Automatic weapon automaton fault criterion construction method based on design parameter association

Abstract

The invention provides an automatic weapon automaton fault criterion construction method based on design parameter association, and relates to the technical field of computer simulation. The method comprises the steps of constructing a sample matrix based on design parameter distribution and statistics of an automatic weapon, inputting the sample matrix into an automatic weapon automaton dynamics simulation model, extracting motion parameter response, and fitting the motion parameter response into a proxy model; establishing an association relationship of the design parameters, calculating the contribution degree of each design parameter based on the association relationship, and determining the design parameters of which the contribution degrees are greater than a contribution degree threshold value as key design parameters; and constructing a Bayesian network based on the key design parameters and the proxy model, further performing forward and reverse reasoning to trace the design parameters influencing the fault, obtaining a design parameter constraint range and a motion parameter constraint range, and establishing an automaton fault criterion. The method can be used for quickly positioning fault reasons, improving design parameters and further guiding automatic weapon reliability design.

Description

Technical Field

[0001] The invention relates to the technical field of computer simulation, and in particular to a method for constructing fault criterion of an automatic weapon automaton based on design parameter association. Background Art

[0002] The prior art provides a fault diagnosis method for complex equipment based on a fault tree using fuzzy Bayesian network reasoning. The method establishes a fault tree model for the complex equipment by analyzing the structural composition of the complex equipment. The method uses a fault tree transformation method to construct a Bayesian network topology based on the fault tree. The method uses causal reasoning and diagnostic reasoning in fuzzy Bayesian network reasoning to diagnose the fault (potential fault) nodes in the case.

[0003] The existing technology uses causal reasoning and diagnostic reasoning in Bayesian network reasoning to diagnose the faulty (potential faulty) nodes in the case. However, since the association relationship between design parameters has not been established and the design parameters are relatively independent, they can only be used to diagnose faults, but cannot form fault judgment criteria and are difficult to guide design. Summary of the invention

[0004] In view of the current technical problems that the failure mechanism of automatic weapon automatons is unclear, the failure formation mechanism is unclear, the failure location is difficult, no fault judgment criterion has been formed, the analysis and solution of the fault only rely on qualitative analysis, and the design parameters and motion parameters that affect the fault have not been quantified, the present invention proposes a method and system for constructing automatic weapon automatons fault judgment criterion based on design parameter association.

[0005] The first aspect of the present invention discloses a method for constructing a fault criterion of an automatic weapon automaton based on design parameter association, the method comprising:

[0006] Step S1, constructing a sample matrix based on the design parameter distribution and statistics of the automatic weapon, and inputting the sample matrix into the automatic weapon automaton dynamics simulation model to extract the motion parameter response, and then fitting the automatic weapon automaton motion parameter response proxy model;

[0007] Step S2, establishing an association relationship among design parameters, and calculating the contribution of each design parameter based on the association relationship, and determining a design parameter whose contribution is greater than a contribution threshold as a key design parameter;

[0008] Step S3, constructing a Bayesian network based on the key design parameters and the automatic weapon automaton motion parameter response proxy model, and then tracing back the key design parameters that affect the fault through forward and reverse reasoning, obtaining the key design parameter constraint range and the motion parameter constraint range, and establishing the automaton fault judgment criteria.

[0009] Optionally, the step S1 specifically includes:

[0010] Step S11, constructing a sample matrix based on the design parameter distribution and statistics of the automatic weapon and through Latin hypercube sampling;

[0011] Step S12, inputting the sample matrix into the automatic weapon automaton dynamics simulation model to calculate and extract the motion parameter response;

[0012] Step S13, based on the correspondence between each sample in the sample matrix and the extracted motion parameter response, a proxy model is fitted to obtain an automatic weapon automaton motion parameter response proxy model; the automatic weapon automaton motion parameter response proxy model is used to obtain the motion parameter response corresponding to the fault criterion according to the input design parameters.

[0013] Optionally, in step S13, the type of the automatic weapon robot motion parameter response agent model is one or more of a Kriging model, a radial basis neural network, a support vector regression model, an extended adaptive hybrid agent model, an optimization-based two-layer integrated agent model, a weighted average combination agent model, a weighted average combination agent model, a moving least squares-based variable-fidelity agent model, and a Gaussian process regression model.

[0014] Optionally, the step S1 further includes:

[0015] Step S14, calculate and output at least one motion parameter response through the automatic weapon automaton motion parameter response proxy model, and perform a one-to-one comparison analysis with the motion parameter response calculated by the automatic weapon automaton dynamics simulation model; if the errors are all less than the preset values, it is determined that the automatic weapon automaton motion parameter response proxy model meets the requirements; otherwise, repeat steps S11-S13 until the errors are all less than the preset values.

[0016] Optionally, in step S2, establishing an association relationship of design parameters specifically includes:

[0017] Step S21, dimensionless and normalized the design parameters to obtain standardized design parameters;

[0018] Step S22, by dividing the normalized design parameters by evaluation thresholds, a transaction database is obtained, and then data combinations are performed on the design parameters in the transaction database, and the support of each data combination is calculated to screen out all data combinations greater than the minimum support threshold as frequent item sets, so as to obtain a frequent pattern growth tree;

[0019] Step S23, mining association rules on the frequent pattern growth tree in a bottom-up manner to obtain a corresponding design parameter set; wherein the design parameter set includes a redundant association design parameter set M, a complementary association design parameter set N and an independent design parameter set K.

[0020] Optionally, in step S2, the contribution of each design parameter is calculated based on the association relationship, specifically including:

[0021] Step S24, solving the nominal contribution of the design parameters in the redundant associated design parameter set M and the complementary associated design parameter set N by maximizing the Marichal entropy; wherein,

[0022] The formula for maximizing Marichal entropy is:

[0023]

[0024] In the formula, The actual contribution is The value of Maricha entropy of the redundant associated design parameter set M when Actual contribution Hyperparameters of is the u-th design parameter in the redundant associated design parameter set M, i is numerically equal to |M|, indicating the number of all design parameters in the redundant associated design parameter set M; S is any subset of all subsets in the redundant associated design parameter set M; The weight coefficient of the subset S in the parameter set M is designed for redundant association; Calculate the entropy value; It is the nominal contribution; To merge, is the design parameter The Shapley value in the cooperative game of ; T is a specific subset of size t in the redundant association parameter set M;

[0025] Step S25, based on the pointers of the single design parameter frequent itemsets in all design parameter sets, the nominal contributions of the design parameters in the redundant associated design parameter set M and the complementary associated design parameter set N are corrected to obtain the actual contributions of the design parameters; wherein,

[0026] The uth design parameter in the redundant associated design parameter set M The actual contribution for:

[0027]

[0028] In the formula, is the u-th design parameter in the redundant associated design parameter set M Nominal contribution of vis the pointer of the frequent item set of a single design parameter; m1, m2, ..., mi are the serial numbers of the design parameters in the redundant associated design parameter set M; n1, n2, ..., nj are the serial numbers of the design parameters in the complementary associated design parameter set N; k1, k2, ..., kl are the serial numbers of the design parameters in the independent design parameter set K;

[0029] Complementary association design parameter set The uth design parameter in The actual contribution for:

[0030]

[0031] In the formula, Designing parameter sets for complementary associations The uth design parameter in Nominal contribution of

[0032] Step S26, based on the pointers of the single design parameter frequent itemsets in all design parameter sets, calculate the actual contribution of the design parameters in the independent design parameter set K; wherein,

[0033]

[0034] In the formula, is the u-th design parameter in the independent design parameter set K The actual contribution of ku is the u-th design parameter in the independent design parameter set K Pointer to the frequent itemsets.

[0035] Optionally, in step S3, by tracing back the key design parameters that affect the fault through forward and reverse reasoning, the constraint range of the key design parameters and the constraint range of the motion parameters are obtained, and the automaton fault judgment criterion is established, which specifically includes:

[0036] Step S31, changing the value of the key design parameter, and inputting the changed key design parameter into the automatic weapon automaton motion parameter response agent model, obtaining the variation range of the motion parameter, and determining the automaton fault type based on the forward reasoning ability of the Bayesian network;

[0037] Step S32, changing the fault type, using the reverse reasoning capability of the Bayesian network to obtain the range of motion parameters, and inputting the range of motion parameters into the automatic weapon automaton motion parameter response proxy model, thereby obtaining the constraint range of key design parameters;

[0038] Step S33, based on the reasoning results of step S31 and step S32, a causal tracing analysis is performed to obtain the constraint range of key design parameters and the constraint range of motion parameters, and to establish the automaton fault judgment criteria.

[0039] Optionally, the key design parameters include: extraction timing, chamfer of the outer edge of the shell bottom edge, extractor spring force, ejection window size, distance between the extractor and the ejection window, bullet pushing height, bullet holding mouth size, return spring force, extractor mirror distance, support spring force and feeding resistance.

[0040] The second aspect of the present invention discloses a system for constructing fault criteria of an automatic weapon automaton based on design parameter association, the system comprising:

[0041] The first processing module is configured to construct a sample matrix based on the design parameter distribution and statistics of the automatic weapon, and input the sample matrix into the automatic weapon automaton dynamics simulation model to extract the motion parameter response, and then fit the automatic weapon automaton motion parameter response proxy model;

[0042] The second processing module is configured to establish an association relationship among the design parameters, calculate a contribution of each design parameter based on the association relationship, and determine a design parameter whose contribution is greater than a contribution threshold as a key design parameter;

[0043] The third processing module is configured to build a Bayesian network based on the key design parameters and the automatic weapon automaton motion parameter response agent model, and then trace the key design parameters that affect the fault through forward and reverse reasoning, obtain the key design parameter constraint range and the motion parameter constraint range, and establish the automaton fault judgment criteria.

[0044] The third aspect of the present invention discloses an electronic device. The electronic device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps in the method for constructing a fault criterion of an automatic weapon automaton based on design parameter association described in the first aspect of the present invention are implemented.

[0045] The fourth aspect of the present invention discloses a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the method for constructing a fault criterion of an automatic weapon automaton based on design parameter association described in the first aspect of the present invention are implemented.

[0046] In summary, the present invention establishes the association relationship between design parameters and parameter weights (i.e., parameter contribution), establishes the constraint range of design parameters through fault reproduction, reverse reasoning and tracing, establishes the constraint range of automatic machine motion parameters, and forms fault judgment criteria, which can be used to quickly locate the cause of the fault, improve the design parameters, and further guide the reliability design of automatic weapons. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0048] Figure 1 A flow chart is constructed for the fault judgment criteria of the automatic weapon automaton of the embodiment of the present invention;

[0049] Figure 2 A schematic diagram of a proxy model fitting process according to an embodiment of the present invention;

[0050] Figure 3 This is a flow chart of forward reasoning analysis of automatic weapon automaton faults according to an embodiment of the present invention;

[0051] Figure 4 It is a flow chart of reverse reasoning analysis of automatic weapon automaton fault according to an embodiment of the present invention;

[0052] Figure 5 This is a flow chart of the causal tracing analysis of the automatic weapon automaton failure according to an embodiment of the present invention;

[0053] Figure 6 The figure is a structural diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0055] The first aspect of the present invention discloses a method for constructing fault criterion of automatic weapon automaton based on design parameter association, see Figure 1 , the method comprising:

[0056] Step S1, constructing a sample matrix based on the design parameter distribution and statistics of the automatic weapon, and inputting the sample matrix into the automatic weapon automaton dynamics simulation model to extract the motion parameter response, and then fitting the automatic weapon automaton motion parameter response proxy model;

[0057] Optionally, see Figure 2 , the step S1 specifically comprises:

[0058] Step S11, constructing a sample matrix based on the design parameter distribution and statistics of the automatic weapon and through Latin hypercube sampling;

[0059] Step S12, inputting the sample matrix into the automatic weapon automaton dynamics simulation model to calculate and extract the motion parameter response;

[0060] Optionally, the extracted motion parameter response includes shell ejection speed, recoil speed, bullet pushing speed, return speed, recoil displacement, return displacement, shell position coordinates, bullet position coordinates, etc.

[0061] Step S13, based on the correspondence between each sample in the sample matrix and the extracted motion parameter response, a proxy model is fitted to obtain an automatic weapon automaton motion parameter response proxy model; the automatic weapon automaton motion parameter response proxy model is used to obtain the motion parameter response corresponding to the fault criterion according to the input design parameters.

[0062] Optionally, in step S13, the type of the automatic weapon robot motion parameter response agent model is one or more of a Kriging model, a radial basis neural network, a support vector regression model, an extended adaptive hybrid agent model, an optimization-based two-layer integrated agent model, a weighted average combination agent model, a weighted average combination agent model, a moving least squares-based variable-fidelity agent model, and a Gaussian process regression model.

[0063] Optionally, the agent model is prioritized in the following order: Kriging model, radial basis neural network, support vector regression, extended adaptive hybrid agent model, optimization-based two-layer ensemble agent model, weighted average combination agent model, weighted average combination agent model, moving least squares-based variable-fidelity agent model, Gaussian process regression.

[0064] Optionally, the step S1 further includes:

[0065] Step S14, calculate and output at least one motion parameter response through the automatic weapon automaton motion parameter response proxy model, and perform a one-to-one comparison analysis with the motion parameter response calculated by the automatic weapon automaton dynamics simulation model; if the errors are all less than the preset values, it is determined that the automatic weapon automaton motion parameter response proxy model meets the requirements; otherwise, repeat steps S11-S13 until the errors are all less than the preset values.

[0066] In this step, the proxy model is used to calculate and output four parameters, namely, shell ejection speed, recoil speed, projectile push speed, and return speed. A one-to-one comparison analysis is performed with the four parameters, namely, shell ejection speed, recoil speed, projectile push speed, and return speed, calculated by the automatic weapon automaton dynamics simulation model. The error is less than 10% to meet the requirements. If the requirements are not met, the proxy model needs to be rebuilt by increasing the sample size and checking the response extraction until the errors are less than 10%.

[0067] Step S2, establishing an association relationship among design parameters, and calculating the contribution of each design parameter based on the association relationship, and determining a design parameter whose contribution is greater than a contribution threshold as a key design parameter;

[0068] There are many design parameters affecting automatic weapons, and the relationship between design parameters is complex. Design parameters are mainly divided into three categories: key part size parameters, spring parameters, and chamber pressure load parameters. Accurately identifying associations and optimizing key design parameters are the key to constructing automatic weapon fault judgment criteria. Optionally, the FP-Tree algorithm (frequent pattern growth algorithm) is used to mine the associations and association types between design parameters.

[0069] Optionally, in step S2, establishing an association relationship of design parameters specifically includes:

[0070] Step S21, dimensionless and normalized the design parameters to obtain standardized design parameters;

[0071] Step S22, by dividing the normalized design parameters by evaluation thresholds, a transaction database is obtained, and then data combinations are performed on the design parameters in the transaction database, and the support of each data combination is calculated to screen out all data combinations greater than the minimum support threshold as frequent item sets, so as to obtain a frequent pattern growth tree (FP-Free); each branch of the FP-Free is a frequent item set;

[0072] Step S23, mining association rules on the frequent pattern growth tree in a bottom-up manner to obtain a corresponding design parameter set;

[0073] For the rules to be mined, expert knowledge should be used to screen and determine the association rules, and finally determine the association type (complementary and redundant) between the design parameters. The mining of design parameter association rules can be divided into two stages. The first stage is to find all frequent item sets that meet the minimum support threshold from the design parameter set, and the second stage is to generate association rules that meet the minimum confidence from the frequent item sets.

[0074] After mining association rules, three types of design parameter sets are obtained, namely, redundant associated design parameter set M with redundant associations between design parameters, complementary associated design parameter set N with complementary associations between design parameters, and independent design parameter set K with independent design parameters.

[0075] Optionally, in step S2, the contribution of each design parameter is calculated based on the association relationship, specifically including:

[0076] Step S24, solving the nominal contribution of the design parameters in the redundant associated design parameter set M and the complementary associated design parameter set N by maximizing the Marichal entropy; wherein,

[0077] The formula for maximizing Marichal entropy is:

[0078]

[0079] In the formula, The actual contribution is The value of Maricha entropy of the redundant associated design parameter set M when Actual contribution Hyperparameters of is the u-th design parameter in the redundant associated design parameter set M, i is numerically equal to |M|, indicating the number of all design parameters in the redundant associated design parameter set M; S is any subset of all subsets in the redundant associated design parameter set M; The weight coefficient of the subset S in the parameter set M is designed for redundant association; Calculate the entropy value; It is the nominal contribution; To merge, is the design parameter The Shapley value in the cooperative game of ; T is a specific subset of size t in the redundant association parameter set M;

[0080] In this step, the fuzziness It is a modeling of design parameter values, which is used to express the comprehensive importance of multiple design parameters and the relationship between multiple design parameters, that is, the actual contribution. Marichal entropy (based on information entropy) is used to construct the objective function and establish an optimization model. The contribution of design parameters in the redundant associated design parameter set M is solved by maximizing Marichal entropy.

[0081] In the above formula:

[0082] ;

[0083] ;

[0084] Step S25, based on the pointers of the single design parameter frequent itemsets in all design parameter sets, the nominal contributions of the design parameters in the redundant associated design parameter set M and the complementary associated design parameter set N are corrected to obtain the actual contributions of the design parameters; wherein,

[0085] The uth design parameter in the redundant associated design parameter set M The actual contribution for:

[0086]

[0087] In the formula, is the u-th design parameter in the redundant associated design parameter set M Nominal contribution of v is the pointer of the frequent item set of a single design parameter, which is used to normalize the contribution value; m1, m2, ..., mi are the serial numbers of the design parameters in the redundant associated design parameter set M; n1, n2, ..., nj are the serial numbers of the design parameters in the complementary associated design parameter set N; k1, k2, ..., kl are the serial numbers of the design parameters in the independent design parameter set K;

[0088] The uth design parameter in the complementary associated design parameter set N The actual contribution for:

[0089]

[0090] In the formula, is the u-th design parameter in the complementary associated design parameter set N Nominal contribution of

[0091] The solution of the design parameter contribution in the complementary association design parameter set N is similar to the redundant association design parameter type. The optimization model is constructed with the maximum Marichal entropy as the objective function to solve the complementary association design parameter contribution. The nominal contribution is corrected to obtain the u-th design parameter in the complementary association design parameter set N. The actual contribution .

[0092] Step S26, based on the pointers of the single design parameter frequent itemsets in all design parameter sets, calculate the actual contribution of the design parameters in the independent design parameter set K; wherein,

[0093]

[0094] In the formula, is the u-th design parameter in the independent design parameter set K The actual contribution of kuis the u-th design parameter in the independent design parameter set K Pointer to the frequent itemsets.

[0095] Step S3, constructing a Bayesian network based on the key design parameters and the automatic weapon automaton motion parameter response proxy model, and then tracing back the key design parameters that affect the fault through forward and reverse reasoning, obtaining the key design parameter constraint range and the motion parameter constraint range, and establishing the automaton fault judgment criteria.

[0096] Optionally, in step S3, by tracing back the key design parameters that affect the fault through forward and reverse reasoning, the constraint range of the key design parameters and the constraint range of the motion parameters are obtained, and the automaton fault judgment criterion is established, which specifically includes:

[0097] Step S31, changing the value of the key design parameter, and inputting the changed key design parameter into the automatic weapon automaton motion parameter response agent model, obtaining the variation range of the motion parameter, and determining the automaton fault type based on the forward reasoning ability of the Bayesian network;

[0098] See also Figure 3 In this step, we focus on four typical faults: jammed shells, jammed bullets, empty chambers, and incomplete recoil. We make full use of the Bayesian network forward reasoning model to change the values ​​of design parameters such as the shell extraction timing, recoil spring force, shell ejection height, and shell ejection window size of automatic weapons, analyze the changes in motion parameters such as shell ejection angle, shell ejection speed, and shell ejection speed, as well as the changes in fault occurrence, obtain the range of motion parameters, analyze the occurrence of automatic weapon automaton faults, determine the range of motion parameters based on the fault boundary, and obtain the fault judgment criterion.

[0099] Step S32, changing the fault type, using the reverse reasoning capability of the Bayesian network to obtain the range of motion parameters, and inputting the range of motion parameters into the automatic weapon automaton motion parameter response proxy model, thereby obtaining the constraint range of key design parameters;

[0100] See also Figure 4 ,In this step, the reverse reasoning mode makes full use of the reverse reasoning ,capture ,jam, empty chamber, and inadequate recoil faults are ,set respectively by focusing on the jamming, bullet jamming, empty chamber ,and inadequate recoil speed, according to the changes in motion parameters such as shell extraction timing, recoil spring force, bullet pushing height, ,shell ejection window size, and other design parameters are ,analyzed to guide the design.

[0101] Step S33, based on the reasoning results of step S31 and step S32, a causal tracing analysis is performed to obtain the constraint range of key design parameters and the constraint range of motion parameters, and to establish the automaton fault judgment criteria.

[0102] See also Figure 5In this step, the reasoning results of step S31 and step S32 are fully utilized to carry out causal tracing analysis. Through the constraints of the value ranges of design parameters such as the extraction timing of automatic weapons, the recoil spring force, the ejection height, and the ejection window size, the influence of motion parameters such as the shell ejection angle, the shell ejection speed, and the ejection speed is analyzed. Combined with the fault reproduction, the value ranges of motion parameters such as the shell ejection angle, the shell ejection speed, and the ejection speed that are closely related to the fault are determined.

[0103] Optionally, the key design parameters include: extraction timing, chamfer of the outer edge of the shell bottom edge, extractor spring force, ejection window size, distance between the extractor and the ejection window, bullet pushing height, bullet holding mouth size, return spring force, extractor mirror distance, support spring force and feeding resistance.

[0104] The second aspect of the present invention discloses a system for constructing fault criteria of an automatic weapon automaton based on design parameter association, the system comprising:

[0105] The first processing module is configured to construct a sample matrix based on the design parameter distribution and statistics of the automatic weapon, and input the sample matrix into the automatic weapon automaton dynamics simulation model to extract the motion parameter response, and then fit the automatic weapon automaton motion parameter response proxy model;

[0106] The second processing module is configured to establish an association relationship among the design parameters, calculate a contribution of each design parameter based on the association relationship, and determine a design parameter whose contribution is greater than a contribution threshold as a key design parameter;

[0107] The third processing module is configured to build a Bayesian network based on the key design parameters and the automatic weapon automaton motion parameter response agent model, and then trace the key design parameters that affect the fault through forward and reverse reasoning, obtain the key design parameter constraint range and the motion parameter constraint range, and establish the automaton fault judgment criteria.

[0108] The third aspect of the present invention discloses an electronic device. The electronic device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps in the method for constructing a fault criterion of an automatic weapon automaton based on design parameter association described in the first aspect of the present invention are implemented.

[0109] Figure 6 is a structural diagram of an electronic device according to an embodiment of the present invention, such as Figure 6As shown, the electronic device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the electronic device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, near field communication (NFC) or other technologies. The display screen of the electronic device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the electronic device can be a touch layer covered on the display screen, or a key, trackball or touchpad set on the housing of the electronic device, or an external keyboard, touchpad or mouse, etc.

[0110] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a structural diagram of the part related to the technical solution of the present disclosure, and does not constitute a limitation on the electronic device to which the technical solution of the present application is applied. The specific electronic device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0111] The fourth aspect of the present invention discloses a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the method for constructing a fault criterion of an automatic weapon automaton based on design parameter association described in the first aspect of the present invention are implemented.

[0112] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them. Although the present invention has been described in detail with reference to the above embodiments, it should be understood by those skilled in the art that the technical solutions described in the above embodiments can still be modified, or some or all of the technical features therein can be replaced by equivalents, and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for constructing fault criteria of automatic weapon automaton based on design parameter association, characterized in that: The method comprises: Step S1, constructing a sample matrix based on the design parameter distribution and statistics of the automatic weapon, and inputting the sample matrix into the automatic weapon automaton dynamics simulation model to extract the motion parameter response, and then fitting the automatic weapon automaton motion parameter response proxy model; Step S2, establishing an association relationship among design parameters, and calculating the contribution of each design parameter based on the association relationship, and determining a design parameter whose contribution is greater than a contribution threshold as a key design parameter; Step S3, constructing a Bayesian network based on the key design parameters and the automatic weapon automaton motion parameter response proxy model, and then tracing back the key design parameters that affect the fault through forward and reverse reasoning, obtaining the key design parameter constraint range and the motion parameter constraint range, and establishing the automaton fault judgment criteria.

2. The method according to claim 1, characterized in that The step S1 specifically includes: Step S11, constructing a sample matrix based on the design parameter distribution and statistics of the automatic weapon and through Latin hypercube sampling; Step S12, inputting the sample matrix into the automatic weapon automaton dynamics simulation model, calculating and extracting the motion parameter response; Step S13, based on the correspondence between each sample in the sample matrix and the extracted motion parameter response, a proxy model is fitted to obtain an automatic weapon automaton motion parameter response proxy model; the automatic weapon automaton motion parameter response proxy model is used to obtain the motion parameter response corresponding to the fault criterion according to the input design parameters.

3. The method according to claim 2, characterized in that In step S13, the type of the automatic weapon robot motion parameter response agent model is one or more of a Kriging model, a radial basis neural network, a support vector regression model, an extended adaptive hybrid agent model, an optimization-based two-layer integrated agent model, a weighted average combination agent model, a moving least squares-based variable-fidelity agent model, and a Gaussian process regression model.

4. The method according to claim 3, characterized in that The step S1 further comprises: Step S14, calculate and output at least one motion parameter response through the automatic weapon automaton motion parameter response proxy model, and perform a one-to-one comparison analysis with the motion parameter response calculated by the automatic weapon automaton dynamics simulation model; if the errors are all less than the preset values, it is determined that the automatic weapon automaton motion parameter response proxy model meets the requirements; otherwise, repeat steps S11-S13 until the errors are all less than the preset values.

5. The method according to claim 1, characterized in that: In step S2, establishing an association relationship of design parameters specifically includes: Step S21, dimensionless and normalized the design parameters to obtain standardized design parameters; Step S22, by dividing the normalized design parameters by evaluation thresholds, a transaction database is obtained, and then data combinations are performed on the design parameters in the transaction database, and the support of each data combination is calculated to screen out all data combinations greater than the minimum support threshold as frequent item sets, so as to obtain a frequent pattern growth tree; Step S23, mining association rules on the frequent pattern growth tree in a bottom-up manner to obtain a corresponding design parameter set; wherein the design parameter set includes a redundant association design parameter set M, a complementary association design parameter set N and an independent design parameter set K.

6. The method according to claim 5, characterized in that In step S2, the contribution of each design parameter is calculated based on the association relationship, specifically including: Step S24, solving the nominal contribution of the design parameters in the redundant associated design parameter set M and the complementary associated design parameter set N by maximizing the Marichal entropy; wherein, The formula for maximizing Marichal entropy is: ； ； In the formula, The actual contribution is The value of Maricha entropy of the redundant associated design parameter set M when Actual contribution Hyperparameters of is the u-th design parameter in the redundant associated design parameter set M, i is numerically equal to |M|, indicating the number of all design parameters in the redundant associated design parameter set M; S is any subset of all subsets in the redundant associated design parameter set M; The weight coefficient of the subset S in the parameter set M is designed for redundant association; Calculate the entropy value; It is the nominal contribution; To merge, is the design parameter The Shapley value in the cooperative game of ; T is a specific subset of size t in the redundant association parameter set M; Step S25, based on the pointers of the single design parameter frequent itemsets in all design parameter sets, the nominal contributions of the design parameters in the redundant associated design parameter set M and the complementary associated design parameter set N are corrected to obtain the actual contributions of the design parameters; wherein, The uth design parameter in the redundant associated design parameter set M The actual contribution for: ； In the formula, is the u-th design parameter in the redundant associated design parameter set M Nominal contribution of v is the pointer of the frequent item set of a single design parameter, v is the serial number variable; m1, m2, ..., mi are the serial numbers of each design parameter in the redundant associated design parameter set M; n1, n2, ..., nj are the serial numbers of each design parameter in the complementary associated design parameter set N; k1, k2, ..., kl are the serial numbers of each design parameter in the independent design parameter set K; Complementary association design parameter set The uth design parameter in The actual contribution for: ； In the formula, Designing parameter sets for complementary associations The uth design parameter in Nominal contribution of Step S26, based on the pointers of the single design parameter frequent itemsets in all design parameter sets, calculate the actual contribution of the design parameters in the independent design parameter set K; wherein, ； In the formula, is the u-th design parameter in the independent design parameter set K The actual contribution of ku is the u-th design parameter in the independent design parameter set K Pointer to the frequent itemsets.

7. The method according to claim 1, characterized in that In step S3, by tracing back the key design parameters that affect the fault through forward and reverse reasoning, the constraint range of the key design parameters and the constraint range of the motion parameters are obtained, and the automaton fault judgment criteria are established, which specifically includes: Step S31, changing the value of the key design parameter, and inputting the changed key design parameter into the automatic weapon automaton motion parameter response agent model, obtaining the variation range of the motion parameter, and determining the automaton fault type based on the forward reasoning ability of the Bayesian network; Step S32, changing the fault type, using the reverse reasoning capability of the Bayesian network to obtain the range of motion parameters, and inputting the range of motion parameters into the automatic weapon automaton motion parameter response proxy model, thereby obtaining the constraint range of key design parameters; Step S33, based on the reasoning results of step S31 and step S32, a causal tracing analysis is performed to obtain the constraint range of key design parameters and the constraint range of motion parameters, and to establish the automaton fault judgment criteria.

8. The method according to any one of claims 1 to 7, characterized in that The key design parameters include: extraction timing, chamfer of the outer edge of the shell base, extractor spring force, ejection window size, distance between the extractor and the ejection window, bullet pushing height, bullet holding mouth size, return spring force, extractor mirror distance, support spring force and feeding resistance.

9. A system for constructing fault criteria of automatic weapon automaton based on design parameter association, characterized in that: The system comprises: The first processing module is configured to construct a sample matrix based on the design parameter distribution and statistics of the automatic weapon, and input the sample matrix into the automatic weapon automaton dynamics simulation model to extract the motion parameter response, and then fit the automatic weapon automaton motion parameter response proxy model; The second processing module is configured to establish an association relationship among the design parameters, calculate a contribution of each design parameter based on the association relationship, and determine a design parameter whose contribution is greater than a contribution threshold as a key design parameter; The third processing module is configured to build a Bayesian network based on the key design parameters and the automatic weapon automaton motion parameter response agent model, and then trace the key design parameters that affect the fault through forward and reverse reasoning, obtain the key design parameter constraint range and the motion parameter constraint range, and establish the automaton fault judgment criteria.

10. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the steps in the method for constructing an automatic weapon automatic machine fault judgment criterion based on design parameter association as described in claim 8 are implemented.

Images