0110 0.1246 0.0139]

[0141] ζ + =[0.0555 0 0.4000 0.4000 0.0555 0.2450 0.0555 0.0555 0 0.0555 0 00]

[0142] ζ - =[0.8186 0.7040 0.8186 0.7040 0.9552 0.6086 0.8186 0.4473 0.95520.7953 0.7890 0.6754 0.7861]

[0143] (2) Secondly, based on the components of the optimal solution and the worst solution obtained in (1), calculate the similarity between the 9 targets and the positive ideal solution and the negative ideal solution. This is used as an element to form the similarity matrix S between the target and the positive ideal solution and the negative ideal solution. + ,S - as follows:

[0144] S + =[0.6621 0.7893 0.8000 0.8708 0.8338 0.7693 0.7143 0.7464 0.6003]

[0145] S - =[0.6672 0.4021 0.3879 0.2839 0.3228 0.5803 0.5491 0.5024 0.8072]

[0146] (3) Finally, based on the obtained similarity, calculate the relative closeness p of each target i , thus obtaining the comprehensive threat level of the target Th i , the calculated comprehensive threat matrix Th is as follows:

[0147] Th=[0.4981 0.6625 0.6734 0.7541 0.7209 0.5700 0.5654 0.5977 0.4265]

[0148] The comprehensive threat levels of the targets are ranked, and the ranking results are as follows:

[0149] M9<M1<M7<M6<M8<M2<M3<M5<M4

[0150] In the specific implementation process, the threat assessment method can be implemented as a computer program. When importing the original data of the incoming target, it should be noted that the speed value of the target needs to meet certain boundary conditions. If the original value of the target speed obtained is 0, it is necessary to adjust the value to 1km / h and import it into the program as the target speed value for calculation, so that the program runs more smoothly, avoiding operation errors, and will not have a decisive impact on the assessment results. In addition, when the distance between the enemy and us is less than 300 meters, the accuracy of the assessment results will be reduced, which is somewhat different from the actual situation. Therefore, the distance of the target object is preferably greater than 300 meters to ensure the accuracy of the assessment results.

Claims

1. A ground target threat assessment method based on intuitionistic fuzzy TOPSIS and adaptive weight, characterized in that: include: Step 1: construct the target dynamic and static attribute evaluation indicators, select dynamic attribute indicators and static attribute indicators, and quantify the dynamic and static attribute indicators and convert them into intuitive fuzzy numbers; among them, the static attribute indicators are converted into intuitive fuzzy numbers using fuzzy evaluation language; the dynamic attribute indicators are calculated by the membership function value as a real value, and then the real number is converted into an intuitive fuzzy number; Step 2, the static attribute index and the dynamic attribute index are jointly calculated, and the weight of the static attribute index and the constant weight of the dynamic attribute index are obtained by using a nonlinear programming model based on entropy minimization, wherein the constant weight of the dynamic attribute index is used to further calculate its adaptive weight vector; Step 3: Use variable weight theory to solve the state variable weight vector of the dynamic attribute index, so as to obtain the adaptive weight vector of each target dynamic attribute index as the weight of the dynamic attribute index, which together with the static attribute weight constitutes the combined weight of the dynamic and static attribute indicators; Step 4: Based on the intuitive fuzzy values ​​and weights of dynamic and static attribute indicators, the intuitive fuzzy TOPSIS method is used to calculate the positive and negative ideal solutions of the intuitive fuzzy values ​​of the target set, and the similarity between each target and the positive and negative ideal solutions is solved to obtain the comprehensive threat degree of each time-sensitive target on the ground and its ranking.

2. The ground target threat assessment method based on intuitionistic fuzzy TOPSIS and adaptive weight according to claim 1 is characterized in that: The intuitionistic fuzzy number of the static attribute index is: In the formula, τ ik The positive intuitive fuzzy number representing the kth static attribute index of the ith target, ζ ik Represents the negative intuitionistic fuzzy number of the kth static attribute index of the i-th target, μ m ,v m is the initial intuitive fuzzy number corresponding to the intermediate scale α = (L-1) / 2, L is the number of divided fuzzy evaluation language scales, and t represents the degree of certainty of the decision maker on the target fuzzy evaluation language he or she obtains.

3. The ground target threat assessment method based on intuitionistic fuzzy TOPSIS and adaptive weight according to claim 1 is characterized in that: The intuitive fuzzy number of the dynamic attribute index is: Where, T il is the membership function value of the lth dynamic attribute index of the ith target, τ il The positive intuitive fuzzy number representing the lth dynamic attribute index of the ith target, ζ il Represents the negative intuitive fuzzy number of the lth dynamic attribute index of the ith target, β g ∈[0,1] is the additional fuzzy factor, and m is the number of incoming targets.

4. The ground target threat assessment method based on intuitionistic fuzzy TOPSIS and adaptive weight according to claim 1 is characterized in that: Step 2 specifically includes: Convert the target dynamic and static attribute indicators into weight analysis values: In the formula, E j is the weight analysis value of the jth attribute index, τ ij is the positive intuitive fuzzy number of the jth attribute index of the ith target, ζ ij is the negative intuitive fuzzy number of the jth attribute index of the i-th target, and m is the number of incoming targets; Establish a quadratic nonlinear programming model based on the weight analysis value of each target attribute weight: In the formula, ω j ≥0, indicating the static attribute objective weight or dynamic attribute constant weight of the jth attribute index, n is the sum of the number of dynamic and static attribute indexes; Solve the quadratic nonlinear programming model to obtain the weights of static attribute indicators and the constant weights of dynamic attribute indicators.

5. The ground target threat assessment method based on intuitionistic fuzzy TOPSIS and adaptive weight according to claim 1 is characterized in that: Step 3 specifically includes: Based on the dynamic attribute constant weights solved in step 2, the state variable weight vector of each target dynamic attribute is obtained using variable weight theory; Perform Hadamard product of the dynamic attribute constant weight vector and the state variable weight vector to calculate the adaptive weight vector of the dynamic attribute index; The adaptive weights of dynamic attributes and the objective weights of static attributes together constitute the combined weight matrix W of dynamic and static attribute indicators.

6. The ground target threat assessment method based on intuitionistic fuzzy TOPSIS and adaptive weight according to claim 5 is characterized in that: The components of the combined weight matrix W are: Where W ij is the weight value of the jth attribute index of the i-th target; is the objective weight value of the jth static attribute index of the ith target, is the adaptive weight of the jxth dynamic attribute index of the ith target, x is the number of static attribute indicators, and n is the sum of the number of dynamic and static attribute indicators.

7. The ground target threat assessment method based on intuitionistic fuzzy TOPSIS and adaptive weight according to claim 1 is characterized in that: Step 4 specifically includes: Select the elements of the best and worst solutions for each attribute indicator; According to the elements of the best solution and the worst solution obtained and the weights of each attribute index, the similarity between each target and the positive ideal solution and the negative ideal solution is calculated; According to the obtained similarity value, the relative closeness of each target is calculated to obtain the comprehensive threat degree of the target and its ranking.

8. The ground target threat assessment method based on intuitionistic fuzzy TOPSIS and adaptive weight according to claim 1 is characterized in that: The formula for calculating the comprehensive threat level is as follows: In the formula, p i is the relative closeness of the i-th target, Th i is the quantitative value of the threat level of the i-th target, is the similarity between the ith target and the positive ideal solution, S i - is the similarity between the ith target and the negative ideal solution, is the similarity between the intuitionistic fuzzy number and the positive ideal solution and the negative ideal solution.

9. The ground target threat assessment method based on intuitionistic fuzzy TOPSIS and adaptive weight according to claim 1 is characterized in that: Respectively represent the positive intuitive fuzzy number τ of the jth attribute index ij Positive ideal solution and negative intuitionistic fuzzy number ζ ij The positive ideal solution of , the negative ideal solution of positive intuitionistic fuzzy number, and the negative ideal solution of negative intuitionistic fuzzy number, m is the number of incoming targets.

Images