Contribution evaluation method for UAV swarm system based on extended confidence rule reasoning

Abstract

The present invention provides a method for evaluating the contribution of a drone cluster system based on extended confidence rule reasoning. The extended confidence rule reasoning method is used to improve the individual matching calculation formula in the extended confidence rule reasoning method, thereby avoiding the defect that the original method cannot process some special data, and applying it to the problem of evaluating the contribution of a drone cluster combat system, which can effectively process the advantages of uncertain information and realize the evaluation of the contribution of a drone cluster combat system. In addition, by improving the calculation formula of the individual matching, the present invention effectively solves the defect that the original extended confidence rule reasoning method cannot operate normally when the individual matching value is negative.

Description

Technical Field

[0001] The present invention relates to the field of combat effectiveness evaluation of unmanned aerial vehicle (UAV) swarm systems, and in particular to a method for evaluating the contribution of unmanned aerial vehicle (UAV) swarm combat systems. Background Art

[0002] With the development of control technology and information technology, drone swarm combat will become a key means of future battlefield confrontation. System contribution, as an indicator for evaluating the role of weapons and equipment in the combat system's completion of combat missions, plays an important role in promoting the development of drone equipment and the construction of drone swarm combat systems.

[0003] At present, some scholars have proposed to use theories and methods such as combat ring (Chinese patent "CN112417710B Weapon equipment system contribution evaluation method based on combat ring") and sensitivity analysis (Chinese patent "CN114139902B Aviation equipment personnel search and rescue system contribution evaluation method based on sensitivity analysis") to solve the problem of system contribution evaluation under certain information conditions. Due to the complexity of the UAV swarm combat system, there are difficulties in obtaining information during the system contribution evaluation process, experts find it difficult to give accurate evaluation information, and some indicators are difficult to obtain accurate values, resulting in some of the information obtained being uncertain information. Existing methods cannot effectively handle this uncertain information, so they are no longer applicable. Summary of the invention

[0004] In order to overcome the shortcomings of the prior art, the present invention provides a method for evaluating the contribution of a drone cluster system based on extended confidence rule reasoning. By using the extended confidence rule reasoning method, the present invention improves the individual matching calculation formula in the extended confidence rule reasoning method to avoid the defect that the original method cannot handle some special data, and applies it to the problem of evaluating the contribution of a drone cluster combat system, which can effectively handle the advantages of uncertain information and realize the evaluation of the contribution of a drone cluster combat system. In addition, by improving the calculation formula of the individual matching, the present invention effectively solves the defect that the original extended confidence rule reasoning method cannot operate normally when the individual matching is negative.

[0005] The present invention aims at the fact that the existing system contribution evaluation method cannot handle the uncertain information in the process of evaluating the contribution of the drone swarm combat system, and uses the extended confidence rule reasoning method to solve the deficiency of the existing method that cannot handle uncertain information. The present invention defines the contribution of the drone swarm combat system as: the degree of gain in the combat effectiveness of the system after the new drone equipment is added to the drone swarm combat system.

[0006] The technical solution adopted by the present invention to solve the technical problem comprises the following steps:

[0007] Step 1: Constructing the UAV swarm combat system contribution evaluation index system;

[0008] Aiming at the problem of evaluating the contribution of drone swarm combat system, an indicator system with three indicator layers is constructed, such as Figure 1 As shown in the figure, in the evaluation index system, the first layer is the combat effectiveness, including the combat effectiveness of the UAV cluster combat system; the second layer is the combat mission layer, including three combat tasks: reconnaissance and perception tasks, command and decision-making tasks, and coordinated combat tasks; the third layer is the combat capability layer, including nine combat capabilities: intelligence search capability, target identification capability, information sharing capability, situation awareness capability, coordinated control capability, mission planning capability, coordinated penetration capability, coordinated strike capability, and damage assessment capability;

[0009] Step 2: Single indicator evaluation;

[0010] Step 2.1: Calculate the individual matching degree of the extended confidence rule;

[0011] Step 2.2: Calculate activation weights;

[0012] Step 2.3: Aggregate the activated extended confidence rules;

[0013] There are T activated extended confidence rules in the extended confidence rule base. The T extended confidence rules are aggregated using the method of evidence reasoning. The calculation formula is:

[0014]

[0015]

[0016] Among them, β n Represents the nth evaluation level D in the evaluation result of the upper-level indicator Z n The confidence level of ; T represents the number of extended confidence rules activated; ω t represents the activation weight of the tth activated extended confidence rule, t = 1, 2, ..., T; represents the confidence of the nth evaluation level of the upper indicator Z in the tth activated extended confidence rule; λ is the intermediate quantity in the calculation process of the evidence reasoning algorithm;

[0017] After evidence reasoning, we get the aggregation result X of the upper-level indicator Z. Z It is expressed as:

[0018] X Z ={(D n ,β n )|n=1,2,…,N} (9);

[0019] Step 2.4: Calculation of single indicator evaluation conclusion;

[0020] According to formula (10) and formula (11), the evaluation conclusion (x Z ,ε Z );

[0021]

[0022]

[0023] Among them, x Z Represents the evaluation value of the upper-level indicator Z; ε Z Represents the evaluation value x Z confidence level;

[0024] Step 3: Calculation of evaluation conclusions of combat mission level indicators;

[0025] In the combat capability layer of the UAV swarm combat system contribution evaluation index system, the index information of intelligence search capability c1, target identification capability c2, information sharing capability c3, situation awareness capability c4, collaborative control capability c5, mission planning capability c6, collaborative penetration capability c7, collaborative strike capability c8 and damage assessment capability c9 correspond to (x c1 ,ε c1 )、(x c2 ,ε c2 )、(x c3 ,ε c3 )、(x c4 ,ε c4 )、(x c5 ,ε c5 )、(x c6 ,ε c6 )、(x c7 ,ε c7 )、(x c8 ,ε c8 ) and (x c9 ,ε c9 );

[0026] According to the corresponding relationship between the combat mission layer and the combat capability layer in the indicator system of step 2, the evaluation conclusions of the reconnaissance and perception task m1, the command and decision-making task m2, and the coordinated combat task m3 are calculated as (x m1 ,ε m1 )、(x m2 ,ε m2 ) and (x m3 ,ε m3 );

[0027] Step 4: Calculation of combat effectiveness of drone swarm combat system;

[0028] (4.1) According to step 2, based on the corresponding relationship between the UAV swarm combat system layer and the combat mission layer in the indicator system, the evaluation conclusion of the UAV swarm combat system is calculated as (x s ,ε s ), where x s represents the evaluation value of the combat system, ε s Indicates the combat system evaluation value x s Confidence level;

[0029] (4.2) Calculate the combat effectiveness value of the UAV swarm combat system;

[0030] Step 5: Calculate the combat effectiveness of new UAV equipment after it is added to the combat system;

[0031] After the new UAV equipment joins the combat system, the index information of intelligence search capability c1, target identification capability c2, information sharing capability c3, situation awareness capability c4, collaborative control capability c5, mission planning capability c6, collaborative penetration capability c7, collaborative strike capability c8 and damage assessment capability c9 will change to Refer to steps 3 to 4 to calculate the combat effectiveness value of the drone swarm combat system after the new drone equipment is added to the combat system.

[0032] Step 6: Calculation of contribution of drone swarm combat system;

[0033] The system contribution of new UAV equipment to the UAV swarm combat system is calculated. By calculating the system contribution of new UAV equipment to the UAV swarm combat system, the gain effect of the new UAV equipment on the system after joining the UAV swarm combat system can be effectively determined, and it can be judged whether the new UAV equipment can effectively improve the combat effectiveness of one's own combat system.

[0034] The steps of calculating the individual matching degree of the extended confidence rule are:

[0035] In the indicator system, the lower-level indicators corresponding to an upper-level indicator Z are {U1,U2,…,U i ,…,U M}, where U i Indicates the i-th lower-level indicator corresponding to the upper-level indicator Z of the indicator system, M indicates the number of lower-level indicators corresponding to the upper-level indicator Z, i = 1, 2, ..., M;

[0036] The i-th lower level indicator U i The evaluation level is The reference value set corresponding to the evaluation level is Among them, A i,j represents the i-th lower layer index Ui The jth evaluation level of i , J i Represents the lower layer index U i The number of evaluation levels; a i,j Indicates evaluation level A i,j The corresponding reference value;

[0037] The evaluation level of the upper index Z is {D1,D2,…,D n ,…,D N}, the reference value set corresponding to the evaluation level is {d1,d2,…,d n ,…,d N}, where D n represents the nth evaluation level of the upper-level indicator Z, n=1,2,…,N, N represents the number of evaluation levels of the upper-level indicator Z, d n Indicates evaluation level D n The corresponding reference value;

[0038] For the upper-level indicator Z, there exists an extended confidence rule base containing K extended confidence rules, where the kth extended confidence rule R k It is expressed as:

[0039] if

[0040] Then Z is

[0041] Among them, R k represents the kth extended confidence rule in the extended confidence rule base, k = 1, 2, ..., K; represents the i-th lower level indicator U in the k-th extended confidence rule i For the jth evaluation level A i,j Confidence level; represents the nth evaluation level D of the upper indicator Z in the kth extended confidence rule n Confidence of k is the rule weight of the kth extended confidence rule in the extended confidence rule base; definition is the relative weight set of the lower-level indicators of the k-th extended confidence rule, where represents the i-th lower level indicator U of the k-th extended confidence rule i The relative weight of

[0042] The indicator information of the lower-level indicators constitutes a set {(x1,ε1),(x2,ε2),…,(x i ,ε i ),…,(x M ,ε M )}, where xi represents the i-th lower layer index U i The value of ε i Indicates the indicator value x i confidence level;

[0043] (2.1.1) Calculate the i-th lower layer index U according to formula (2): i The value of x i Compared with its evaluation grade A i,j Corresponding reference value a i,j Confidence of:

[0044]

[0045] Among them, φ(x i ,a i,j ) is the index value x i Compared with the assessment grade reference value a i,j confidence level;

[0046] (2.1.2) Calculate the i-th lower layer index U according to formula (3): i The value of x i Similarity with the kth extended confidence rule

[0047]

[0048] (2.1.3) Calculate the i-th lower layer index U according to formula (4): i The indicator information (x i ,ε i ) and the individual matching degree of the kth extended confidence rule

[0049]

[0050] Thus, the individual matching degree between each indicator information and each extended confidence rule is obtained.

[0051] The steps of calculating the activation weight are:

[0052] (2.2.1) According to formula (5), the normalized relative weight of the lower-level indicators is calculated:

[0053]

[0054] in, represents the normalized lower-level index U of the k-th extended confidence rule i The relative weight of max{δ k} represents the maximum value of the relative weight set of the lower-level indicators of the k-th extended confidence rule;

[0055] (2.2.2) Calculate the activation weight of each extended confidence rule in the extended confidence rule base according to formula (6):

[0056]

[0057] Among them, ω k represents the activation weight of the kth extended confidence rule;

[0058] After the activation weight of each extended confidence rule is calculated, the extended confidence rules whose activation weight is not 0 are defined as activated extended confidence rules.

[0059] In step 4.2, the formula for calculating the combat effectiveness value of the human-machine cluster combat system is:

[0060]

[0061] Among them, Y s Represents the combat effectiveness value of the UAV swarm combat system; N s Indicates the number of evaluation levels of the UAV swarm combat system; d s,1 and They represent the first and Nth evaluation levels in the evaluation level of the UAV swarm combat system. s A reference value for each assessment level.

[0062] In step 6, the formula for calculating the system contribution of the new drone equipment to the drone cluster combat system is:

[0063]

[0064] Among them, con represents the system contribution of new UAV equipment to the UAV swarm combat system.

[0065] The beneficial effect of the present invention is that the contribution evaluation of the UAV swarm combat system under uncertain information is realized based on the extended confidence rule reasoning method, which effectively solves the problem that the existing method cannot handle the uncertain information in the system contribution evaluation process, and provides decision support for the demonstration, development and application of new UAV equipment. In addition, the present invention effectively solves the defect that the original extended confidence rule reasoning method cannot operate normally when the individual matching degree is negative by improving the calculation formula of the individual matching degree. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 It is a schematic diagram of the contribution evaluation index system of the UAV swarm combat system of the present invention. DETAILED DESCRIPTION

[0067] The present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0068] Step 1: Construction of the UAV swarm combat system contribution evaluation index system

[0069] Establish as Figure 1 The contribution evaluation index system of the UAV swarm combat system is shown in Figure 1. In this evaluation index system, the solid arrows indicate that there is an influence relationship between the upper and lower level indicators, for example, "command decision-making tasks" have an influence relationship with "information sharing capability", "cooperative control capability" and "task planning capability".

[0070] Step 2: Single indicator evaluation

[0071] The following is a detailed explanation using the evaluation of the "command decision-making task" indicator as an example.

[0072] In the indicator evaluation process, the upper-level indicator Z is "command decision-making task", and the corresponding lower-level indicators are: U1-"information sharing capability", U2-"cooperative control capability" and U3-"task planning capability".

[0073] Assume that the evaluation levels of the lower-level indicators “information sharing capability”, “collaborative control capability” and “task planning capability” are all {A1, A2, A3, A4, A5}, representing low, lower, medium, higher and high respectively, and the reference value set corresponding to the evaluation level is {a1=0, a2=0.25, a3=0.5, a4=0.75, a5=1}.

[0074] Assume that the evaluation level of the upper-level indicator “command decision-making task” is {D1, D2, D3, D4, D5}, which respectively represent low, lower, medium, higher and high, and the reference value set corresponding to the evaluation level is {d1=0, d2=0.25, d3=0.5, d4=0.75, d5=1}.

[0075] Assume that there are 125 extended confidence rules for the “command decision-making task” indicator, among which the first extended confidence rule R1 can be expressed as:

[0076] If U1 is {(A1,1.000)}∧U2 is {(A1,1.000)}∧U3 is {(A3,0.9409)(A4,0.0591)}

[0077] Then Z is {(D1,0.8527),(D2,0.1473),(D3,0.000),(D4,0.000),(D5,0.000)}

[0078] The above 125 extended confidence rules can constitute the extended confidence rule base of “command decision-making task”, as shown in Table 1.

[0079] It is assumed that the relative weights of the lower-level indicators “information sharing capability”, “cooperative control capability” and “task planning capability” are δ1=0.3536, δ2=0.2874 and δ3=0.3590 respectively.

[0080] Table 1 Extended confidence rule base for “command decision-making task”

[0081]

[0082]

[0083] Step 2.1: Calculate the individual matching degree of the extended confidence rule

[0084] Assume that the indicator information of "information sharing capability", "cooperative control capability" and "task planning capability" constitutes the set {(0.78, 0.94), (0.73, 0.93), (0.79, 0.91)}, among which the indicator value of "information sharing capability" is 0.78, and the confidence of the indicator value is 0.94.

[0085] (a) According to formula (2), the confidence level of the “information sharing capability” index value of 0.78 and the evaluation level reference value of 0.75 is calculated as follows:

[0086]

[0087] Similarly, the confidence level of the indicator value of “information sharing capability” of 0.78 and the evaluation level reference value of 1 is 0.12, and the confidence level of the other evaluation level reference values ​​is 0.

[0088] Similarly, it can be calculated that the confidence levels of the “collaborative control capability” index value of 0.73 and the reference values ​​of 0, 0.25, 0.5, 0.75 and 1 are 0, 0, 0.0744, 0.8556 and 0 respectively; the confidence levels of the “task planning capability” index value of 0.79 and the reference values ​​of 0, 0.25, 0.5, 0.75 and 1 are 0, 0, 0, 0.7644 and 0.1456 respectively.

[0089] (b) According to formula (3), the similarity between each lower-level indicator and each extended confidence rule is calculated. Taking the similarity and individual matching degree calculation between "information sharing ability" and the 13th extended confidence rule in Table 1 as an example, the similarity between "information sharing ability" and the 13th extended confidence rule is:

[0090]

[0091] (c) According to formula (4), the individual matching degree between each lower-level indicator and each extended confidence rule is calculated. Taking the similarity and individual matching degree calculation between "information sharing ability" and the 13th extended confidence rule in Table 1 as an example, the individual matching degree between "information sharing ability" and the 13th extended confidence rule is:

[0092]

[0093] Similarly, the individual matching degree between “information sharing capability”, “cooperative control capability” and “task planning capability” and each extended confidence rule can be calculated.

[0094] Step 2.2: Calculate activation weights

[0095] (a) According to formula (5), the normalized relative weights of “information sharing capability”, “cooperative control capability” and “task planning capability” are: and

[0096] (b) According to formula (6), the activation weight of each extended confidence rule in the extended confidence rule base is calculated, and the activation weight calculation of the 13th extended confidence rule in Table 1 is used as an example. The rule weight of this extended confidence rule is 0.5790, and the individual matching degrees of "information sharing capability", "cooperative control capability" and "task planning capability" with this extended confidence rule are 0.8519, 0.9088 and 0.8223 respectively. According to formula (6), the activation weight of the 13th extended confidence rule can be calculated to be 0.1729.

[0097] Similarly, the activation weights of all extended confidence rules in the “command decision-making task” extended confidence rule base are calculated, and a total of 5 activated extended confidence rules are obtained, which are the 13th, 14th, 16th, 123rd and 124th extended confidence rules in Table 1.

[0098] Step 2.3: Aggregate activated extended confidence rules

[0099] According to equations (7) and (8), the above five extended confidence rules are aggregated to obtain the aggregation result of “command decision-making task”:

[0100] {(D1,0),(D2,0),(D3,0.0521),(D4,0.7842),(D5,0.1637)}

[0101] Step 2.4: Calculation of single indicator evaluation conclusion

[0102] According to formula (10) and formula (11), the evaluation value of “command decision-making task” is calculated as

[0103] 0×0+0.25×0+0.5×0.0521+0.75×0.7842+1×0.1637=0.7651. The confidence level of the evaluation value is 0.0521+0.7722+0.1594=0.9846. The evaluation conclusion of “command decision-making task” is (0.7651, 0.9846).

[0104] Step 3: Calculation of evaluation conclusions of combat mission level indicators

[0105] Referring to step 2, the evaluation conclusions of the “command decision-making task” and the “cooperative combat task” are calculated.

[0106] Step 4: Calculation of the combat effectiveness of the UAV swarm combat system

[0107] Referring to step 2, the evaluation conclusion of the UAV swarm combat system is calculated. Assuming that the evaluation conclusion of the UAV swarm combat system is (0.7964, 0.9978), the combat effectiveness value of the UAV swarm combat system is calculated according to formula (12):

[0108] Step 5: Calculate the combat effectiveness of new UAV equipment after it joins the combat system

[0109] Refer to steps 3 to 4 to calculate the combat effectiveness value of the system after the new UAV equipment joins the combat system. Assuming that after the new communication UAV joins the system, the "information sharing capability" changes from (0.78, 0.94) to (0.86, 0.96), and the combat effectiveness value of the UAV swarm combat system increases from 0.8041 to 0.8366.

[0110] Step 6: Calculation of contribution of drone swarm combat system

[0111] According to formula (13), the system contribution of the new communication UAV to the UAV swarm combat system is calculated as follows:

Claims

1. A method for evaluating the contribution of drone swarm systems based on extended confidence rule reasoning, characterized in that The steps include: Step 1: Constructing the UAV swarm combat system contribution evaluation index system; Aiming at the problem of evaluating the contribution of drone swarm combat system, an indicator system with three indicator layers is constructed. In the evaluation indicator system, the first layer is the combat effectiveness, which includes the combat effectiveness of drone swarm combat system; the second layer is the combat mission layer, which includes three combat missions: reconnaissance and perception mission, command and decision-making mission, and coordinated combat mission; the third layer is the combat capability layer, which includes nine combat capabilities: intelligence search capability, target identification capability, information sharing capability, situation awareness capability, coordinated control capability, mission planning capability, coordinated penetration capability, coordinated strike capability, and damage assessment capability; Step 2: Single indicator evaluation; Step 2.1: Calculate the individual matching degree of the extended confidence rule; Step 2.2: Calculate activation weights; Step 2.3: Aggregate the activated extended confidence rules; There are T activated extended confidence rules in the extended confidence rule base. The T extended confidence rules are aggregated using the method of evidence reasoning. The calculation formula is: Among them, β n Represents the nth evaluation level D in the evaluation result of the upper-level indicator Z n The confidence level of ; T represents the number of extended confidence rules activated; ω t represents the activation weight of the tth activated extended confidence rule, t = 1, 2, ..., T; β n t represents the confidence of the nth evaluation level of the upper indicator Z in the tth activated extended confidence rule; λ is the intermediate quantity in the calculation process of the evidence reasoning algorithm; After evidence reasoning, we get the aggregation result X of the upper-level indicator Z. Z It is expressed as: X Z ={(D n ,b n )|n=1,2,…,N} (9); Step 2.4: Calculation of single indicator evaluation conclusion; According to formula (10) and formula (11), the evaluation conclusion (x Z ,ε Z ); Among them, x Z Represents the evaluation value of the upper-level indicator Z; ε Z Represents the evaluation value x Z confidence level; Step 3: Calculation of evaluation conclusions of combat mission level indicators; In the combat capability layer of the UAV swarm combat system contribution evaluation index system, the index information of intelligence search capability c1, target identification capability c2, information sharing capability c3, situation awareness capability c4, collaborative control capability c5, mission planning capability c6, collaborative penetration capability c7, collaborative strike capability c8 and damage assessment capability c9 correspond to (x c1 ,ε c1 )、(x c2 ,ε c2 )、(x c3 ,ε c3 )、(x c4 ,ε c4 )、(x c5 ,ε c5 )、(x c6 ,ε c6 )、(x c7 ,ε c7 )、(x c8 ,ε c8 ) and (x c9 ,ε c9 ); According to the corresponding relationship between the combat mission layer and the combat capability layer in the indicator system of step 2, the evaluation conclusions of the reconnaissance and perception task m1, the command and decision-making task m2, and the coordinated combat task m3 are calculated as (x m1 ,ε m1 )、(x m2 ,ε m2 ) and (x m3 ,ε m3 ); Step 4: Calculation of combat effectiveness of drone swarm combat system; (4.1) According to step 2, based on the corresponding relationship between the UAV swarm combat system layer and the combat mission layer in the indicator system, the evaluation conclusion of the UAV swarm combat system is calculated as (x s ,ε s ), where x s represents the evaluation value of the combat system, ε s Indicates the combat system evaluation value x s confidence level; (4.2) Calculate the combat effectiveness value of the UAV swarm combat system; Step 5: Calculate the combat effectiveness of new UAV equipment after it is added to the combat system; After the new UAV equipment joins the combat system, the index information of intelligence search capability c1, target identification capability c2, information sharing capability c3, situation awareness capability c4, collaborative control capability c5, mission planning capability c6, collaborative penetration capability c7, collaborative strike capability c8 and damage assessment capability c9 will change to and Refer to steps 3 to 4 to calculate the combat effectiveness value of the drone swarm combat system after the new drone equipment is added to the combat system. Step 6: Calculation of contribution of drone swarm combat system; The system contribution of new UAV equipment to the UAV swarm combat system is calculated. By calculating the system contribution of new UAV equipment to the UAV swarm combat system, the gain effect of the new UAV equipment on the system after joining the UAV swarm combat system can be effectively determined, and it can be judged whether the new UAV equipment can effectively improve the combat effectiveness of one's own combat system.

2. The method for evaluating the contribution of a drone cluster system based on extended confidence rule reasoning according to claim 1 is characterized in that: The steps of calculating the individual matching degree of the extended confidence rule are: In the indicator system, the lower-level indicators corresponding to an upper-level indicator Z are {U1,U2,…,U i ,…,U M }, where U i Indicates the i-th lower-level indicator corresponding to the upper-level indicator Z of the indicator system, M indicates the number of lower-level indicators corresponding to the upper-level indicator Z, i = 1, 2, ..., M; The i-th lower level indicator U i The evaluation level is The reference value set corresponding to the evaluation level is Among them, A i,j represents the i-th lower layer index U i The jth evaluation level of i , J i Represents the lower layer index U i The number of evaluation levels; a i,j Indicates evaluation level A i,j The corresponding reference value; The evaluation level of the upper index Z is {D1,D2,…,D n ,…,D N }, the reference value set corresponding to the evaluation level is {d1,d2,…,d n ,…,d N }, where D n represents the nth evaluation level of the upper-level indicator Z, n=1,2,…,N, N represents the number of evaluation levels of the upper-level indicator Z, d n Indicates evaluation level D n The corresponding reference value; For the upper-level indicator Z, there exists an extended confidence rule base containing K extended confidence rules, where the kth extended confidence rule R k It is expressed as: Among them, R k represents the kth extended confidence rule in the extended confidence rule base, k = 1, 2, ..., K; represents the i-th lower level indicator U in the k-th extended confidence rule i For the jth evaluation level A i,j confidence level; represents the nth evaluation level D of the upper indicator Z in the kth extended confidence rule n Confidence of k is the rule weight of the kth extended confidence rule in the extended confidence rule base; definition is the relative weight set of the lower-level indicators of the k-th extended confidence rule, where represents the i-th lower level indicator U of the k-th extended confidence rule i The relative weight of The indicator information of the lower-level indicators constitutes a set {(x1,ε1),(x2,ε2),…,(x i ,ε i ),…,(x M ,ε M )}, where x i represents the i-th lower layer index U i The value of ε i Indicates the indicator value x i confidence level; (2.1.1) Calculate the i-th lower layer index U according to formula (2): i The value of x i Compared with its evaluation grade A i,j Corresponding reference value a i,j Confidence of: Among them, φ(x i ,a i,j ) is the index value x i Compared with the assessment grade reference value a i,j confidence level; (2.1.2) Calculate the i-th lower layer index U according to formula (3): i The value of x i Similarity with the kth extended confidence rule (2.1.3) Calculate the i-th lower layer index U according to formula (4): i The indicator information (x i ,ε i ) and the individual matching degree of the kth extended confidence rule Thus, the individual matching degree between each indicator information and each extended confidence rule is obtained.

3. The method for evaluating the contribution of drone cluster systems based on extended confidence rule reasoning according to claim 1 is characterized in that: The steps of calculating the activation weight are: (2.2.1) According to formula (5), the normalized relative weight of the lower-level indicators is calculated: in, represents the normalized lower-level index U of the k-th extended confidence rule i The relative weight of max{δ k } represents the maximum value of the relative weight set of the lower-level indicators of the k-th extended confidence rule; (2.2.2) Calculate the activation weight of each extended confidence rule in the extended confidence rule base according to formula (6): Among them, ω k represents the activation weight of the kth extended confidence rule; After the activation weight of each extended confidence rule is calculated, the extended confidence rules whose activation weight is not 0 are defined as activated extended confidence rules.

4. The method for evaluating the contribution of drone cluster systems based on extended confidence rule reasoning according to claim 1 is characterized in that: In step 4.2, the formula for calculating the combat effectiveness value of the human-machine cluster combat system is: Among them, Y s Represents the combat effectiveness value of the UAV swarm combat system; N s Indicates the number of evaluation levels of the UAV swarm combat system; d s,1 and They represent the first and Nth evaluation levels in the evaluation level of the UAV swarm combat system. s A reference value for each assessment level.

5. The method for evaluating the contribution of drone cluster systems based on extended confidence rule reasoning according to claim 1 is characterized in that: In step 6, the formula for calculating the system contribution of the new drone equipment to the drone cluster combat system is: Among them, con represents the system contribution of new UAV equipment to the UAV swarm combat system.

Images