An evaluation method for anti-missile early warning combat plan

Abstract

The present invention provides an evaluation method for an anti-missile warning combat plan, the purpose of which is to evaluate multiple anti-missile warning combat plans according to combat capabilities and give a ranking of the advantages and disadvantages of the combat plans. The technical scheme is: firstly, an anti-missile warning combat plan evaluation index system is constructed, and each evaluation index is modeled, then the subjective and objective fusion weights of each evaluation index are calculated, and the evaluation matrix of each combat plan is weighted and calculated, and then each combat plan is evaluated by an improved approximate ideal solution sorting method to achieve the ranking of the advantages and disadvantages of each combat plan. The evaluation index system established by the present invention can comprehensively evaluate the combat capability of the combat plan, and the proposed hierarchical analysis method, the subjective and objective combined weighting method combining standard deviation and independent coefficient, the weight calculation is more scientific and reasonable, and the principal component analysis is introduced into the approximate ideal solution sorting method. By comparing the relative closeness of the principal components of each plan, the originally indistinguishable combat plans are distinguished, and the scope of application of the evaluation method is improved.

Description

Technical Field

[0001] The present invention relates to the field of anti-missile warning technology, and in particular to an evaluation method for an anti-missile warning combat plan. Background Art

[0002] Due to the long range, high speed and strong penetration capability of ballistic missiles, anti-missile early warning operations are highly complex and time-sensitive. On-the-spot decision-making cannot meet the needs of agile command and control. It is necessary to formulate combat plans in advance according to the possible battlefield situation and form a plan library for wartime decision-making. Under the specific battlefield situation, quickly selecting the optimal combat plan from the combat plan library is conducive to shortening the decision-making time and improving the decision-making efficiency. Therefore, the evaluation of combat plans is of great significance for anti-missile early warning operations. To this end, domestic and foreign scholars have conducted relevant research and proposed some evaluation methods, among which the representative one is the approach to ideal solution sorting method.

[0003] The approach to ideal solution sorting method defines the ideal solution and negative ideal solution of the evaluation problem, and evaluates the operation plan based on the distance from the ideal solution and the negative ideal solution. The closer the distance to the ideal solution and the farther the distance to the negative ideal solution, the better the operation plan. This evaluation method has the following advantages: simple principle, can evaluate multiple objects at the same time; fast calculation, high resolution of results; objective evaluation, good rationality and applicability.

[0004] However, when there are several operational plans that are equally far from the ideal solution and the negative ideal solution, the approach to the ideal solution ranking method cannot distinguish these operational plans. In addition, the approach to the ideal solution ranking method does not provide a method for calculating the weights of the evaluation indicators. Summary of the invention

[0005] The purpose of the present invention is to address the deficiencies of the above-mentioned prior art, provide an evaluation method for an anti-missile early warning combat plan, construct an anti-missile early warning combat plan evaluation index system, model each evaluation index, and improve the approach to ideal solution sorting method.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] The present invention provides an evaluation method for an anti-missile early warning combat plan, comprising the following steps:

[0008] S1. Construct an evaluation index system for anti-missile early warning combat plans and model each of the evaluation indicators;

[0009] S2. Construct an evaluation matrix for each operational plan and perform standardization;

[0010] S3, using the analytic hierarchy process to calculate the subjective weight of each evaluation indicator;

[0011] S4, calculating the objective weight of each evaluation index by standard deviation and independent coefficient, and integrating it with the subjective weight;

[0012] S5, weighting the evaluation matrix;

[0013] S6. Calculate the virtual best combat plan and the virtual worst combat plan;

[0014] S7, calculate the relative closeness of each combat plan, and rank them according to their merits to obtain the evaluation results of the combat plans;

[0015] S8. Perform principal component analysis on the combat plans that cannot be sorted, and recalculate the relative closeness after dimensionality reduction.

[0016] Further, the S1 includes:

[0017] S101, Earliest warning time for:

[0018] (1);

[0019] in, The moment when it is detected by the anti-missile early warning system; The time when the ballistic missile is launched; is the total flight time;

[0020] S102, Target Acquisition Probability for:

[0021] (2);

[0022] in, is the number of targets detected by the anti-missile early warning system; The total number of ballistic missile targets that flew through the airspace of the anti-missile early warning system;

[0023] S103, airspace coverage factor for:

[0024] (3);

[0025] in, The detection distance of the early warning equipment; Scan azimuth for the antenna; is the pitch angle; The volume of responsible airspace for early warning equipment;

[0026] S104: The target tracking accuracy is determined by the angle measurement accuracy and the distance measurement accuracy. for:

[0027] (4);

[0028] in, is the weighting factor, and , ; is the angle measurement accuracy; is the ranging accuracy;

[0029] S105, Target Capacity for:

[0030] (5);

[0031] Assume that The tracking time of each target is ;

[0032] S106, Target Capacity for:

[0033] (6);

[0034] in, is the search interval time; For search time; To track the interval time; is the tracking beam dwell time.

[0035] Furthermore, the S1 further includes:

[0036] S107, using the recognition rate and the recognition accuracy to measure the target category recognition quality, the target category recognition quality for:

[0037] (7);

[0038] in, The number of warheads correctly identified; is the total number of warheads; The number of targets judged to be warheads; ;

[0039] S108, target recognition completion time for:

[0040] (8);

[0041] Assume that the target The recognition completion time is ;

[0042] S109, trajectory estimation accuracy for:

[0043] (9);

[0044] Assume that time Trajectory within, prediction points, the predicted value is ,in ,this The actual value of the trajectory corresponding to the predicted value is ; For the The coordinates of the predicted values;

[0045] S110, Estimation accuracy of the emission point for:

[0046] (10);

[0047] in, The real coordinates of the ballistic missile launch point; Estimated launch point coordinates for the early warning system;

[0048] S111, Falling point prediction accuracy for:

[0049] (11)

[0050] in, The coordinates of the landing point predicted by the early warning system.

[0051] Further, the S2 includes:

[0052] S201, calculate the evaluation index value of each combat plan, form an 11-dimensional vector, and record the combat plan The vector is:

[0053]

[0054]

[0055] If yes A battle plan is constructed The evaluation matrix is:

[0056] (13);

[0057] S202, Evaluation Matrix The 1st, 4th, 8th, 9th, 10th and 11th columns are cost indicators, which are standardized as follows:

[0058] (14)

[0059] ;

[0060] Columns 2, 3, 5, 6, and 7 are benefit indicators, which are standardized as follows:

[0061] (15);

[0062]

[0063] The standardized matrix is ​​recorded as .

[0064] Further, the S3 includes:

[0065] S301, compare the evaluation indicators in pairs, and assign quantitative values ​​from 1 to 9 according to their relative importance; the larger the value, the more important the previous indicator is relative to the next indicator. If the previous indicator is not as important as the next indicator, assign a value of to 1, the judgment matrix formed by the comparison results for:

[0066] (16);

[0067] in, For the The indicator for The quantitative value of the relative importance of each indicator;

[0068] S302: Calculate the judgment matrix The consistency test coefficient is:

[0069] (17);

[0070] in, for The maximum eigenvalue of

[0071] like ,but If the consistency requirements are met, proceed to S303.

[0072] like , then you need to Check and adjust until ;

[0073] S303, calculation The maximum eigenvalue of The corresponding eigenvector , and normalized, the calculation formula is:

[0074] (18);

[0075] in, for No. elements; by all Construct the normalized vector ; is the subjective weight of each evaluation indicator.

[0076] Further, the S4 includes:

[0077] S401, calculating the average value of each evaluation index in each combat plan :

[0078] (19);

[0079] Calculate the standard deviation of the values ​​of each evaluation indicator in each combat plan. The standard deviation of the evaluation index is recorded as ,but:

[0080] (20);

[0081] S402, calculate the The evaluation indicators and Correlation coefficient of evaluation index for:

[0082] (twenty one);

[0083] Calculate the independent coefficients of each evaluation index, The independence coefficient of the evaluation index is recorded as ,but:

[0084] (twenty two);

[0085] S403, combined standard deviation and independent coefficient are:

[0086] (twenty three);

[0087] in, is the weighting factor, and , ;

[0088] right The normalization process is:

[0089] (twenty four);

[0090] By all Construct the normalized vector , That is, the objective weight of each evaluation indicator;

[0091] S404: The subjective weight and the objective weight are integrated to obtain the final weight:

[0092] (25);

[0093] in, is the fusion factor, and .

[0094] According to the method for evaluating an anti-missile early warning combat plan according to claim 6, it is characterized in that the specific method for implementing S5 is:

[0095] (26).

[0096] Further, in S6, the calculation of the vector element value is:

[0097] (27);

[0098] in, To remember the best virtual combat plan; It is a virtual worst-case scenario.

[0099] Further, the S7 includes:

[0100] S701. The calculation formula of the Euclidean distance between the combat plan and the virtual optimal combat plan is:

[0101] (28);

[0102] in, For combat plan The Euclidean distance to the virtual optimal combat plan; For combat plan No. element value; The best virtual combat plan element value;

[0103] S702, calculating the Euclidean distance between the combat plan and the virtual worst combat plan:

[0104] (29);

[0105] in, For combat plan The Euclidean distance to the virtual worst combat plan; For combat plan No. element value; The worst possible scenario for the virtual element value;

[0106] S703. Calculate the relative closeness of each combat plan as follows:

[0107] (30);

[0108] in, For combat plan The relative closeness of , the larger the value, the better the combat plan; that is, according to The order of size can be used to evaluate and rank each combat plan;

[0109] If the relative closeness of the various combat plans is not equal, the evaluation ends;

[0110] If there are combat plans with equal relative closeness and cannot be sorted, go to S8.

[0111] Further, the S8 includes:

[0112] S801, assuming that the relative closeness of the combat plans are , recorded in matrix form:

[0113] (31);

[0114] Calculate the mean and standard deviation of each column using the formula:

[0115] (32);

[0116] (33);

[0117] Will Each column is processed using formula (34) to make the variance of each column equal to 1, and the matrix is ​​obtained :

[0118] (34);

[0119] S802, calculation matrix The correlation coefficient of each column is as follows:

[0120] (35);

[0121] in, For the Column and The correlation coefficients of the columns, all the correlation coefficients constitute the correlation coefficient matrix :

[0122] (36);

[0123] S803: Calculate the correlation coefficient matrix The characteristic value of , , the corresponding eigenvector is , the number of principal components for:

[0124] (37);

[0125] in, is the cumulative contribution rate to be achieved, , in order to retain more information after dimensionality reduction, ;

[0126] After determining the number of principal components, the operational plan data is projected onto the principal axis to achieve dimensionality reduction. The projection method is to linearly combine the operational plan data with the eigenvectors as coefficients:

[0127] (38);

[0128] in, For the The first battle plan principal component; is the eigenvector, through projection, Mapping , Reason The column vector composed of

[0129] S804, As input, the relative closeness is calculated according to S6 and S7, and the combat plan is adjusted according to the relative closeness. Sort.

[0130] The beneficial effects of the present invention are as follows: the constructed anti-missile early warning combat plan evaluation index system can comprehensively evaluate the combat capability of the anti-missile early warning combat plan from four aspects, namely, target detection capability, target tracking capability, target recognition capability and trajectory calculation capability.

[0131] Compared with the approach to ideal solution sorting method, it has better operational plan evaluation differentiation. The approach to ideal solution sorting method ranks the advantages and disadvantages of each plan by the relative closeness of each plan. However, when the relative closeness of multiple operational plans is equal, it cannot be sorted. The present invention introduces principal component analysis into the approach to ideal solution sorting method, and distinguishes the originally indistinguishable operational plans by comparing the relative closeness of the principal components of each plan, thereby improving the scope of application of the evaluation method.

[0132] Compared with a single subjective or objective weighting method, the present invention adopts a subjective and objective combined weighting method, which can effectively overcome the shortcomings of a single subjective weighting method that is greatly interfered by subjective factors and is not combined with combat plan data; it can effectively overcome the shortcomings of a single objective weighting method that is overly dependent on objective and precise data and completely eliminates the subjective factors of decision makers. BRIEF DESCRIPTION OF THE DRAWINGS

[0133] Figure 1 This is a flow chart of an evaluation method for an anti-missile early warning combat plan of the present invention;

[0134] Figure 2 An anti-missile early warning combat plan evaluation index system established by S1 of the present invention;

[0135] Figure 3 A flow chart of the evaluation matrix of each combat plan constructed by S2 of the present invention and the standardized processing;

[0136] Figure 4 A flowchart of calculating the subjective weight of each evaluation index in S3 of the present invention;

[0137] Figure 5 Flow chart of calculating the objective weight of each evaluation index and fusing it with the subjective weight in S4 of the present invention;

[0138] Figure 6 The flowchart of S7 of the present invention for calculating the relative closeness of each combat plan and ranking them by merit;

[0139] Figure 7 This is a flow chart of the principal component analysis and ranking of S8 of the present invention. DETAILED DESCRIPTION

[0140] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0141] See also Figure 1 , an evaluation method for an anti-missile early warning operation plan, comprising the following steps:

[0142] S1. Construct an evaluation index system for anti-missile early warning combat plans and model each of the evaluation indicators;

[0143] S2. Construct an evaluation matrix for each operational plan and perform standardization;

[0144] S3, using the analytic hierarchy process to calculate the subjective weight of each evaluation indicator;

[0145] S4, calculating the objective weight of each evaluation index by standard deviation and independent coefficient, and integrating it with the subjective weight;

[0146] S5, weighting the evaluation matrix;

[0147] S6. Calculate the virtual best combat plan and the virtual worst combat plan;

[0148] S7, calculate the relative closeness of each combat plan, and rank them according to their merits to obtain the evaluation results of the combat plans;

[0149] S8. Perform principal component analysis on the combat plans that cannot be sorted, and recalculate the relative closeness after dimensionality reduction.

[0150] The S1 includes:

[0151] S101, Earliest warning time for:

[0152] (1);

[0153] in, The moment when it is detected by the anti-missile early warning system; The time when the ballistic missile is launched; is the total flight time;

[0154] S102, Target Acquisition Probability for:

[0155] (2);

[0156] in, is the number of targets detected by the anti-missile early warning system; The total number of ballistic missile targets that flew through the airspace of the anti-missile early warning system;

[0157] S103, airspace coverage factor for:

[0158] (3);

[0159] in, The detection distance of the early warning equipment; Scan azimuth for the antenna; is the pitch angle; The volume of responsible airspace for early warning equipment;

[0160] S104: The target tracking accuracy is determined by the angle measurement accuracy and the distance measurement accuracy. for:

[0161] (4);

[0162] in, is the weighting factor, and , ; is the angle measurement accuracy; is the ranging accuracy;

[0163] S105, Target Capacity for:

[0164] (5);

[0165] Assume that The tracking time of each target is ;

[0166] S106, Target Capacity for:

[0167] (6);

[0168] in, is the search interval time; For search time; To track the interval time; is the tracking beam dwell time.

[0169] The S1 further comprises:

[0170] S107, using the recognition rate and the recognition accuracy to measure the target category recognition quality, the target category recognition quality for:

[0171] (7);

[0172] in, The number of warheads correctly identified; is the total number of warheads; The number of targets judged to be warheads; ;

[0173] S108, target recognition completion time for:

[0174] (8);

[0175] Assume that the target The recognition completion time is ;

[0176] S109, trajectory estimation accuracy for:

[0177] (9);

[0178] Assume that time Trajectory within, prediction points, the predicted value is ,in ,this The actual value of the trajectory corresponding to the predicted value is ; For the The coordinates of the predicted values;

[0179] S110, Estimation accuracy of the emission point for:

[0180] (10);

[0181] in, The real coordinates of the ballistic missile launch point; Estimated launch point coordinates for the early warning system;

[0182] S111, Falling point prediction accuracy for:

[0183] (11)

[0184] in, The coordinates of the landing point predicted by the early warning system.

[0185] The S2 includes:

[0186] S201, calculate the evaluation index value of each combat plan, form an 11-dimensional vector, and record the combat plan The vector is:

[0187]

[0188]

[0189] If yes A battle plan is constructed The evaluation matrix is:

[0190] (13);

[0191] S202, Evaluation Matrix The 1st, 4th, 8th, 9th, 10th and 11th columns are cost indicators, which are standardized as follows:

[0192] (14)

[0193] ;

[0194] Columns 2, 3, 5, 6, and 7 are benefit indicators, which are standardized as follows:

[0195] (15);

[0196]

[0197] The standardized matrix is ​​recorded as .

[0198] The S3 includes:

[0199] S301, compare the evaluation indicators in pairs, and assign quantitative values ​​from 1 to 9 according to their relative importance; the larger the value, the more important the previous indicator is relative to the next indicator. If the previous indicator is not as important as the next indicator, assign a value of to 1, the judgment matrix formed by the comparison results for:

[0200] (16);

[0201] in, For the The indicator for The quantitative value of the relative importance of each indicator;

[0202] S302: Calculate the judgment matrix The consistency test coefficient is:

[0203] (17);

[0204] in, for The maximum eigenvalue of

[0205] like ,but If the consistency requirements are met, proceed to S303.

[0206] like , then you need to Check and adjust until ;

[0207] S303, calculation The maximum eigenvalue of The corresponding eigenvector , and normalized, the calculation formula is:

[0208] (18);

[0209] in, for No. elements; by all Construct the normalized vector ; is the subjective weight of each evaluation indicator.

[0210] The S4 includes:

[0211] S401, calculating the average value of each evaluation index in each combat plan :

[0212] (19);

[0213] Calculate the standard deviation of the values ​​of each evaluation indicator in each combat plan. The standard deviation of the evaluation index is recorded as ,but:

[0214] (20);

[0215] S402, calculate the The evaluation indicators and Correlation coefficient of evaluation index for:

[0216] (twenty one);

[0217] Calculate the independent coefficients of each evaluation index, The independence coefficient of the evaluation index is recorded as ,but:

[0218] (twenty two);

[0219] S403, combined standard deviation and independent coefficient are:

[0220] (twenty three);

[0221] in, is the weighting factor, and , ;

[0222] right The normalization process is:

[0223] (twenty four);

[0224] By all Construct the normalized vector , That is, the objective weight of each evaluation indicator;

[0225] S404: The subjective weight and the objective weight are integrated to obtain the final weight:

[0226] (25);

[0227] in, is the fusion factor, and .

[0228] The specific method for implementing S5 is:

[0229] (26).

[0230] In S6, the calculation of the vector element value is:

[0231] (27);

[0232] in, To remember the best virtual combat plan; It is a virtual worst-case scenario.

[0233] The S7 includes:

[0234] S701. The calculation formula of the Euclidean distance between the combat plan and the virtual optimal combat plan is:

[0235] (28);

[0236] in, For combat plan The Euclidean distance to the virtual optimal combat plan; For combat plan No. element value; The best virtual combat plan element value;

[0237] S702, calculating the Euclidean distance between the combat plan and the virtual worst combat plan:

[0238] (29);

[0239] in, For combat plan The Euclidean distance to the virtual worst combat plan; For combat plan No. element value; The worst possible scenario for the virtual element value;

[0240] S703. Calculate the relative closeness of each combat plan as follows:

[0241] (30);

[0242] in, For combat plan The relative closeness of , the larger the value, the better the combat plan; that is, according to The order of size can be used to evaluate and rank each combat plan;

[0243] If the relative closeness of the various combat plans is not equal, the evaluation ends;

[0244] If there are combat plans with equal relative closeness and cannot be sorted, go to S8.

[0245] The S8 includes:

[0246] S801, assuming that the relative closeness of the combat plans are , recorded in matrix form:

[0247] (31);

[0248] Calculate the mean and standard deviation of each column using the formula:

[0249] (32);

[0250] (33);

[0251] Will Each column is processed using formula (34) to make the variance of each column equal to 1, and the matrix is ​​obtained :

[0252] (34);

[0253] S802, calculation matrix The correlation coefficient of each column is as follows:

[0254] (35);

[0255] in, For the Column and The correlation coefficients of the columns, all the correlation coefficients constitute the correlation coefficient matrix :

[0256] (36);

[0257] S803: Calculate the correlation coefficient matrix The characteristic value of , , the corresponding eigenvector is , the number of principal components for:

[0258] (37);

[0259] in, is the cumulative contribution rate to be achieved, , in order to retain more information after dimensionality reduction, ;

[0260] After determining the number of principal components, the operational plan data is projected onto the principal axis to achieve dimensionality reduction. The projection method is to linearly combine the operational plan data with the eigenvectors as coefficients:

[0261] (38);

[0262] in, For the The first battle plan principal component; is the eigenvector, through projection, Mapping , Reason The column vector composed of

[0263] S804, As input, the relative closeness is calculated according to S6 and S7, and the combat plan is adjusted according to the relative closeness. Sort.

[0264] Table 1 is an example of the anti-missile early warning operation plan evaluation of the present invention. The example has 5 operation plans. Table 1 shows the values ​​of the evaluation indicators of the 5 operation plans. The weight vector calculated by the present invention is [0.111, 0.113, 0.079, 0.103, 0.110, 0.083, 0.101, 0.073, 0.088, 0.067, 0.072]. Among them, the weights of the first, second, fourth, fifth, and seventh indicators are relatively large (all greater than 0.1, and the weights of other indicators are all less than 0.1). The corresponding indicators are the earliest warning time, target interception probability, target tracking accuracy, average tracking time, and target type recognition probability. Anti-missile early warning operations need to provide fast, accurate, continuous, and complete intelligence. The earliest warning time is used to evaluate "fast", the target tracking accuracy and the target type recognition probability are used to evaluate "accurate", the average tracking time is used to evaluate "continuous", and the target interception probability is used to evaluate "complete". It is reasonable to give these five indicators a large weight. The evaluation results of the combat plans are ranked from best to worst as 2>4>3>1>5, among which, all indicators of Plan 2 are the best, so it ranks first, and all indicators of Plan 5 are the worst, so it ranks last. Although some indicators of Plan 4 are not very good, its 1st, 2nd, 4th, 5th, and 7th indicators are only second to Plan 2 and better than any other plan, so it ranks second. Plan 3 and Plan 1 are between Plan 4 and Plan 5. From the above analysis, it can be seen that the evaluation results of the combat plans of the present invention are reasonable.

[0265] Table 1

[0266]

[0267] Table 2 is a performance comparison between the present invention and the approach to ideal solution sorting method. Table 2 (a) is the combat plan data used for testing, with a total of 5 plans, and Table 2 (b) is the evaluation result. The evaluation result of the approach to ideal solution sorting method is that plan 1 is the best and plan 3 is the worst, but it is impossible to evaluate and rank plans 2, 4, and 5; the evaluation result of the present invention method is 1>2>5>4>3, and plans 2, 4, and 5 that cannot be evaluated by the approach to ideal solution sorting method are ranked. This is because the principal component analysis of plans 2, 4, and 5 improves the discrimination of combat plans.

[0268] Table 2(a)

[0269]

[0270]

[0271] The above-mentioned embodiments only express the implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention should be based on the attached claims.

Claims

1. A method for evaluating an anti-missile early warning combat plan, characterized in that: The following steps are involved: S1. Construct an evaluation index system for anti-missile early warning combat plans and model each of the evaluation indicators; S2. Construct an evaluation matrix for each operational plan and perform standardization; S3, using the analytic hierarchy process to calculate the subjective weight of each evaluation indicator; S4, calculating the objective weight of each evaluation index by standard deviation and independent coefficient, and integrating it with the subjective weight; S5, weighting the evaluation matrix; S6. Calculate the virtual best combat plan and the virtual worst combat plan; S7, calculate the relative closeness of each combat plan, and rank them according to their merits to obtain the evaluation results of the combat plans; S8. Perform principal component analysis on the operational plans that cannot be sorted, reduce the dimension, and then recalculate the relative closeness; The S1 includes: S101, Earliest warning time for: （1）； in, The moment when it is detected by the anti-missile early warning system; The time when the ballistic missile is launched; is the total flight time; S102, Target Acquisition Probability for: （2）； in, is the number of targets detected by the anti-missile early warning system; The total number of ballistic missile targets that flew through the airspace of the anti-missile early warning system; S103, airspace coverage factor for: （3）； in, The detection distance of the early warning equipment; Scanning azimuth for the antenna; is the pitch angle; The volume of responsible airspace for early warning equipment; S104: The target tracking accuracy is determined by the angle measurement accuracy and the distance measurement accuracy. for: （4）； in, is the weighting factor, and ; is the angle measurement accuracy; is the ranging accuracy; S105, Target Capacity for: （5）； Among them, The tracking time of each target is ; S106, Target Capacity for: （6）； in, is the search interval; For search time; To track the interval time; To track the beam dwell time; The S1 further comprises: S107, using the recognition rate and the recognition accuracy to measure the target category recognition quality, the target category recognition quality for: （7）； in, The number of warheads correctly identified; is the total number of warheads; The number of targets judged to be warheads; ; S108, target recognition completion time for: （8）； Among them, the target The recognition completion time is ; S109, trajectory estimation accuracy for: （9）； Among them, time Trajectory within, prediction points, the predicted value is , ,this The actual value of the trajectory corresponding to the predicted value is ; For the The coordinates of the predicted values; S110, Estimation accuracy of the emission point for: （10）； in, The real coordinates of the ballistic missile launch point; Estimated launch point coordinates for the early warning system; S111, Falling point prediction accuracy for: （11）; in, The coordinates of the landing point predicted by the early warning system; The S2 includes: S201, calculate the evaluation index value of each combat plan, form an 11-dimensional vector, and record the combat plan The vector is: ； If yes A battle plan is constructed The evaluation matrix is: （13）； S202, Evaluation Matrix The 1st, 4th, 8th, 9th, 10th and 11th columns are cost indicators, which are standardized as follows: （14）； Columns 2, 3, 5, 6, and 7 are benefit indicators, which are standardized as follows: （15）； The standardized matrix is ​​recorded as ; The S3 includes: S301, compare the evaluation indicators in pairs, and assign quantitative values ​​from 1 to 9 according to their relative importance; the larger the value, the more important the previous indicator is relative to the next indicator. If the previous indicator is not as important as the next indicator, assign a value of to 1, the judgment matrix formed by the comparison results for: （16）； in, For the The indicator for The quantitative value of the relative importance of each indicator; S302: Calculate the judgment matrix The consistency test coefficient is: （17）； in, for The maximum eigenvalue of like ,but If the consistency requirements are met, proceed to S303. like , then Check and adjust until ; S303, calculation The maximum eigenvalue of The corresponding eigenvector , and normalized, the calculation formula is: （18）； in, for No. elements; by all Construct the normalized vector ; is the subjective weight of each evaluation indicator; The S4 includes: S401, calculating the average value of each evaluation index in each combat plan : （19）； Calculate the standard deviation of the values ​​of each evaluation indicator in each combat plan. The standard deviation of the evaluation index is recorded as ,but: （20）； S402, calculate the The evaluation indicators and Correlation coefficient of evaluation index for: （21）； Calculate the independent coefficients of each evaluation index, The independence coefficient of the evaluation index is recorded as ,but: （22）； S403, combined standard deviation and independent coefficient are: （23）； in, is the weighting factor, and ; right The normalization process is: （24）； By all Construct the normalized vector , That is, the objective weight of each evaluation indicator; S404: The subjective weight and the objective weight are integrated to obtain the final weight: （25）； in, is the fusion factor, and ; The specific method for implementing S5 is: （26）; In S6, the calculation of the vector element value is: （27）； in, To provide the best virtual combat plan; It is a virtual worst-case scenario; The S7 includes: S701. The calculation formula of the Euclidean distance between the combat plan and the virtual optimal combat plan is: （28）； in, For combat plan The Euclidean distance to the virtual optimal combat plan; For combat plan No. element value; The best virtual combat plan element value; S702, calculating the Euclidean distance between the combat plan and the virtual worst combat plan: （29）； in, For combat plan The Euclidean distance to the virtual worst combat plan; For combat plan No. element value; The worst possible scenario for the virtual element value; S703. Calculate the relative closeness of each combat plan as follows: （30）； in, For combat plan The relative closeness of the two operations is the same. The larger the value, the better the operation plan. If the relative closeness of the operation plans is not equal, the evaluation ends. If there are combat plans with equal relative closeness and cannot be sorted, go to S8; The S8 includes: S801, assuming that the relative closeness of the combat plans are , recorded in matrix form: （31）； Calculate the mean and standard deviation of each column using the formula: （32）； （33）； Will Each column is processed using formula (34) to make the variance of each column equal to 1, and the matrix is ​​obtained : （34）； S802, calculation matrix The correlation coefficient of each column is as follows: （35）； in, For the Column and The correlation coefficients of the columns, all the correlation coefficients constitute the correlation coefficient matrix : （36）； S803: Calculate the correlation coefficient matrix The characteristic value of , , the corresponding eigenvector is , the number of principal components for: （37）； in, is the cumulative contribution rate, ; After determining the number of principal components, the operational plan data is projected onto the principal axis to achieve dimensionality reduction. The projection method is to linearly combine the operational plan data with the eigenvectors as coefficients: （38）； in, For the The first battle plan principal component; is the feature vector No. element values, through projection, Mapping , Reason The column vector composed of S804, As input, the relative closeness is calculated according to S6 and S7, and the combat plan is adjusted according to the relative closeness. Sort.

Images