Multi-stage missile reliability test data fusion factor determination method and system based on entropy theory

Abstract

The invention discloses a multi-stage missile reliability test data fusion factor determination method and system based on an entropy theory, and belongs to the technical field of equipment general quality characteristic evaluation. The method comprises the following steps: determining a first field sample conditional entropy under the condition of no historical sample data by utilizing field sample data; determining a second field sample conditional entropy under the condition of the historical sample data of each stage by utilizing the historical sample data and the field sample data of each stage; calculating a missile reliability test data information entropy loss ratio of each stage by using the first field sample conditional entropy and the second field sample conditional entropy; and performing normalization processing on the missile reliability test data information entropy loss ratio of each stage to determine a missile reliability test data fusion factor of each stage. According to the method, contributions of test sample data in different stages are taken as fusion factors, and support is provided for reliability fusion evaluation of test data in subsequent stages.

Description

Technical Field

[0001] The present invention belongs to the technical field of general quality characteristic evaluation of equipment, and particularly relates to a method and system for determining a multi-stage missile reliability test data fusion factor based on entropy theory. Background Technique

[0002] Different from the evaluation indexes of maintainability, testability, supportability, safety, and environmental adaptability, traditional missile reliability is mostly evaluated by the classical binomial distribution during the type approval stage, that is, the missile is regarded as a success or failure product. This method evaluates the launch flight reliability of the missile only relying on the target range identification flight sub-sample data. When the launch flight reliability and confidence level of the missile are both relatively high, the required number of identification flight test sub-samples and flight success rate will be unacceptably high. Especially for medium and long-range missiles, due to the limitations of R & D costs and development progress, the number of missiles used for identification is often limited. Therefore, comprehensively using the test data in stages such as performance tests, combat tests, and in-service assessments to evaluate missile reliability has become the main method for missile reliability evaluation under limited samples.

[0003] Due to the differences in the characteristics of test data in different stages, especially the test data generated in the ground tests during the performance verification stage. The test data in this stage is generated by simulating and assessing the missile in combination with the mission profile through the ground environmental test chamber. Some scholars fuse the test data in different stages through environmental factors, and determine that the environmental reduction coefficients for the accompanied flight test and flight test are 1, and the environmental reduction coefficient for the environmental test of single-item ultimate stress environment is 1.5. There are also scholars who reduce the pre-test equivalent flight test data on the ground and the target range identification flight test data through the reduction coefficient, and this value usually takes 0.6 - 0.9. However, the above methods are all determined by experience, overly rely on rigid rules, and are greatly affected by subjectivity. However, due to reasons such as the test environment and the missile, the test results of missiles in different stages are different, and the reliability information they contain is also different. Combining the reliability information of the test data itself is of great value for realizing the fusion of test data in each stage. Summary of the Invention

[0004] One of the purposes of the present invention is to provide a method for determining a multi-stage missile reliability test data fusion factor based on entropy theory. This method takes the contributions of test sample data in different stages as the fusion factor, providing support for the reliable fusion evaluation of test data in subsequent stages.

[0005] Another purpose of the present invention is to provide a system for determining a multi-stage missile reliability test data fusion factor based on entropy theory.

[0006] To achieve the above purpose, the present invention is implemented by adopting the following technical solutions:

[0007] A method for determining the data fusion factor of multi-stage missile reliability test based on entropy theory, the method for determining the data fusion factor of multi-stage missile reliability test includes:

[0008] Step S1: Use the on-site sample data to determine the first on-site sample conditional entropy in the case of no historical sample data;

[0009] Step S2: Use the historical sample data of each stage and the on-site sample data to determine the second on-site sample conditional entropy in the case of the historical sample data of each stage;

[0010] Step S3: Use the first on-site sample conditional entropy and the second on-site sample conditional entropy to calculate the information entropy loss ratio of the missile reliability test data of each stage;

[0011] Step S4: Normalize the information entropy loss ratio of the missile reliability test data of each stage to determine the data fusion factor of the missile reliability test data of each stage.

[0012] Furthermore, the specific implementation process of step S1 includes:

[0013] Step S11: Obtain the on-site sample data, and use the beta distribution to determine the first posterior distribution function of the missile reliability of the on-site sample data in the case of no historical sample data;

[0014] In step S11, the on-site sample data includes the on-site test times and the on-site success times;

[0015] Step S12: Perform logarithmic calculation on the first posterior distribution function;

[0016] Step S13: Multiply the first posterior distribution function by the logarithmically calculated first posterior distribution function, and then perform integration to obtain the first on-site sample conditional entropy in the case of no historical sample data.

[0017] Furthermore, in step S11, the first posterior distribution function of the missile reliability of the on-site sample data in the case of no historical sample data is:

[0018]

[0019] where, π 0 (R|X) is the first posterior distribution function of the missile reliability of the on-site sample data in the case of no historical sample data; R is the missile reliability of the on-site sample data X; Γ(*) is the Γ function; n and x are the on-site test times and the on-site success times in the on-site sample data X respectively;

[0020] In step S13, the first on-site sample conditional entropy is:

[0021]

[0022] Among them, H 0 (R) is the conditional entropy of the first scene sample.

[0023] Furthermore, the specific implementation process of step S2 includes:

[0024] Step S21, obtaining historical sample data of each stage, and determining the prior distribution hyperparameters of the missile reliability of the historical sample data of each stage;

[0025] In the step S21, the historical sample data includes the historical number of tests and the historical number of successes;

[0026] Step S22, using the prior distribution hyperparameters and the field sample data, determining a second posterior distribution function of the reliability of the field sample data missile under the historical sample data of each stage;

[0027] Step S23, performing logarithmic calculation on the second posterior distribution function of the missile reliability of the on-site sample data in the case of the historical sample data of each stage;

[0028] Step S24, multiply the second posterior distribution function of the missile reliability of the field sample data under the historical sample data of each stage by the second posterior distribution function of the missile reliability of the field sample data under the historical sample data of the corresponding stage after logarithmic calculation, and then integrate them to obtain the second field sample conditional entropy under the historical sample data of each stage.

[0029] Further, in step S22, the second posterior distribution function is:

[0030] π i (R|(X,X i * )) = β(a+x,b+nx);

[0031] Among them, π i (R|(X,X i * )) is the historical sample data X of the i-th stage i * The second posterior distribution function of the reliability of the field sample data X missile under the condition; R is the reliability of the field sample data X missile; β(*) is the Beta distribution function; n and x are the number of trials and the number of successes in the field sample data X respectively; a and b are the hyperparameters of the prior distribution;

[0032] In step S24, the second on-site sample condition entropy under the historical sample data of each stage is:

[0033]

[0034] Among them, H i (R) is the historical sample data X of the i-th stage i * The conditional entropy of the second on-site sample under the condition of...

[0035] Furthermore, in the step S3, the information entropy loss ratio of the missile reliability test data in each stage is as follows:

[0036]

[0037] Among them, is the information entropy loss ratio of the missile reliability test data in the i-th stage

[0038] Furthermore, in the step S4, the data fusion factor of the missile reliability test data in each stage is as follows:

[0039]

[0040] Among them, ρ i is the data fusion factor of the missile reliability test data in the i-th stage; is the information entropy loss ratio of the missile reliability test data in the j-th stage, j = 1, 2,..., N, and N is the number of stages.

[0041] In order to achieve the second above-mentioned purpose, the present invention is implemented by adopting the following technical solution:

[0042] A system for determining the data fusion factor of multi-stage missile reliability test based on entropy theory, the system for determining the data fusion factor of multi-stage missile reliability test includes:

[0043] The first determination module is used to determine the conditional entropy of the first on-site sample without historical sample data by using the on-site sample data;

[0044] The second determination module is used to determine the conditional entropy of the second on-site sample in the case of historical sample data in each stage by using the historical sample data in each stage and the on-site sample data;

[0045] The calculation module is used to calculate the information entropy loss ratio of the missile reliability test data in each stage by using the conditional entropy of the first on-site sample and the conditional entropy of the second on-site sample;

[0046] The normalization processing module is used to perform normalization processing on the information entropy loss ratio of the missile reliability test data in each stage to determine the data fusion factor of the missile reliability test data in each stage.

[0047] Further, the first determination module includes:

[0048] A first acquisition sub-module, configured to acquire on-site sample data, and use the beta distribution to determine a first posterior distribution function of the missile reliability of the on-site sample data in the case of no historical sample data;

[0049] The on-site sample data includes the number of on-site tests and the number of on-site successes;

[0050] A first logarithm calculation module, configured to perform a logarithm calculation on the first posterior distribution function;

[0051] A first integration sub-module, configured to multiply the first posterior distribution function by the logarithmically calculated first posterior distribution function and then perform integration to obtain a first on-site sample conditional entropy in the case of no historical sample data.

[0052] Further, the second determination module includes:

[0053] A second acquisition sub-module, configured to acquire historical sample data for each stage and determine prior distribution hyperparameters of the missile reliability of the historical sample data for each stage;

[0054] The historical sample data includes the number of historical tests and the number of historical successes;

[0055] A determination sub-module, configured to use the prior distribution hyperparameters and the on-site sample data to determine a second posterior distribution function of the missile reliability of the on-site sample data in the case of the historical sample data for each stage;

[0056] A second logarithm calculation module, configured to perform a logarithm calculation on the second posterior distribution function of the missile reliability of the on-site sample data in the case of the historical sample data for each stage;

[0057] A second integration sub-module, configured to multiply the second posterior distribution function of the missile reliability of the on-site sample data in the case of the historical sample data for each stage by the logarithmically calculated second posterior distribution function of the missile reliability of the on-site sample data in the case of the corresponding stage's historical sample data and then perform integration to obtain a second on-site sample conditional entropy in the case of the historical sample data for each stage.

[0058] In summary, the technical solution of the present invention has the following technical effects:

[0059] The present invention calculates the information entropy loss ratio of the missile reliability test data for each stage by using the first on-site sample conditional entropy in the case of no historical sample data determined from the on-site sample data, and the second on-site sample conditional entropy in the case of historical sample data for each stage determined from the historical sample data and on-site sample data for each stage; and normalizes the information entropy loss ratio of the missile reliability test data for each stage to determine the missile reliability test data fusion factor for each stage, realizing the construction of the missile reliability with data fusion at different stages through the entropy theory, taking the contributions of the test samples at different stages as the fusion factors, and providing support for the reliable fusion evaluation of the test data at subsequent stages. Brief Description of the Drawings

[0060] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for use in the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0061] Figure 1 It is a schematic flow chart of a method for determining the missile reliability test data fusion factor based on the entropy theory according to an embodiment of the present invention. Detailed Embodiments

[0062] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some of the embodiments of the present invention, rather than all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0063] This embodiment provides a method for determining the missile reliability test data fusion factor based on the entropy theory. Referring to Figure 1 , the method for determining the missile reliability test data fusion factor includes:

[0064] Step S1: Use the on-site sample data to determine the first on-site sample conditional entropy in the case of no historical sample data.

[0065] The on-site sample data in this embodiment includes the number of on-site tests and the number of on-site successes. After obtaining the on-site sample data X = (n, x), using the uniform distribution on (0, 1) as the prior distribution of the missile reliability R of the on-site sample data X, the first posterior distribution function of the missile reliability of the on-site sample data in the case of no historical sample data is:

[0066]

[0067] where, π 0 (R|X) is the first posterior distribution function of the missile reliability of the on-site sample data in the case of no historical sample data; R is the missile reliability of the on-site sample data X; Γ(*) is the Γ function; n and x are the number of on-site tests and the number of on-site successes in the on-site sample data X, respectively.

[0068] According to the following formula, calculate the conditional entropy of the on-site sample data in the case of no historical sample data, that is, the information entropy (the first on-site sample conditional entropy) of the missile reliability R of the on-site sample data itself in the case of no historical sample data. The first on-site sample conditional entropy is:

[0069]

[0070] where, H 0 (R) is the first on-site sample conditional entropy.

[0071] In summary, the specific implementation process of this step includes:

[0072] Step S11: Obtain the on-site sample data, and use the beta distribution to determine the first posterior distribution function of the missile reliability of the on-site sample data in the case of no historical sample data;

[0073] Step S12: Perform logarithmic calculation on the first posterior distribution function;

[0074] Step S13: Multiply the first posterior distribution function by the first posterior distribution function after logarithmic calculation, and then perform integration to obtain the first on-site sample conditional entropy in the case of no historical sample data.

[0075] Step S2: Use the historical sample data of each stage and the on-site sample data to determine the second on-site sample conditional entropy in the case of the historical sample data of each stage.

[0076] The historical sample data in this embodiment includes the historical number of tests and the historical number of successes.. Considering the historical sample data X of each stage i * =(n i * ,x i * ), assume the prior distribution of the missile reliability R of the on-site sample data as its conjugate prior distribution β(a, b) (where the hyperparameters a and b can be solved from the historical stage samples), calculate the posterior distribution considering the on-site sample data X=(n, x), and solve the conditional entropy (that is, the second on-site sample conditional entropy in the case of the historical sample data).

[0077] In summary, the specific implementation process of this step includes:

[0078] Step S21, obtaining historical sample data of each stage, and determining the prior distribution hyperparameters of the missile reliability of the historical sample data of each stage;

[0079] Step S22, using the prior distribution hyperparameters and the field sample data, determining a second posterior distribution function of the reliability of the field sample data missile under the historical sample data of each stage;

[0080] The second posterior distribution function in this embodiment is:

[0081] π i (R|(X,X i * )) = β(a+x,b+nx);

[0082] Among them, π i (R|(X,X i * )) is the historical sample data X of the i-th stage i * The second posterior distribution function of the missile reliability of the field sample data X under the condition; R is the missile reliability of the field sample data X; β(*) is the Beta distribution function; n and x are the number of trials and the number of successes in the field sample data X respectively; a and b are the hyperparameters of the prior distribution.

[0083] Step S23, performing logarithmic calculation on the second posterior distribution function of the missile reliability of the on-site sample data in the case of the historical sample data of each stage;

[0084] Step S24, multiply the second posterior distribution function of the missile reliability of the field sample data under the historical sample data of each stage by the second posterior distribution function of the missile reliability of the field sample data under the historical sample data of the corresponding stage after logarithmic calculation, and then integrate them to obtain the second field sample conditional entropy under the historical sample data of each stage.

[0085] The second on-site sample condition entropy under the historical sample data of each stage in this embodiment is:

[0086]

[0087] Among them, H i (R) is the historical sample data X of the i-th stage i * The second scene sample conditional entropy of the case.

[0088] Step S3: Calculate the information entropy loss ratio of the missile reliability test data for each stage by using the first on-site sample conditional entropy and the second on-site sample conditional entropy.

[0089] The information entropy loss ratio of the missile reliability test data for each stage in this embodiment is:

[0090]

[0091] where is the information entropy loss ratio of the missile reliability test data for the i-th stage.

[0092] Step S4: Normalize the information entropy loss ratio of the missile reliability test data for each stage to determine the data fusion factor of the missile reliability test for each stage.

[0093] The data fusion factor of the missile reliability test data for each stage in this embodiment is:

[0094]

[0095] where ρ i is the data fusion factor of the missile reliability test data for the i-th stage; is the information entropy loss ratio of the missile reliability test data for the j-th stage, j = 1, 2,..., N, and N is the number of stages.

[0096] If there is only historical sample data for one stage, then the information entropy loss ratio of the missile reliability test data for this stage does not need to be normalized, and this value is the fusion factor ρ.

[0097] Suppose a certain type of missile has undergone flight tests in two stages. In stage 1, it flew 14 times and the number of successful flights was 14. In the current stage, it flew 22 times and the number of successful flights was 22.

[0098] 1. Calculate the first on-site sample conditional entropy (in the case of no historical sample data):

[0099] n = 22, x = 22;

[0100] a 0 = x + 1 = 23, b 0 = n - x + 1 = 1;

[0101] The first on-site sample conditional entropy is 0.0531.

[0102] 2. Calculate the second on-site sample conditional entropy (in the case of having historical sample data):

[0103] For the ground test converted to (n * = 14, x *= 14.). Calculate the prior distribution hyperparameters of the missile reliability for the historical sample data.

[0104] a 1 = x * + 1 = 15;

[0105] b 1 = n * - x * + 1 = 1;

[0106] The updated posterior distribution parameters are:

[0107]

[0108]

[0109] The conditional entropy of the second - scene sample is 0.34.

[0110] 3. The information entropy loss ratio of the missile reliability test data at each stage:

[0111]

[0112] 4. The data fusion factor of the missile reliability test data at each stage

[0113] Since there is only 1 stage for this historical sample data, the fusion factor is equal to the contribution rate, then:

[0114]

[0115] ρ 2 = 1 - ρ 1 = 0.6403.

[0116] In this embodiment, by using the conditional entropy of the first - scene sample under the condition of no historical sample data determined by the on - site sample data, and the conditional entropy of the second - scene sample under the condition of historical sample data at each stage determined by the historical sample data and on - site sample data at each stage, calculate the information entropy loss ratio of the missile reliability test data at each stage; and normalize the information entropy loss ratio of the missile reliability test data at each stage to determine the data fusion factor of the missile reliability test data at each stage, realizing the construction of the missile reliability with data fusion in different stages through the entropy theory, taking the contributions of test samples in different stages as the fusion factor, and providing support for the reliable fusion evaluation of test data in subsequent stages.

[0117] The technical solution of the above - mentioned embodiment can be implemented by using the technical solution given in the following embodiment:

[0118] Another embodiment provides a system for determining a multi-stage missile reliability test data fusion factor based on entropy theory. The system for determining a multi-stage missile reliability test data fusion factor includes:

[0119] A first determination module, configured to use on-site sample data to determine a first on-site sample conditional entropy in the case of no historical sample data;

[0120] A second determination module, configured to use historical sample data of each stage and the on-site sample data to determine a second on-site sample conditional entropy in the case of historical sample data of each stage;

[0121] A calculation module, configured to calculate an information entropy loss ratio of missile reliability test data for each stage by using the first on-site sample conditional entropy and the second on-site sample conditional entropy;

[0122] A normalization processing module, configured to perform normalization processing on the information entropy loss ratio of missile reliability test data for each stage to determine a missile reliability test data fusion factor for each stage.

[0123] Further, the first determination module includes:

[0124] A first acquisition sub-module, configured to acquire on-site sample data and use the beta distribution to determine a first posterior distribution function of the missile reliability of on-site sample data in the case of no historical sample data;

[0125] The on-site sample data includes the number of on-site tests and the number of on-site successes;

[0126] A first logarithm calculation module, configured to perform logarithm calculation on the first posterior distribution function;

[0127] A first integral sub-module, configured to multiply the first posterior distribution function by the first posterior distribution function after logarithm calculation and then perform integration to obtain a first on-site sample conditional entropy in the case of no historical sample data.

[0128] Further, the second determination module includes:

[0129] A second acquisition sub-module, configured to acquire historical sample data of each stage and determine prior distribution hyperparameters of the missile reliability of historical sample data of each stage;

[0130] The historical sample data includes the number of historical tests and the number of historical successes;

[0131] A determination sub-module, configured to use the prior distribution hyperparameters and on-site sample data to determine a second posterior distribution function of the missile reliability of on-site sample data in the case of historical sample data of each stage;

[0132] The second logarithm calculation module is configured to calculate the logarithm of the second posterior distribution function of the on-site sample data missile reliability under the historical sample data condition of each stage;

[0133] The second integration sub-module is configured to multiply the second posterior distribution function of the on-site sample data missile reliability under the historical sample data condition of each stage by the second posterior distribution function of the on-site sample data missile reliability under the historical sample data condition of the corresponding stage after logarithm calculation, and then perform integration to obtain the second on-site sample conditional entropy under the historical sample data condition of each stage.

[0134] The principles, formulas, and their parameter definitions involved in the above embodiments are all applicable and will not be elaborated here one by one.

[0135] The above embodiments only illustrate several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the appended claims.

Claims

1. A method for determining data fusion factors of multi-stage missile reliability test based on entropy theory, characterized in that: The method for determining the multi-stage missile reliability test data fusion factor comprises: Step S1, using the on-site sample data, determining the first on-site sample conditional entropy in the absence of historical sample data; Step S2, using the historical sample data of each stage and the field sample data, determining the second field sample conditional entropy under the historical sample data of each stage; Step S3, using the first on-site sample condition entropy and the second on-site sample condition entropy, calculating the missile reliability test data information entropy loss ratio of each stage; Step S4: normalize the information entropy loss ratio of the missile reliability test data at each stage to determine the fusion factor of the missile reliability test data at each stage.

2. The method for determining the data fusion factor of the multi-stage missile reliability test according to claim 1 is characterized in that: The specific implementation process of step S1 includes: Step S11, obtaining on-site sample data, and using Beta distribution to determine the first posterior distribution function of the missile reliability of the on-site sample data in the absence of historical sample data; In the step S11, the field sample data includes the number of field tests and the number of field successes; Step S12, performing logarithmic calculation on the first posterior distribution function; Step S13: multiply the first posterior distribution function by the first posterior distribution function after logarithm calculation, and then integrate them to obtain the first on-site sample conditional entropy in the absence of historical sample data.

3. The method for determining the data fusion factor of the multi-stage missile reliability test according to claim 2 is characterized in that: In step S11, the first posterior distribution function of the missile reliability of the on-site sample data in the absence of historical sample data is: Wherein, π0(R|X) is the first posterior distribution function of the missile reliability of the field sample data without historical sample data; R is the missile reliability of the field sample data X; Γ(*) is the Γ function; n and x are the number of field tests and the number of field successes in the field sample data X, respectively; In step S13, the first scene sample conditional entropy is: Among them, H0(R) is the conditional entropy of the first on-site sample.

4. The method for determining the data fusion factor of the multi-stage missile reliability test according to claim 3 is characterized in that: The specific implementation process of step S2 includes: Step S21, obtaining historical sample data of each stage, and determining the prior distribution hyperparameters of the missile reliability of the historical sample data of each stage; In the step S21, the historical sample data includes the historical number of tests and the historical number of successes; Step S22, using the prior distribution hyperparameters and the field sample data, determining a second posterior distribution function of the reliability of the field sample data missile under the historical sample data of each stage; Step S23, performing logarithmic calculation on the second posterior distribution function of the missile reliability of the on-site sample data in the case of the historical sample data of each stage; Step S24, multiply the second posterior distribution function of the missile reliability of the field sample data under the historical sample data of each stage by the second posterior distribution function of the missile reliability of the field sample data under the historical sample data of the corresponding stage after logarithmic calculation, and then integrate them to obtain the second field sample conditional entropy under the historical sample data of each stage.

5. The method for determining the data fusion factor of the multi-stage missile reliability test according to claim 4 is characterized in that: In step S22, the second posterior distribution function is: p i (R|(X,X i * ))=β(a+x,b+nx); Among them, π i (R|(X,X i * )) is the historical sample data X of the i-th stage i * The second posterior distribution function of the reliability of the field sample data X missile under the condition; R is the reliability of the field sample data X missile; β(*) is the Beta distribution function; n and x are the number of trials and the number of successes in the field sample data X respectively; a and b are the hyperparameters of the prior distribution; In step S24, the second on-site sample condition entropy under the historical sample data of each stage is: Among them, H i (R) is the historical sample data X of the i-th stage i * The second scene sample conditional entropy of the case.

6. The method for determining the data fusion factor of the multi-stage missile reliability test according to claim 5 is characterized in that: In step S3, the information entropy loss ratio of the missile reliability test data in each stage is: Among them, K ΔHi(R) is the information entropy loss ratio of missile reliability test data in the i-th phase.

7. The method for determining the data fusion factor of the multi-stage missile reliability test according to claim 6 is characterized in that: In step S4, the missile reliability test data fusion factor of each stage is: Among them, ρ i is the missile reliability test data fusion factor of the i-th phase; is the information entropy loss ratio of the missile reliability test data in the jth stage, j = 1, 2, ..., N, where N is the number of stages.

8. A multi-stage missile reliability test data fusion factor determination system based on entropy theory, characterized in that: The multi-stage missile reliability test data fusion factor determination system includes: A first determination module is used to determine the first on-site sample conditional entropy without historical sample data using on-site sample data; A second determination module is used to determine the second on-site sample condition entropy under the historical sample data of each stage by using the historical sample data of each stage and the on-site sample data; A calculation module, used to calculate the information entropy loss ratio of the missile reliability test data in each stage by using the first on-site sample condition entropy and the second on-site sample condition entropy; The normalization processing module is used to normalize the information entropy loss ratio of the missile reliability test data at each stage to determine the fusion factor of the missile reliability test data at each stage.

9. The multi-stage missile reliability test data fusion factor determination system according to claim 8 is characterized in that: The first determining module comprises: A first acquisition submodule is used to acquire on-site sample data and determine the first posterior distribution function of the missile reliability of the on-site sample data without historical sample data using Beta distribution; The field sample data includes the number of field tests and the number of field successes; A first logarithmic calculation module, used for performing logarithmic calculation on the first posterior distribution function; The first integral submodule is used to multiply the first posterior distribution function by the first posterior distribution function after logarithm calculation, and then integrate to obtain the first on-site sample conditional entropy in the absence of historical sample data.

10. The multi-stage missile reliability test data fusion factor determination system according to claim 9, characterized in that: The second determining module comprises: The second acquisition submodule is used to acquire the historical sample data of each stage and determine the hyperparameters of the prior distribution of the missile reliability of the historical sample data of each stage; The historical sample data includes the historical number of tests and the historical number of successes; A determination submodule, for determining a second posterior distribution function of the missile reliability of the field sample data under the historical sample data condition of each stage by using the prior distribution hyperparameters and the field sample data; A second logarithmic calculation module is used to perform logarithmic calculation on a second posterior distribution function of the missile reliability of the on-site sample data under the historical sample data of each stage; The second integration submodule is used to multiply the second posterior distribution function of the missile reliability of the on-site sample data under the historical sample data of each stage by the second posterior distribution function of the missile reliability of the on-site sample data under the historical sample data of the corresponding stage after logarithmic calculation, and then integrate to obtain the second on-site sample conditional entropy under the historical sample data of each stage.

Images