Multi-fidelity agent model driven solid attitude control engine thrust performance reliability analysis method, device and equipment and storage medium

Abstract

The invention discloses a multi-fidelity agent model driven solid attitude control engine single valve thrust performance reliability analysis method, device and equipment and a storage medium, and relates to the technical field of solid attitude control engines. Generating a first number of first-class samples according to the zero-dimensional inner trajectory mathematical model and the uncertainty variable of the solid attitude control engine; according to the uncertainty variables and the three-dimensional inner trajectory simulation model of the solid attitude control engine, generating a second number of second-class samples; the first number is greater than the second number, and the accuracy of the first-class samples is less than that of the second-class samples; constructing a first agent model according to the first number of the first type of sample thrust and the first type of sample variable, and constructing a second agent model according to the first agent model, the second number of the second type of sample thrust and the second type of sample variable; and analyzing the thrust performance reliability of the solid attitude control engine by using the second agent model. The method can improve the analysis efficiency under the condition of ensuring the analysis accuracy.

Description

Technical Field

[0001] This application relates to the technical field of solid attitude control engines, and in particular, to a method, device, equipment, and storage medium for reliability analysis of the thrust performance of a solid attitude control engine driven by a multi-fidelity surrogate model. Background Art

[0002] As an actuator of a kinetic energy interceptor, the attitude control engine has a simple structure, is safe and reliable, and is particularly suitable for interception weapon systems with high safety requirements such as air-based and sea-based systems. The magnitude and variation law of the thrust of the attitude control engine can directly reflect the working performance of the engine. During the working process, by adjusting the relative position of the throttle plug and the nozzle, real-time adjustment of the nozzle flow rate and thrust is achieved. Due to the intervention of the throttle plug, the internal flow field simulation, heat transfer, and ablation mechanism of the throttle plug nozzle become complex, making it difficult to establish an accurate simulation model, resulting in a difference between the actual thrust and the predetermined thrust, thereby affecting the control performance. Therefore, it is of great significance to perform reliability analysis of the maximum output thrust performance under the conditions of uncertain variables and increase the fault tolerance of the control system design.

[0003] Currently, the actual maximum thrust can be calculated based on simulation experiments or physical experiments. However, since the thrust of the attitude control engine is affected by multiple uncertain variables, although a relatively accurate thrust can be obtained through simulation and experimental methods, this method is extremely time-consuming, especially when many groups of uncertain variables need to be determined, and the analysis efficiency is extremely low. Summary of the Invention

[0004] This application provides a method, device, equipment, and storage medium for reliability analysis of the thrust performance of a solid attitude control engine driven by a multi-fidelity surrogate model, which can improve the analysis efficiency while ensuring the analysis accuracy.

[0005] To achieve the above object, this application adopts the following technical solutions: In a first aspect, this application provides a method for reliability analysis of the thrust performance of a solid attitude control engine driven by a multi-fidelity surrogate model, and the method includes: Obtain the uncertainty variables of the solid attitude control engine, where the uncertainty variables include combustion chamber pressure, throttle plug displacement, throttle plug diameter, throttle plug head length, throat diameter, and convergence transition arc radius; Generate a first number of first-class samples according to the zero-dimensional internal ballistics mathematical model of the solid attitude control engine and the uncertainty variables, where the first-class samples include corresponding first-class sample thrusts and first-class sample variables; According to the uncertainty variables and the three-dimensional interior ballistic simulation model of the solid attitude control engine, a second quantity of second-class samples are generated, and the second-class samples include correspondingly paired second-class sample thrusts and second-class sample variables; the first quantity is greater than the second quantity, and the accuracy of the first-class samples is less than the accuracy of the second-class samples. According to the first quantity of first-class sample thrusts and first-class sample variables, a first surrogate model is constructed, and according to the first surrogate model, the second quantity of second-class sample thrusts and second-class sample variables, a second surrogate model is constructed. Using the second surrogate model, the thrust performance reliability of the solid attitude control engine is analyzed.

[0006] In some possible implementation manners, the generating the first quantity of first-class samples according to the zero-dimensional interior ballistic mathematical model of the solid attitude control engine and the uncertainty variables includes: Substituting the first quantity of uncertainty variables into the following formula to obtain the first quantity of first-class sample thrusts:

[0007]

[0008] where, is the first-class sample thrust, is the thrust coefficient, is the combustion chamber pressure, is the first-class sample throat plug displacement, is the first-class sample throat plug diameter, is the first-class sample throat plug head length, is the first-class sample throat diameter, is the first-class sample convergence transition arc radius, is the first-class sample equivalent throat area; the correspondingly paired first-class sample thrust and first-class sample variables form the first-class samples, and the first-class sample variables include the first-class sample throat plug displacement, the first-class sample throat plug diameter, the first-class sample throat plug head length, the first-class sample throat diameter, and the first-class sample convergence transition arc radius.

[0009] In some possible implementation manners, the thrust coefficient is determined by the following formula:

[0010]

[0011] where, is the thrust coefficient, is the specific heat ratio, is the propellant density, is the ambient pressure, is the cross-sectional area at the nozzle exit.

[0012] In some possible implementation manners, the analyzing the thrust performance reliability of the solid attitude control engine by using the second surrogate model includes: Inputting the variable to be tested into the second surrogate model to obtain a predicted thrust value corresponding to the variable to be tested.

[0013] In some possible implementation manners, the method further includes: Comparing the predicted thrust value with a target thrust value to obtain a comparison result; If the comparison result indicates that the predicted thrust value is greater than or equal to the target thrust value, obtaining a result that the variable to be tested meets the requirements; If the comparison result indicates that the predicted thrust value is less than the target thrust value, obtaining a result that the variable to be tested does not meet the requirements.

[0014] In a second aspect, the present application provides a thrust analysis device for a solid attitude control engine, and the device includes: An acquisition module, configured to acquire uncertainty variables of the solid attitude control engine, where the uncertainty variables include combustion chamber pressure, plug displacement, plug diameter, plug head length, throat diameter, and radius of convergence transition arc; A generation module, configured to generate a first quantity of first-class samples according to the zero-dimensional internal ballistic mathematical model of the solid attitude control engine and the uncertainty variables, where the first-class samples include corresponding first-class sample thrusts and first-class sample variables; generating a second quantity of second-class samples according to the uncertainty variables and the three-dimensional internal ballistic simulation model of the solid attitude control engine, where the second-class samples include corresponding second-class sample thrusts and second-class sample variables; the first quantity is greater than the second quantity, and the accuracy of the first-class samples is less than the accuracy of the second-class samples; A construction module, configured to construct a first surrogate model according to the first quantity of first-class sample thrusts and first-class sample variables, and construct a second surrogate model according to the first surrogate model, the second quantity of second-class sample thrusts, and second-class sample variables; An analysis module, configured to analyze the thrust performance reliability of the solid attitude control engine by using the second surrogate model.

[0015] In some possible implementation manners, the generation module is specifically configured to: Substituting the first quantity of uncertainty variables into the following formula to obtain the first quantity of first-class sample thrusts:

[0016]

[0017] Among them, is the first type of sample thrust, is the thrust coefficient, is the combustion chamber pressure, is the first type of sample throat plug displacement, is the first type of sample throat plug diameter, is the first type of sample throat plug head length, is the first type of sample throat diameter, is the first type of sample convergent transition arc radius, is the first type of sample equivalent throat area; the corresponding first type of sample thrust and the first type of sample variables form the first type of sample, and the first type of sample variables include the first type of sample throat plug displacement, the first type of sample throat plug diameter, the first type of sample throat plug head length, the first type of sample throat diameter, and the first type of sample convergent transition arc radius.

[0018] In some possible implementation manners, the thrust coefficient is determined by the following formula:

[0019]

[0020] Among them, is the thrust coefficient, is the specific heat ratio, is the propellant density, is the ambient pressure, is the cross-sectional area at the nozzle exit section.

[0021] In some possible implementation manners, the analysis module is specifically configured to input the variable to be tested into the second proxy model to obtain the predicted thrust value corresponding to the variable to be tested.

[0022] In some possible implementation manners, the analysis module is further configured to compare the predicted thrust value with the target thrust value to obtain a comparison result; if the comparison result indicates that the predicted thrust value is greater than or equal to the target thrust value, then obtain the result that the variable to be tested meets the requirements; if the comparison result indicates that the predicted thrust value is less than the target thrust value, then obtain the result that the variable to be tested does not meet the requirements.

[0023] In a third aspect, the present application provides a computing device, including a memory and a processor; Wherein, one or more computer programs are stored in the memory, and the one or more computer programs include instructions; when the instructions are executed by the processor, the computing device is caused to execute the method according to any one of the first aspect.

[0024] In a fourth aspect, the present application provides a computer-readable storage medium for storing a computer program for executing the method according to any one of the first aspect.

[0025] From the above technical solutions, it can be seen that the present application has at least the following beneficial effects: The present application provides a method for reliability analysis of the thrust performance of a solid attitude control engine driven by a multi-fidelity surrogate model. The method includes obtaining uncertainty variables of the solid attitude control engine, where the uncertainty variables include combustion chamber pressure, pintle displacement, pintle diameter, pintle head length, throat diameter, and convergence transition arc radius. Then, according to the zero-dimensional internal ballistics data model of the solid attitude control engine and the uncertainty variables, a first number of first-class samples are generated. The first-class samples include corresponding first-class sample thrusts and first-class sample variables. Alternatively, according to the uncertainty variables and the three-dimensional internal ballistics simulation model of the solid attitude control engine, a second number of second-class samples are generated. The second-class samples include corresponding second-class sample thrusts and second-class sample variables, where the first number is greater than the second number, and the accuracy of the first-class samples is less than the accuracy of the second-class samples. That is to say, the number of high-precision samples is small, and the number of low-precision samples is large. Since generating a large number of high-precision samples requires obtaining them through a three-dimensional simulation model, which is time-consuming, only a small number of high-precision samples are generated in the present application, saving time. To increase the number of samples, a large number of low-precision samples are generated by means of a zero-dimensional mathematical model. Then, a first surrogate model is constructed using the low-precision samples, and then, based on the constructed first surrogate model and the high-precision samples, a second surrogate model is constructed. Since some high-precision samples are introduced in the process of constructing the second surrogate model, the accuracy of the second surrogate model can be guaranteed. Moreover, the second surrogate model is obtained by optimizing the first surrogate model constructed from the low-precision samples, and obtaining the low-precision samples is not time-consuming. It can be seen that in this way, the construction efficiency of the surrogate model can be improved while ensuring the analysis accuracy, and further, the analysis efficiency of the reliability of the thrust performance of the solid attitude control engine can be improved.

[0026] It should be understood that the description of technical features, technical solutions, beneficial effects or similar language in this application does not imply that all features and advantages can be realized in any single embodiment. On the contrary, it is understood that the description of a feature or beneficial effect means that a specific technical feature, technical solution or beneficial effect is included in at least one embodiment. Therefore, the description of a technical feature, technical solution or beneficial effect in this specification does not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions and beneficial effects described in the present embodiment can also be combined in any appropriate manner. Those skilled in the art will understand that the embodiment can be implemented without one or more specific technical features, technical solutions or beneficial effects of a specific embodiment. In other embodiments, additional technical features and beneficial effects can also be identified in specific embodiments that do not embody all embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 A flowchart of a multi-fidelity proxy model-driven solid attitude control engine thrust performance reliability analysis method provided in an embodiment of the present application; Figure 2 A schematic diagram of the appearance of a solid attitude control power valve provided in an embodiment of the present application; Figure 3 A flowchart of another multi-fidelity proxy model-driven solid attitude control engine thrust performance reliability analysis method provided in an embodiment of the present application; Figure 4 A schematic diagram of a thrust performance analysis device for a solid attitude control engine provided in an embodiment of the present application; Figure 5 A schematic diagram of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0028] The terms "first", "second" and "third" in this application specification and the accompanying drawings are used to distinguish different objects rather than to limit a specific order.

[0029] In the embodiments of this application, words such as "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of this application should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0030] Currently, the reliability analysis of the maximum output thrust performance of a solid attitude control engine can be carried out in the following ways. (1) Sampling method. This method requires a large number of sample points for simulation. By means of simulation experiments or physical experiments, the actual maximum thrust of the solid attitude control engine is calculated. When the sample size is large enough, the reliability analysis results of the sampling method are basically consistent with the theoretical results. This method is simple to operate and has high robustness. However, based on the sampling method, a large number of simulation samples need to be generated to ensure the convergence of the estimated value of the failure probability, resulting in low computational efficiency. Especially when the theoretical result is small, directly using numerical simulation for reliability analysis will cause an unbearable computational burden.

[0031] (2) Most probable failure point method. This method approximates the performance function and performs a first-order or second-order Taylor expansion of the performance function corresponding to the failure mode of the maximum output thrust of the solid attitude control engine at the most probable failure point. This method is simple and convenient to implement, effectively avoiding complex calculation processes. However, the calculation accuracy of such methods is closely related to the degree of nonlinearity of the variables. The maximum output thrust of a solid attitude control engine is usually a high-dimensional nonlinear problem, and the performance function is generally an implicit performance function, and it can only solve the reliability analysis problem of the single-valve thrust performance where the performance function is explicit. Therefore, the calculation accuracy is difficult to guarantee.

[0032] (3) Numerical integration method. This method obtains the reliability analysis results by calculating the statistical moments of the thrust response of the solid attitude control engine and estimating the probability distribution of the thrust response of the solid attitude control engine. It is a highly robust reliability analysis method. However, this method generally includes two steps: estimating the statistical moments of the thrust performance and reconstructing the probability model of the thrust performance. However, the calculation of the statistical moments usually involves a multi-dimensional integral that cannot be directly solved. When the number of variables is high, it usually brings an unbearable computational burden.

[0033] In view of this, the present application provides a thrust reliability analysis of a solid attitude control engine. Aiming at the problem of low computational efficiency in the reliability analysis of thrust performance caused by the high experimental cost and computational cost of the solid attitude control engine, a reliability analysis method for the maximum output thrust performance of a solid attitude control engine based on multi-source data fusion is proposed. This method utilizes the high-precision advantages of the samples generated by the three-dimensional interior ballistic simulation model and the low time-consuming advantages of the samples generated by the zero-dimensional interior ballistic mathematical model. By adopting a multi-source data fusion method, the advantages of the two methods are combined, effectively balancing the contradictory relationship between model performance and computational cost. Based on the constructed multi-fidelity surrogate model, the reliability analysis of the maximum output thrust performance is carried out for a specific state, providing theoretical guidance for increasing the fault tolerance of the control system design.

[0034] In order to make the technical solution of the present application clearer and easier to understand, the technical solution of the present application will be introduced below with reference to the accompanying drawings.

[0035] As shown Figure 1 in the figure, this figure is a flowchart of a method for analyzing the reliability of the thrust performance of a solid attitude control engine driven by a multi-fidelity surrogate model provided by an embodiment of the present application. The method includes: S101. Obtain the uncertainty variables of the solid attitude control engine.

[0036] The uncertainty variables include combustion chamber pressure, pintle displacement, pintle diameter, pintle head length, throat diameter, and radius of the convergent transition arc. The uncertainties of these uncertainty variables will affect the thrust of the solid attitude control engine.

[0037] As shown in Table 1, this table shows the uncertainty variables of the solid attitude control engine.

[0038] Table 1:

[0039] Among them, Table 1 shows the value ranges of the uncertainty variables.

[0040] As shown Figure 2 in the figure, this figure is a schematic diagram of the outer shape of a solid attitude control valve provided by an embodiment of the present application. Figure 2 The radius of the convergent transition arc, pintle displacement, pintle radius (half of the pintle diameter), pintle head length, and throat radius (half of the throat diameter) are schematically shown.

[0041] After determining the uncertainty variables and the value ranges of these uncertainty variables, sample generation can be performed based on these uncertainty variables, that is, determine the corresponding thrust magnitudes for subsequent reliability analysis.

[0042] S102. Generate a first number of first-class samples according to the zero-dimensional internal ballistics mathematical model of the solid attitude control engine and the uncertainty variables. The first-class samples include corresponding first-class sample thrusts and first-class sample variables.

[0043] In some embodiments, a zero-dimensional internal ballistics mathematical model can be established first. Refer to the following formula:

[0044]

[0045] Among them, is the first-class sample thrust, is the thrust coefficient, is the combustion chamber pressure, is the first-class sample pintle displacement, is the first-class sample pintle diameter, is the first-class sample pintle head length, is the throat diameter of the first type of sample, is the radius of the convergent transition arc of the first type of sample, is the equivalent throat area of the first type of sample; the thrust of the first type of sample and the variables of the first type of sample that correspond to each other form the first type of sample, and the variables of the first type of sample include the displacement of the throat plug of the first type of sample, the diameter of the throat plug of the first type of sample, the head length of the throat plug of the first type of sample, the throat diameter of the first type of sample, and the radius of the convergent transition arc of the first type of sample.

[0046] The thrust coefficient is determined by the following formula:

[0047]

[0048] where, is the thrust coefficient, is the specific heat ratio, is the propellant density, is the ambient pressure, is the cross-sectional area at the nozzle exit section, is an intermediate variable and will be replaced during the calculation process.

[0049] As can be seen from the above formula, after determining a set of uncertainty variables, the corresponding thrust can be calculated through the above formula. Therefore, the first quantity of uncertainty variables can be substituted into the above formula, and then the thrust of the first type of sample of the first quantity can be obtained. Among them, the first quantity of uncertainty variables substituted can be used as the variables of the first type of sample. Specifically, the variables of the first type of sample can include the displacement of the throat plug of the first type of sample, the diameter of the throat plug of the first type of sample, the head length of the throat plug of the first type of sample, the throat diameter of the first type of sample, and the radius of the convergent transition arc of the first type of sample. The calculation process of the above formula is relatively simple, so a large number of the first type of samples can be generated through this calculation formula.

[0050] The following introduces the derivation process of the above zero-dimensional interior ballistics mathematical model: The mass flow rate per second of the nozzle and the single-valve jet velocity are calculated through the following formula.

[0051]

[0052]

[0053] where, is the mass flow rate per second of the nozzle, is the gas constant of the gas, is the adiabatic combustion temperature of the propellant, is the single-valve jet velocity, is the propellant density, is the specific heat ratio, is the combustion chamber pressure.

[0054]

[0055] Wherein, is the combustion chamber pressure, is the propellant density, is the characteristic velocity of the propellant, is the burning rate coefficient, is the burning surface area, is the pressure exponent.

[0056] Then, according to the thrust theory, the thrust calculation formula is determined:

[0057] is the thrust, is the propellant density, is the ambient pressure, is the cross-sectional area at the nozzle exit section.

[0058] Finally, substituting each part of the formula into the above formula, the zero-dimensional internal ballistic mathematical model of the solid attitude control engine can be obtained.

[0059] S103. Generate a second quantity of second-class samples according to the uncertainty variables and the three-dimensional internal ballistic simulation model of the solid attitude control engine, where the second-class samples include corresponding second-class sample thrusts and second-class sample variables.

[0060] The first quantity is greater than the second quantity, and the accuracy of the first-class samples is less than the accuracy of the second-class samples.

[0061] In some examples, the three-dimensional interior ballistic simulation model of the solid attitude control engine can be based on the compressible Navier-Stokes (N-S) equations and the Realizable k-ε model. During the simulation using the three-dimensional model, since both the throat plug and the nozzle are of rotational symmetry, a two-dimensional axisymmetric method can be used to simplify the nozzle and the throat plug to improve the calculation efficiency. The combustion chamber pressure of the solid attitude control engine changes with the displacement of the throat plug. Therefore, the inlet boundary condition is the gas flowing into the nozzle interface, and the input condition is the pressure of the combustion chamber, such as 7 Mpa. The wall surfaces of the nozzle and the throat plug adopt the adiabatic no-slip wall boundary condition, and the nozzle profile changes with the input conditions. The outlet boundary condition is defined as the gas flowing out of the nozzle interface, and the output conditions are the atmospheric pressure, the ambient pressure, and the temperature is set to the sea-level standard atmospheric parameters. Based on the nozzle profile, nodes are selected at a certain step size to divide the computational domain grid. The profiles of the nozzle and the throat plug of the solid attitude control engine are complex, so triangular unstructured grids are used, and the areas where the physical quantities change violently, such as the boundary layer, the nozzle throat, and the throat plug head, etc., are subjected to grid encryption. Table 2 shows the thrust simulation results and the calculation time under different grid numbers.

[0062] Table 2:

[0063] It can be seen from the table that as the grid number gradually increases, the simulation results of the thrust become more and more accurate. After the grid number reaches 172,564, with the increase of the grid density, the simulation results change insignificantly, the simulation results are less affected by the grid density, and the relative error with the theoretical thrust is 0.08%, and the accuracy meets the calculation requirements. Finally, high-fidelity input data is used to calculate the maximum output thrust of the solid attitude control engine to obtain high-fidelity thrust data, that is, the second type of samples.

[0064] After obtaining the three-dimensional interior ballistic simulation model of the solid attitude control engine, a second quantity of the second type of samples can be generated based on the uncertainty variables. For example, input the second quantity of uncertainty variables into the three-dimensional interior ballistic simulation model to obtain the second quantity of the second type of sample thrust. The input uncertainty variables can be the second type of sample variables, and the second type of sample variables can include the second type of sample combustion chamber pressure, the second type of sample throat plug displacement, the second type of sample throat plug diameter, the second type of sample throat plug head length, the second type of sample throat diameter, and the second type of sample convergence transition arc radius.

[0065] It should be noted that the time required to generate a set of the second type of samples through the three-dimensional interior ballistic simulation model is relatively long, that is, it is time-consuming. However, the accuracy of the generated second type of samples is greater than that of the first type of samples. Therefore, a small number of the second type of samples can be generated, thereby reducing the time-consuming of generating samples, generating a large number of the first type of samples, and supplementing the problem of the small number of samples.

[0066] In some embodiments, low-precision samples can be extracted from the first type of samples with the first quantity through the Latin hypercube algorithm. Based on these low-precision samples, the nearest neighbor sampling method can be used to obtain high-precision sample points, and a low-precision sample set can be established. and a high-precision sample set , contains sample points, where represents the number of uncertainty variables, contains sample points, where represents the ratio of the calculation cost of high-precision data to low-precision data. The low-precision sample set is jointly composed by calculating through the zero-dimensional interior ballistic mathematical model as and , and the high-precision sample set is jointly composed by calculating through the three-dimensional interior ballistic simulation model as and . The number of samples in the low-precision sample set is greater than the number of samples in the high-precision sample set.

[0067] It should be noted that the present application does not specifically limit the execution order of S102 and S103. In some other embodiments, S103 can also be executed first and then S102. S102 and S103 can also be executed simultaneously.

[0068] S104. Construct a first surrogate model according to the first type of sample thrust and the first type of sample variables of the first quantity, and construct a second surrogate model according to the first surrogate model, the second type of sample thrust and the second type of sample variables of the second quantity.

[0069] In some examples, after obtaining the first type of sample thrust and the first type of sample variables, a first surrogate model can be constructed based on the first type of sample thrust and the first type of sample variables. This first surrogate model can be represented by , which means the predicted response of the first surrogate model when the uncertainty variable is x. Here, x can be a vector composed of combustion chamber pressure, throat plug displacement, throat plug diameter, throat plug head length, throat diameter, and convergence transition arc radius, etc.

[0070] After obtaining the first surrogate model, a second surrogate model can be constructed based on the second type of sample thrust, the second type of sample variables, and the first surrogate model. The following formula:

[0071]

[0072] where is the input of the second surrogate model as The predicted value at is a linear combination of basis functions, where represents a random process,

[0073] and is

[0074] a column vector of regression coefficients; is the predicted value of the first surrogate model at the input

[0075] Assume that the random process has a mean of zero and a covariance:

[0076] where is the Gaussian process model between the predicted value of the first surrogate model at the sample point and the predicted value of the first surrogate model at the sample point is the predicted value of the first surrogate model at the input is the process standard deviation that determines the overall size of the variance; is the correlation function between the sample points

[0077] Exemplarily:

[0078] where respectively represent and the predicted values of the first surrogate model at a certain dimension for the sample points is the dimension of the design variable,

[0079] Assume that the response of the second surrogate model can be approximated as a linear combination of the second type of samples:

[0080] where is the simulation value of the high-precision sample point, is the predicted value of the first surrogate model at x, is the predicted value of the second surrogate model at

[0081] For a set , define the design matrix :

[0082] The error between the predicted value at the test point and and can be calculated as:

[0083] where , to ensure the accuracy of the predicted value, an unbiased estimate needs to be satisfied:

[0084] where is the simulation value of the high-precision sample point.

[0085] Under this condition, the mean square error (MSE) of the predicted value is:

[0086] where is , is the mean square error of each sample point.

[0087] To minimize the mean square error, the Lagrange multiplier method is used:

[0088] where is the Lagrange multiplier, is the Lagrange multiplier function.

[0089] The gradient with respect to can be calculated:

[0090] where is the derivative with respect to c.

[0091] Starting from the first-order necessary condition of optimality, the following system of equations can be obtained:

[0092] where is defined as , then the solution is:

[0093] where is an intermediate variable.

[0094] The predicted value of the constructed multi-fidelity surrogate model for the test point x is:

[0095] S105. Analyze the thrust performance reliability of the solid attitude control engine by using the second surrogate model.

[0096] After the construction of the second surrogate model is completed, the second surrogate model can be used to analyze the thrust performance reliability of the solid attitude control engine.

[0097] Exemplarily, input the variable to be tested into the second surrogate model to obtain the predicted thrust value corresponding to the variable to be tested. Since the second surrogate model is constructed based on high-precision second-class samples and the first surrogate model, the accuracy of the data output by the second surrogate model is relatively high. Moreover, only a small number of high-precision samples are required to construct the second surrogate model. Therefore, only a small number of high-precision samples need to be generated, thereby reducing the time-consuming for constructing the second surrogate model.

[0098] Next, compare the predicted thrust value with the target thrust value to obtain a comparison result. If the comparison result indicates that the predicted thrust value is greater than or equal to the target thrust value, it is obtained that the variable to be tested meets the requirements. If the comparison result indicates that the predicted thrust value is less than the target thrust value, it is obtained that the variable to be tested does not meet the requirements.

[0099] Define the functional function for analyzing the maximum output thrust performance reliability of the solid attitude control engine as , where represents the target thrust value, represents the predicted thrust value. When , it indicates that the target thrust value is greater than the predicted thrust value, and the maximum output thrust of the attitude control engine meets the design requirements. When , it indicates that the target thrust value is less than the predicted thrust value, and the maximum output thrust of the attitude control engine fails and does not meet the design requirements. Among them, the target thrust value can be the specified maximum thrust value.

[0100] In the embodiments of the present application, Table 3 shows the model accuracy evaluation results of the maximum output thrust surrogate model of a certain type of attitude control engine.

[0101] Table 3:

[0102] Among them, R 2 is used to evaluate the global accuracy of the model. The closer its value is to 1, the higher the model accuracy. RMAE is used to evaluate the local accuracy of the model. The closer its value is to 0, the higher the local accuracy of the model. It can be seen from Table 3 that its R 2The mean value is 0.998 and the mean value of RMAE is 0.085, indicating that its global accuracy and local accuracy meet the requirements. At the same time, it can be seen from the results of 10 times that its robustness is better.

[0103] Table 4 shows a table of the case failure probability and the number of model calls.

[0104] Table 4:

[0105] Among them, under the parameter uncertainty level shown in Table 1, the probability that the maximum output thrust of the solid attitude control engine is less than 750 N is 5.30%. In the embodiment of the present application, due to the integration of data with different precisions, only 15 high-fidelity simulation models and 60 low-fidelity simulation models are called. The finally predicted failure probability is 0.0530, that is, the probability that the maximum output thrust of the solid attitude control engine is less than 750 N is 5.30%. In summary, the method provided by the embodiment of the present application greatly improves the calculation efficiency when calculating the performance reliability analysis of the maximum output thrust of the attitude control engine.

[0106] In the embodiment of the present application, when constructing the surrogate model of the performance function, different-precision data are comprehensively utilized, and the rapid and accurate prediction of the failure probability of the maximum output thrust of the solid attitude control engine can be realized at a low calculation cost. According to the characteristics of the solid attitude control engine, a zero-dimensional internal ballistic equation is established as a low-fidelity model, and a three-dimensional internal ballistic simulation model is established as a high-fidelity model. The calculation time of the zero-dimensional internal ballistic equation is about several seconds, which can be ignored compared with the one-hour calculation time of the three-dimensional internal ballistic simulation model. By fusing the two kinds of data, the calculation time can be greatly reduced on the premise of ensuring the accuracy.

[0107] Next, in combination with Figure 3 , the overall inventive concept of the present application will be introduced. As Figure 3 shown, this figure is a flowchart of another multi-fidelity surrogate model-driven solid attitude control engine thrust performance reliability analysis method provided by the embodiment of the present application.

[0108] S301. Specify the combustion chamber pressure, the plug displacement, the plug diameter, the plug head length, the throat diameter, and the radius of the convergent transition arc.

[0109] S302. Generate low-precision samples through experimental design.

[0110] S303. Calculate using the zero-dimensional internal ballistic equation.

[0111] S304. Generate high-precision samples using the nearest neighbor method.

[0112] S305. Calculate using three-dimensional internal ballistic simulation.

[0113] Among them, the execution order of S303 and S304 is not specifically limited in this application and can be executed simultaneously or successively. In some other examples, S304 and S305 can be executed first, and then S303 can be executed.

[0114] S306. Obtain multi-fidelity thrust data.

[0115] S307. Construct the basic form of the multi-fidelity model.

[0116] S308. Set the form and value of the random process.

[0117] S309. Solve the model parameters.

[0118] S310. Predict the thrust value of the unknown point.

[0119] S311. Calculate the model accuracy using cross-validation.

[0120] S312. Calculate the performance reliability of the maximum output thrust.

[0121] It should be noted that Figure 3 The details of the steps shown have been introduced in detail above, so they will not be elaborated here. Similar content can be referred to the previous embodiments.

[0122] Above, in combination with Figures 1 to 3 The thrust performance reliability analysis of the solid attitude control engine provided by the embodiments of the present application has been introduced in detail. Next, the devices and equipment provided by the embodiments of the present application will be introduced with reference to the accompanying drawings.

[0123] As Figure 4 shown, this figure is a schematic diagram of a thrust performance analysis device for a solid attitude control engine provided by an embodiment of the present application. The device includes: An acquisition module 301, configured to acquire the uncertainty variables of the solid attitude control engine, where the uncertainty variables include combustion chamber pressure, plug displacement, plug diameter, plug head length, throat diameter, and convergence transition arc radius; A generation module 302, configured to generate a first quantity of first-class samples according to the zero-dimensional internal ballistic mathematical model of the solid attitude control engine and the uncertainty variables, where the first-class samples include corresponding first-class sample thrusts and first-class sample variables; generate a second quantity of second-class samples according to the uncertainty variables and the three-dimensional internal ballistic simulation model of the solid attitude control engine, where the second-class samples include corresponding second-class sample thrusts and second-class sample variables; the first quantity is greater than the second quantity, and the accuracy of the first-class samples is less than the accuracy of the second-class samples; A construction module 303, configured to construct a first surrogate model according to the first type of sample thrust and the first type of sample variables of the first quantity, and construct a second surrogate model according to the first surrogate model, the second type of sample thrust of the second quantity, and the second type of sample variables. An analysis module 304, configured to analyze the thrust performance reliability of the solid attitude control engine by using the second surrogate model.

[0124] In some possible implementation manners, the generation module 302 is specifically configured to: Substitute the first quantity of uncertainty variables into the following formula to obtain the first type of sample thrust of the first quantity:

[0125]

[0126] Wherein, is the first type of sample thrust, is the thrust coefficient, is the combustion chamber pressure, is the first type of sample throat plug displacement, is the first type of sample throat plug diameter, is the first type of sample throat plug head length, is the first type of sample throat diameter, is the first type of sample convergent transition arc radius, is the first type of sample equivalent throat area; the corresponding first type of sample thrust and the first type of sample variables form the first type of sample, and the first type of sample variables include the first type of sample throat plug displacement, the first type of sample throat plug diameter, the first type of sample throat plug head length, the first type of sample throat diameter, and the first type of sample convergent transition arc radius.

[0127] In some possible implementation manners, the thrust coefficient is determined by the following formula:

[0128]

[0129] Wherein, is the thrust coefficient, is the specific heat ratio, is the propellant density, is the ambient pressure, is the cross-sectional area at the nozzle exit section.

[0130] In some possible implementation manners, the analysis module 304 is specifically configured to input the variables to be tested into the second surrogate model to obtain the predicted thrust value corresponding to the variables to be tested.

[0131] In some possible implementations, the analysis module 304 is further configured to compare the predicted thrust value with a target thrust value to obtain a comparison result; if the comparison result indicates that the predicted thrust value is greater than or equal to the target thrust value, a result that the variable to be tested meets the requirements is obtained; if the comparison result indicates that the predicted thrust value is less than the target thrust value, a result that the variable to be tested does not meet the requirements is obtained.

[0132] The thrust analysis device of the solid attitude control engine according to the embodiment of the present application can correspond to execute the method described in the embodiment of the present application, and the above other operations and / or functions of each module / unit of the thrust analysis device of the solid attitude control engine are respectively for implementing Figure 1 or Figure 3 the corresponding processes of the respective methods in the illustrated embodiments. For the sake of brevity, they will not be described herein again.

[0133] The embodiment of the present application further provides a computing device. As Figure 5 shown, this figure is a schematic diagram of a computing device provided by the embodiment of the present application. As Figure 5 shown, the computing device 400 includes a bus 401, a processor 402, a communication interface 403, and a memory 404. The processor 402, the memory 404, and the communication interface 403 communicate with each other through the bus 401.

[0134] The bus 401 may be a peripheral component interconnect (PCI) bus, an extended industry standard architecture (EISA) bus, or the like. The bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of representation, Figure 5 only a thick line is used to represent it in the figure, but it does not mean that there is only one bus or one type of bus.

[0135] The processor 402 may be any one or more of a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP).

[0136] The communication interface 403 is used for external communication.

[0137] The memory 404 may include volatile memory, such as random access memory (RAM). The memory 404 may also include non-volatile memory, such as read-only memory (ROM), flash memory, a hard disk drive (HDD), or a solid state drive (SSD).

[0138] Executable code is stored in the memory 404, and the processor 402 executes the executable code to perform the thrust performance reliability analysis of the aforementioned solid attitude control engine.

[0139] Specifically, in the case of implementing Figure 4 the illustrated embodiment, and Figure 4 when each module or unit of the thrust analysis device of the solid attitude control engine described in the embodiment is implemented by software, the software or program code required to execute the functions of each module / unit in the figure may be partially or fully stored in the memory 404. The processor 402 executes the program code corresponding to each unit stored in the memory 404 to perform the thrust reliability analysis of the aforementioned solid attitude control engine.

[0140] An embodiment of the present application also provides a computer-readable storage medium. The computer-readable storage medium may be any available medium that can be stored by a computing device or a data storage device such as a data center containing one or more available media. The available medium may be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid state drive), etc. The computer-readable storage medium includes instructions that direct the computing device to execute the above method.

[0141] An embodiment of the present application also provides a computer program product, which includes one or more computer instructions. When the computer instructions are loaded and executed on a computing device, the processes or functions according to the embodiments of the present application are fully or partially generated.

[0142] The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, or data center to another website, computer, or data center by wire (such as coaxial cable, fiber optic, digital subscriber line (DSL)) or wirelessly (such as infrared, wireless, microwave, etc.).

[0143] When the computer program product is executed by a computer, the computer executes any of the methods for analyzing the thrust performance reliability of the aforementioned solid attitude control engine. The computer program product can be a software installation package. In the case where any of the methods for analyzing the thrust performance reliability of the aforementioned solid attitude control engine is needed, the computer program product can be downloaded and executed on the computer.

[0144] The descriptions of the processes or structures corresponding to the above respective drawings have their own focuses. For parts not detailed in a certain process or structure, reference can be made to the relevant descriptions of other processes or structures.

[0145] As described above, the above are only specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present application should be covered within the protection scope of the present application.

Claims

1. A reliability analysis method for the single-valve thrust performance of a solid attitude control engine driven by a multi-fidelity surrogate model, characterized in that The method includes: Obtaining the uncertainty variables of the solid attitude control engine, where the uncertainty variables include combustion chamber pressure, plug displacement, plug diameter, plug head length, throat diameter, and radius of the convergent transition arc; Generating a first quantity of first-class samples according to the zero-dimensional internal ballistic mathematical model of the solid attitude control engine and the uncertainty variables, where the first-class samples include corresponding first-class sample thrusts and first-class sample variables; Generating a second quantity of second-class samples according to the uncertainty variables and the three-dimensional internal ballistic simulation model of the solid attitude control engine, where the second-class samples include corresponding second-class sample thrusts and second-class sample variables; the first quantity is greater than the second quantity, and the accuracy of the first-class samples is less than the accuracy of the second-class samples; Constructing a first surrogate model according to the first quantity of first-class sample thrusts and first-class sample variables, and constructing a second surrogate model according to the first surrogate model, the second quantity of second-class sample thrusts, and second-class sample variables; Analyzing the thrust performance reliability of the solid attitude control engine by using the second surrogate model.

2. The method according to claim 1, wherein The generating a first quantity of first-class samples according to the zero-dimensional internal ballistic mathematical model of the solid attitude control engine and the uncertainty variables includes: Substituting the first quantity of uncertainty variables into the following formula to obtain the first quantity of first-class sample thrusts: Wherein, is the first type of sample thrust, is the thrust coefficient, is the combustion chamber pressure, is the first type of sample throat plug displacement, is the first type of sample throat plug diameter, is the first type of sample throat plug head length, is the first type of sample throat diameter, is the first type of sample convergent transition arc radius, is the first type of sample equivalent throat area; the first type of sample thrust and the first type of sample variables that correspond to each other form the first type of sample, and the first type of sample variables include the first type of sample throat plug displacement, the first type of sample throat plug diameter, the first type of sample throat plug head length, the first type of sample throat diameter, and the first type of sample convergent transition arc radius.

3. The method according to claim 2, wherein The thrust coefficient is determined by the following formula: Among them, is the thrust coefficient, is the specific heat ratio, is the propellant density, is the ambient pressure, is the cross-sectional area at the nozzle exit section.

4. The method according to claim 1, wherein The analyzing the thrust performance reliability of the solid attitude control engine by using the second surrogate model includes: Inputting the variable to be tested into the second surrogate model to obtain the predicted thrust value corresponding to the variable to be tested.

5. The method according to claim 4, characterized in that The method further includes: Comparing the predicted thrust value with the target thrust value to obtain a comparison result; If the comparison result indicates that the predicted thrust value is greater than or equal to the target thrust value, obtaining the result that the variable to be tested meets the requirements; If the comparison result indicates that the predicted thrust value is less than the target thrust value, obtaining the result that the variable to be tested does not meet the requirements.

6. A device for analyzing the reliability of the single-valve thrust performance of a solid attitude control engine driven by a multi-fidelity surrogate model, characterized in that The device includes: An obtaining module, configured to obtain the uncertainty variables of the solid attitude control engine, where the uncertainty variables include combustion chamber pressure, plug displacement, plug diameter, plug head length, throat diameter, and radius of the convergent transition arc; A generating module, configured to generate a first quantity of first-class samples according to the zero-dimensional internal ballistic mathematical model of the solid attitude control engine and the uncertainty variables, where the first-class samples include corresponding first-class sample thrusts and first-class sample variables; generating a second quantity of second-class samples according to the uncertainty variables and the three-dimensional internal ballistic simulation model of the solid attitude control engine, where the second-class samples include corresponding second-class sample thrusts and second-class sample variables; the first quantity is greater than the second quantity, and the accuracy of the first-class samples is less than the accuracy of the second-class samples; A construction module, configured to construct a first surrogate model according to the first type of sample thrusts and the first type of sample variables of the first quantity, and construct a second surrogate model according to the first surrogate model, the second type of sample thrusts of the second quantity and the second type of sample variables; An analysis module, configured to analyze the thrust performance reliability of the solid attitude control engine by using the second surrogate model.

7. The device according to claim 6, characterized in that, The generation module is specifically configured to: Substitute the first quantity of uncertainty variables into the following formula to obtain the first type of sample thrusts of the first quantity: Among them, is the first type of sample thrust, is the thrust coefficient, is the combustion chamber pressure, is the first type of sample throat plug displacement, is the first type of sample throat plug diameter, is the first type of sample throat plug head length, is the first type of sample throat diameter, is the first type of sample convergence transition arc radius, is the first type of sample equivalent throat area; the first type of sample thrust and the first type of sample variables that correspond to each other form the first type of sample, and the first type of sample variables include the first type of sample throat plug displacement, the first type of sample throat plug diameter, the first type of sample throat plug head length, the first type of sample throat diameter, and the first type of sample convergence transition arc radius.

8. The device according to claim 7, wherein The thrust coefficient is determined by the following formula: Among them, is the thrust coefficient, is the specific heat ratio, is the propellant density, is the ambient pressure, is the cross-sectional area at the nozzle exit section.

9. A computing device, characterized in that, Including a memory and a processor; Wherein, one or more computer programs are stored in the memory, and the one or more computer programs include instructions; when the instructions are executed by the processor, the computing device is caused to execute the method according to any one of claims 1 to 5.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program, and the computer program is used to execute the method according to any one of claims 1 to 5.

Images