An equipment adaptive evaluation method and device based on Gaussian process regression
An adaptive weapon and equipment evaluation model was constructed by using the Gaussian process regression method, which solved the problem of unreliability of the model in the grey AHP method and achieved higher evaluation accuracy and adaptability.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA ACADEMY OF ELECTRONICS AND INFORMATION TECHNOLOGY OF CHINA ELECTRONICS TECHNOLOGY GROUP CORPORATION
- Filing Date
- 2022-07-29
- Publication Date
- 2026-04-14
AI Technical Summary
Existing weapon system combat capability assessment methods based on the grey AHP method incorporate subjective factors when constructing the judgment matrix, which prevents the model from adaptively adjusting and reduces the reliability and usability of the assessment model.
The Gaussian process regression method is adopted to obtain a set of performance indicators for multiple levels of weapon systems, calculate the correlation between adjacent levels, construct the connection relationship, and use the Gaussian kernel function to train the evaluation model parameters to establish an adaptive evaluation model.
It improves the accuracy of evaluation calculations, establishes a model that is more adapted to the evaluation data, enhances the model's adaptability, and approximates the real evaluation results.
Smart Images
Figure CN115375103B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of data processing technology, and in particular to an adaptive evaluation method and apparatus for equipment based on Gaussian process regression. Background Technology
[0002] In existing technologies, the weapon system combat capability assessment based on the grey AHP method is based on the theory of scheme decision-making. It compares the importance of each indicator at the same level with respect to the indicators belonging to the previous level in pairs, constructs a pairwise comparison judgment matrix, performs a consistency check on the judgment matrix, and then calculates the relative weight of each indicator from the judgment matrix. After calculating the relative weights of all levels, the system's total effectiveness is obtained by linear weighting and aggregation layer by layer. Specific steps include determining the hierarchical structure, calculating weight coefficients, establishing scoring standards, building an evaluation sample matrix, determining the evaluation grey level, calculating grey evaluation coefficients, calculating grey evaluation vectors and evaluation matrices, comprehensively evaluating Vi and U, calculating and ranking the comprehensive evaluation values, etc.
[0003] The weapon system combat capability assessment method based on the grey AHP method is easy to implement, but the subjective factors introduced when constructing the judgment matrix make the model unable to adapt to the application scenario, thus reducing the reliability and usability of the assessment model. Summary of the Invention
[0004] This application provides an adaptive evaluation method and apparatus for equipment based on Gaussian process regression, which addresses the problems of low reliability and availability in the evaluation of the combat capabilities of existing weapon systems.
[0005] This application provides an equipment adaptive evaluation method based on Gaussian process regression, including:
[0006] Acquire a set of performance indicators at multiple levels of the weapon system;
[0007] The MIC method is used to calculate the correlation between the performance index sets of adjacent levels in turn, so as to construct the connection relationship between adjacent levels based on the correlation.
[0008] An evaluation model for weapons and equipment is constructed based on the set of effectiveness indicators and the correlation between adjacent levels.
[0009] The model parameters of the evaluation model for the weapon system are trained using the following method:
[0010] Select one indicator from the set of upper-level performance indicators as the upper-level indicator, and take the lower-level indicators from the evaluation model that have a connection relationship with the upper-level indicator to form an evaluation unit.
[0011] The evaluation unit is used to establish a mapping relationship with the Gaussian kernel function in order to train the model parameters;
[0012] For the target sample of weapons and equipment to be evaluated:
[0013] Based on the target sample, determine the set of effectiveness indicators for each layer of the weapon system to be evaluated;
[0014] The determined set of performance indicators for each layer of the weapon system to be evaluated is input into the evaluation model to obtain the evaluation results.
[0015] Optionally, the performance indicator set at multiple levels includes, from top to bottom: system-level indicator set, system-level indicator set, and function-level indicator set;
[0016] The correlation between performance index sets at adjacent levels is calculated sequentially using the MIC method, including:
[0017] The importance of system-level indicators to overall system-level indicators is calculated using the MIC method, resulting in an importance matrix:
[0018]
[0019] Among them im ij ∈[0,1],i∈[1,N],j∈[1,M] represents the importance of the i-th system-level indicator to the j-th system-level indicator.
[0020] Based on the importance threshold α, the connection matrix between system-level indicators and phylogenetic indicators satisfies:
[0021]
[0022] in c ij =1 indicates that there is a connection between the i-th system-level indicator and the j-th system-level indicator;
[0023] Based on this, the connection relationship between the system-level indicator set and the functional-level indicator set is calculated.
[0024] Optionally, constructing an evaluation model for weaponry based on the set of effectiveness indicators and the correlation between adjacent levels includes:
[0025] Based on the set of performance indicators and the correlation between adjacent levels, a hierarchical networked evaluation model is constructed to satisfy the following:
[0026] H = (V, E)
[0027] Where node V and edge E are represented as follows:
[0028]
[0029] Where S3 represents the functional level indicator set, S2 represents the system level indicator set, and S1 represents the architecture level indicator set. These represent the connection relationships between indicators at each level.
[0030] Optionally, constructing a mapping relationship between the evaluation unit and the Gaussian kernel function to perform model parameter training includes:
[0031] Determine the Gaussian kernel function, construct a mapping relationship using the evaluation unit and the Gaussian kernel function, and use the conjugate gradient method and solve for the hyperparameters to satisfy:
[0032]
[0033] In the formula XX m XX l Let m and l be the m-th and l-th groups of n lower-level indicator data, where m, l = 1, 2, ..., n; θ0, η d θ1 and θ1 are hyperparameters in the Gaussian kernel function, d = 1, 2, ..., i c ;Xx m,d 、Xx l,d Indicates XX m XX l The value on the d-th dimension;
[0034] Calculate the covariance matrix The element values of the matrix in row m and column l satisfy:
[0035] C m,l =C(XX) m ,XX l )=k m,l +β -1 δ m,l
[0036] In the formula, β is the noise variable:
[0037]
[0038] The constructed weapon and equipment evaluation model satisfies:
[0039] Model = {H, Param}
[0040] Where Param represents the set of evaluation unit parameters obtained during training. in This represents an n-dimensional column vector composed of the upper-level index values of the evaluation unit. This represents the n×i values of the lower-level indicators of the evaluation unit. c matrix.
[0041] Optionally, if the target sample only contains index values corresponding to functional level indicators, determining the set of effectiveness indicators for each layer of the weapon or equipment to be evaluated based on the target sample includes:
[0042] The system-level metrics are calculated as follows:
[0043] Take system-level indicators Y one by one i As a higher-level indicator, its corresponding set of evaluation unit parameters is taken. This represents the covariance matrix of the evaluation unit. This represents Y in the system-level indicator data matrix D2. i The index value column vector, Indicator Y i lower-level indicators The set of corresponding indicator value matrices corresponds to the functional level indicators Z1, Z2, ..., Z. Q A subset of;
[0044] Take D in The median index Y i The lower-level indicator set corresponds to the indicator data. Pick row j Y represents the j-th sample value of the current evaluation unit, and is calculated using the following formula. i The value of y i :
[0045]
[0046]
[0047]
[0048] This yields the system-level indicator value set D. middle ={d1',d2',…,d N '}, where d i ' represents the value of the calculated i-th system-level indicator;
[0049] Based on this, system-level indicators are calculated.
[0050] Optionally, the determined set of effectiveness indicators for each layer of the weapon system to be evaluated is input into the evaluation model to obtain the evaluation results, including:
[0051] Each system-level indicator is used as a higher-level indicator. The corresponding evaluation unit parameters and lower-level indicator sets and data are then used to calculate the system-level indicator results (D). middle ={d1',d2',…,d N'} is used as the input argument;
[0052] The system-level indicators are calculated as dependent variables, resulting in the set of system-level indicator values D. out ={d1”,d2”,…,d M "}, as the evaluation result.
[0053] This application also proposes an equipment adaptive evaluation device based on Gaussian process regression, including a processor and a memory. The memory stores a computer program, which, when executed by the processor, implements the steps of the aforementioned equipment adaptive evaluation method based on Gaussian process regression.
[0054] This application also proposes a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the aforementioned equipment adaptive evaluation method based on Gaussian process regression.
[0055] This application embodiment mines the connection relationship between indicators at different levels through indicator correlation analysis, and establishes an evaluation model structure that adapts to the evaluation data. The method of this application can fit the objectively existing correlation relationship between evaluation indicators well, and more closely approximate the real evaluation model, thereby improving the accuracy of evaluation calculation.
[0056] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description
[0057] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:
[0058] Figure 1 This is a schematic diagram of the basic process of the equipment adaptive evaluation method according to an embodiment of this application. Detailed Implementation
[0059] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
[0060] This application provides an adaptive equipment evaluation method based on Gaussian process regression, such as... Figure 1 As shown, it includes the following steps:
[0061] In step S101, a set of performance indicators at multiple levels of the weapon system is obtained. In some embodiments, the set of performance indicators at multiple levels includes, from top to bottom: a system-level indicator set, a system-level indicator set, and a functional-level indicator set. Specifically, based on the system assessment requirements, relevant indicators at the system-level, system-level, and functional-level levels can be extracted to form an assessment indicator set.
[0062] Among them, system-level indicators are often performance indicators related to mission tasks, such as task completion rate and task completion time; system-level indicators are often indicators related to system applications, such as target positioning accuracy and decision support time; and functional-level indicators are indicators related to basic resources, such as network transmission speed, information transmission latency, and network bandwidth.
[0063] In this embodiment, the system-level index set is denoted as S1 = {X1, X2, ..., X...} M}, where X i Let i = 1, 2, ..., M represent the i-th system-level indicator, and M represent the number of system-level indicators. Let the number of system-level indicator data sets be n. Where x ji ,i∈[1,M],j∈[1,n] represent the i-th system-level index X i The j-th sample value.
[0064] Let the system-level index set be S2={Y1,Y2,…,Y N}, where Y i Let i = 1, 2, ..., N represent the i-th system-level indicator, and N represent the number of system-level indicators. Let the system-level indicator data be denoted as... Where y ji ,i∈[1,N],j∈[1,n] represent the i-th system-level index Y i The j-th sample value.
[0065] Let the functional level index set be S3={Z1,Z2,…,Z P}, where Z i Let i = 1, 2, ..., Q represent the i-th functional level indicator, and Q represent the number of functional level indicators. Let the functional level indicator data be denoted as... Where z ji ,i∈[1,Q],j∈[1,n] represent the i-th functional level index Z i The j-th sample value.
[0066] In step S102, the MIC method is used to sequentially calculate the correlation between performance indicator sets at adjacent levels, so as to construct the connection relationship between adjacent levels based on the correlation. Specifically, this example uses a top-down analysis method, first analyzing the connection relationship between system-level indicators and system-level indicators. Based on importance analysis, according to the importance threshold, it is determined whether there is a connection relationship between system-level indicators and system-level indicators.
[0067] In step S103, an evaluation model for weapon equipment is constructed based on the set of effectiveness indicators and the degree of correlation between adjacent levels.
[0068] The model parameters of the evaluation model for the weapon system are trained using the following method:
[0069] In step S104, an indicator is selected from the set of upper-level performance indicators as an upper-level indicator, and a lower-level indicator that has a connection relationship with the upper-level indicator is taken from the evaluation model to form an evaluation unit.
[0070] In step S105, a mapping relationship is established between the evaluation unit and the Gaussian kernel function to perform training of the model parameters;
[0071] For the target sample of weapons and equipment to be evaluated:
[0072] In step S106, the set of effectiveness indicators for each layer of the weapon system to be evaluated is determined based on the target sample.
[0073] In step S107, the determined set of performance indicators for each layer of the weapon system to be evaluated is input into the evaluation model to obtain the evaluation results.
[0074] This application embodiment mines the connection relationship between indicators at different levels through indicator correlation analysis, and establishes an evaluation model structure that adapts to the evaluation data. The method of this application can fit the objectively existing correlation relationship between evaluation indicators well, and more closely approximate the real evaluation model, thereby improving the accuracy of evaluation calculation.
[0075] In some embodiments, calculating the correlation between performance index sets of adjacent levels sequentially using the MIC method includes:
[0076] The importance of system-level indicators to overall system-level indicators is calculated using the MIC method, resulting in an importance matrix:
[0077]
[0078] Among them im ij ∈[0,1],i∈[1,N],j∈[1,M] represents the importance of the i-th system-level indicator to the j-th system-level indicator, im ijThe closer a value is to 1, the more important the i-th system-level indicator is to the j-th system-level indicator; conversely, the less important it is, the lower the importance.
[0079] Based on the importance threshold α, the connection matrix between system-level indicators and phylogenetic indicators satisfies:
[0080]
[0081] in c ij =1 indicates that there is a connection between the i-th system-level indicator and the j-th system-level indicator, c ij =1 indicates that there is a connection between the i-th system-level indicator and the j-th system-level indicator, c ij =0 indicates that there is no connection between the i-th system-level indicator and the j-th system-level indicator.
[0082] Based on this, the connection relationship between the system-level indicator set and the functional-level indicator set is calculated. Specifically, the connection relationship between functional-level indicators and system-level indicators can be analyzed based on the structural analysis of the system-level evaluation model. Similar to the steps mentioned above, the importance of functional indicators to system-level indicators is obtained, the connection relationship between functional-level and system-level indicators is obtained, and thus the system-level evaluation model is obtained.
[0083] In some embodiments, constructing an evaluation model for weaponry based on the set of performance indicators and the correlation between adjacent levels includes:
[0084] Based on the set of performance indicators and the correlation between adjacent levels, a hierarchical networked evaluation model is constructed to satisfy the following:
[0085] H = (V, E)
[0086] Where node V and edge E are represented as follows:
[0087]
[0088] Where S3 represents the functional level indicator set, S2 represents the system level indicator set, and S1 represents the architecture level indicator set. These represent the connections between indicators at each level. The networked evaluation model structure reflects the cascading relationships from functional-level indicators to system-level indicators and then to framework-level indicators.
[0089] The evaluation of the model parameter training process, specifically step S104, first involves selecting one indicator X from the system-level indicator set S1. i i = 1, 2, ..., M, as upper-level indicators, are extracted from the evaluation model structure H and related to X. i System-level metrics with connections As X i The lower-level indicators, using X represents i The lower-level indicator set, i c Indicates with X i The number of lower-level indicators. X i Together with its lower-level indicators, they constitute X i The evaluation unit. Evaluation model parameter training involves extracting evaluation units and calculating unit parameters for each indicator in the system-level and system-level indicator sets. All evaluation units and their corresponding parameters constitute the evaluation model. In some embodiments, using the evaluation units and Gaussian kernel functions to construct a mapping relationship to perform model parameter training includes:
[0090] Determine the Gaussian kernel function, construct a mapping relationship using the evaluation unit and the Gaussian kernel function, and use the conjugate gradient method and solve for the hyperparameters to satisfy:
[0091]
[0092] In the formula XX m XX l Let m and l be the m-th and l-th groups of n lower-level indicator data, where m, l = 1, 2, ..., n; θ0, η d θ1 and θ1 are hyperparameters in the Gaussian kernel function, d = 1, 2, ..., i c ;Xx m,d 、Xx l,d Indicates XX m XX l The value on the d-th dimension;
[0093] Calculate the covariance matrix The element values of the matrix in row m and column l satisfy:
[0094] C m,l =C(XX) m ,XX l )=k m,l +β -1 δ m,l
[0095] In the formula, β is the noise variable:
[0096]
[0097] The constructed weapon and equipment evaluation model satisfies:
[0098] Model = {H, Param}
[0099] Where Param represents the set of evaluation unit parameters obtained during training. in This represents an n-dimensional column vector composed of the upper-level indicator values of the evaluation unit, which can be understood as the independent variable. This represents the n×i values of the lower-level indicators of the evaluation unit. c A matrix can be understood as a dependent variable.
[0100] In some embodiments, for the sample to be predicted, if only the functional level index S3 = {Z1, Z2, ..., Z} is input, Q The corresponding set of index values z in ={z1,z2,…,z Q Based on the target sample, the set of effectiveness indicators for each layer of the weapon system to be evaluated is determined, including:
[0101] The system-level metrics are calculated as follows:
[0102] Take system-level indicators Y one by one i As a higher-level indicator, its corresponding set of evaluation unit parameters is taken. This represents the covariance matrix of the evaluation unit. This represents Y in the system-level indicator data matrix D2. i The index value column vector, Indicator Y i lower-level indicators The set of corresponding indicator value matrices corresponds to the functional level indicators Z1, Z2, ..., Z. Q A subset of;
[0103] Take D in The median index Y i The lower-level indicator set corresponds to the indicator data. Pick row j Y represents the j-th sample value of the current evaluation unit, and is calculated using the following formula. i The value of y i :
[0104]
[0105]
[0106]
[0107] This yields the system-level indicator value set D. middle ={d1', d2', ..., d N '}, where d i ' represents the value of the calculated i-th system-level indicator;
[0108] Each system-level indicator is used as a higher-level indicator, and the corresponding evaluation unit parameters, lower-level indicator sets and data are taken to calculate the system-level indicator.
[0109] In some embodiments, the determined set of effectiveness indicators for each layer of the weapon system to be evaluated is input into the evaluation model to obtain evaluation results, including:
[0110] Each system-level indicator is used as a higher-level indicator. The corresponding evaluation unit parameters and lower-level indicator sets and data are then used to calculate the system-level indicator results (D). middle ={d1',d2',…,d N '} is used as the input argument;
[0111] The system-level indicators are calculated as dependent variables, resulting in the set of system-level indicator values D. out ={d1”,d2”,…,d M "}, as the evaluation result.
[0112] This application employs an adaptive evaluation model construction strategy based on data mining analysis. By mining the relationships between evaluation indicators from objective data, the strategy determines the connections between indicators at different levels and the smallest evaluation unit constituting the evaluation model through indicator correlation analysis during the model construction process. Based on the indicator data, the parameters of the Gaussian process regression model corresponding to the evaluation unit are trained and calculated, thereby realizing the construction and calculation of the evaluation model. Establishing an evaluation model adapted to the scenario data enhances the model's adaptability; utilizing the Gaussian process regression model to quantify the degree of nonlinear correlation between indicators improves the evaluation model's ability to fit the real evaluation object, thus improving the accuracy of the evaluation calculation.
[0113] This application also proposes an equipment adaptive evaluation device based on Gaussian process regression, including a processor and a memory. The memory stores a computer program, which, when executed by the processor, implements the steps of the aforementioned equipment adaptive evaluation method based on Gaussian process regression.
[0114] This application also proposes a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the aforementioned equipment adaptive evaluation method based on Gaussian process regression.
[0115] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.
[0116] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.
[0117] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.
[0118] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims. All of these forms are within the protection scope of this application.
Claims
1. An adaptive equipment evaluation method based on Gaussian process regression, characterized in that, include: Acquire a set of performance indicators at multiple levels of the weapon system; The MIC method is used to calculate the correlation between the performance index sets of adjacent levels in turn, so as to construct the connection relationship between adjacent levels based on the correlation. An evaluation model for weapons and equipment is constructed based on the set of effectiveness indicators and the correlation between adjacent levels. The model parameters of the evaluation model for the weapon system are trained using the following method: Select one indicator from the set of upper-level performance indicators as the upper-level indicator, and take the lower-level indicators from the evaluation model that have a connection relationship with the upper-level indicator to form an evaluation unit. The evaluation unit is used to establish a mapping relationship with the Gaussian kernel function in order to train the model parameters; For the target sample of weapons and equipment to be evaluated: Based on the target sample, determine the set of effectiveness indicators for each layer of the weapon system to be evaluated; The set of performance indicators for each layer of the weapon or equipment to be evaluated is input into the evaluation model to obtain the evaluation results; The performance indicator sets at multiple levels, from top to bottom, include: system-level indicator set, system-level indicator set, and function-level indicator set; The correlation between performance index sets at adjacent levels is calculated sequentially using the MIC method, including: The importance of system-level indicators to overall system-level indicators is calculated using the MIC method, resulting in an importance matrix: in Indicates the first The first system-level indicator for the first The importance of each system-level indicator Based on importance threshold The connection matrix between system-level indicators and system-level indicators satisfies: in , Indicates the first The system-level indicator and the first There are interconnections between the system-level indicators; Based on this, the connection relationship between the system-level indicator set and the functional-level indicator set is calculated. The evaluation model for weapons and equipment, based on the aforementioned set of performance indicators and the correlation between adjacent levels, includes: Based on the set of performance indicators and the correlation between adjacent levels, a hierarchical networked evaluation model is constructed to satisfy the following: Among the nodes and connecting edges They are represented as follows: in Represents a set of functional level indicators. Represents a system-level indicator set. This represents a system-level indicator set. , These are the connection relationships between indicators at each level; The process of establishing a mapping relationship between the evaluation unit and the Gaussian kernel function to train the model parameters includes: Determine the Gaussian kernel function, construct a mapping relationship using the evaluation unit and the Gaussian kernel function, and use the conjugate gradient method and solve for the hyperparameters to satisfy: In the formula , for The first in the lower-level indicator data of the group Group and No. Group, ; , and This is a hyperparameter in the Gaussian kernel function. ; , express , The Middle The value in the dimension; Calculate the covariance matrix The matrix OK The element values of the column satisfy: In the formula For noise variables: The constructed weapon and equipment evaluation model satisfies: in Represents the set of evaluation unit parameters obtained from training. , ,in This indicates the composition of the upper-level indicator values of the evaluation unit. 3D column vector, This indicates the composition of the lower-level indicator values of the evaluation unit. matrix.
2. The equipment adaptive evaluation method based on Gaussian process regression as described in claim 1, characterized in that, When the target sample only contains index values corresponding to functional level indicators, determining the set of effectiveness indicators for each level of the weapon system to be evaluated based on the target sample includes: The system-level metrics are calculated as follows: Take system-level indicators one by one As a higher-level indicator, its corresponding set of evaluation unit parameters is taken. , This represents the covariance matrix of the evaluation unit. Represents system-level indicator data matrix middle The index value column vector, Indicators lower-level indicators The set of corresponding indicator value matrices corresponds to the functional level indicators. A subset of; Pick medium indicators The lower-level indicator set corresponds to the indicator data. ,Pick The OK Indicates the current evaluation unit's number The sample values are calculated using the following formula. The value of : This yields the set of system-level indicator values. ,in , indicating the calculated first The values of each system-level indicator; Based on this, system-level indicators are calculated.
3. The equipment adaptive evaluation method based on Gaussian process regression as described in claim 2, characterized in that, The determined set of effectiveness indicators for each layer of the weapon system to be evaluated is input into the evaluation model to obtain the evaluation results, including: Each system-level indicator is used as a higher-level indicator. The corresponding evaluation unit parameters and lower-level indicator sets and data are then used to calculate the system-level indicators. As an independent variable input; The system-level indicators are calculated as dependent variables, resulting in the set of system-level indicator values. As an evaluation result.
4. An adaptive equipment evaluation device based on Gaussian process regression, characterized in that, It includes a processor and a memory, wherein the memory stores a computer program that, when executed by the processor, implements the steps of the equipment adaptive evaluation method based on Gaussian process regression as described in any one of claims 1 to 3.
5. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the equipment adaptive evaluation method based on Gaussian process regression as described in any one of claims 1 to 3.