Reliability Prediction Method for UAV Flight Control System Based on Second Screening Data of Components

Abstract

The present invention belongs to the technical field of unmanned aerial vehicles, and specifically discloses a method for predicting the reliability of an unmanned aerial vehicle flight control system based on the data of secondary screening of components, including: establishing a data matrix of secondary screening of components for the unmanned aerial vehicle flight control system, and estimating the parameters of the component reliability model to obtain the reliability model parameters of each type of component in the unmanned aerial vehicle flight control system; dividing the key functional modules of the unmanned aerial vehicle flight control system and constructing a reliability block diagram of the unmanned aerial vehicle flight control system; based on the reliability model parameters of each type of component in the unmanned aerial vehicle flight control system, the key functional modules of the unmanned aerial vehicle flight control system, and the structure of the reliability block diagram of the unmanned aerial vehicle flight control system, using the Monte Carlo sampling method to predict the reliability of the unmanned aerial vehicle flight control system. The present invention solves the problems of low data utilization rate, poor model correlation, high modeling complexity, and insufficient randomness processing ability of the existing methods, and realizes the comprehensiveness, scientificity, and efficiency of the reliability prediction of the unmanned aerial vehicle flight control system.

Description

Technical Field

[0001] The present invention belongs to the technical field of unmanned aerial vehicles, and particularly relates to a method for predicting the reliability of an unmanned aerial vehicle flight control system based on second screening data of components. Background Art

[0002] In recent years, unmanned aerial vehicle technology has developed rapidly, and its application scenarios have gradually expanded from traditional military uses to multiple fields such as logistics transportation, agricultural monitoring, environmental protection, and disaster relief. However, as the core part of an unmanned aerial vehicle, the reliability of the flight control system directly determines the safety, stability, and service life of the unmanned aerial vehicle. Electronic components in the flight control system are the key factors affecting reliability. Due to the complex and changeable working environment of unmanned aerial vehicles (such as high temperature, low pressure, vibration, etc.), the failure risk of components increases significantly. Therefore, it is particularly important to predict reliability during the design stage of the flight control system.

[0003] At present, component reliability screening (including first screening and second screening) is an important means to ensure component performance, and the screening results usually include failure data of components under different acceleration conditions. However, these data are only used for preliminary screening of components and are not further utilized in system design. In addition, traditional system reliability analysis methods often rely on empirical values or reliability parameters of single components, and it is difficult to comprehensively reflect the actual reliability of complex flight control systems. This situation makes it an important requirement for technological development to accurately predict the reliability of the flight control system at the initial design stage.

[0004] Currently, the mainstream methods for system reliability analysis include the following several types:

[0005] 1. Method based on fault tree analysis: By analyzing the possible fault modes of the system and their impacts layer by layer, the system reliability is deduced. However, the analysis efficiency of this method for complex systems is low, and it is difficult to use dynamic component data for reliability modeling.

[0006] 2. Method based on reliability block diagram: By constructing the reliability block diagram of the system, the overall reliability of the system is calculated. However, this method usually assumes that the component reliability is known and ignores the randomness and complex failure modes of components in the actual environment.

[0007] 3. Data model based on component accelerated life test: Existing research mainly focuses on the analysis of the accelerated life model of single components (such as the Arrhenius model), but fails to combine the accelerated test data with system-level reliability modeling.

[0008] There are many deficiencies in the reliability analysis of the UAV flight control system in the existing technology, mainly reflected in the failure to fully utilize the secondary screening data of components for system-level reliability modeling. These screening data are usually only used to judge whether the components are qualified, and lack a standardized processing method to convert them into reliability models, resulting in valuable failure data not being used for system-level reliability analysis.

[0009] In addition, the correlation between system-level reliability analysis and component failure data is insufficient. Existing methods usually assume that the reliability parameters of components are fixed values, ignoring the dynamic changes of the accelerated test data and environmental impacts of components. The topology of the UAV flight control system is complex, including various series, parallel, and hybrid structures. Existing modeling methods are inefficient in dealing with these complex logics. Especially in the reliability calculation of redundant structures and parallel modules, it is easy to lead to inaccurate analysis or incomplete modeling problems.

[0010] At the same time, the existing technology has limited ability to handle the randomness of component failure times. It generally uses deterministic analysis methods and is difficult to truly reflect the random characteristics of component failures. Especially the lack of random modeling tools such as Monte Carlo simulation makes it impossible to comprehensively sample and statistically analyze the failure times of a large number of components in the flight control system.

[0011] Finally, existing methods usually can only conduct reliability assessment based on actual test data after the system design is completed, and it is difficult to provide reliable prediction results in the initial stage of design. The lack of a reliability modeling method based on component screening data and design logic in the initial stage of design makes the reliability prediction unable to meet the high standards of safety and stability of the UAV flight control system. Summary of the Invention

[0012] The purpose of the present invention is to solve the problems of low data utilization rate, poor model correlation, high modeling complexity, and insufficient randomness processing ability of existing methods, and propose a reliability prediction method for UAV flight control systems based on secondary screening data of components.

[0013] The technical solution of the present invention is as follows: A reliability prediction method for UAV flight control systems based on secondary screening data of components, including the following steps:

[0014] S1. Establish a secondary screening data matrix of components for the UAV flight control system;

[0015] S2. According to the secondary screening data matrix of components, estimate the parameters of the component reliability model to obtain the reliability model parameters of each type of component in the UAV flight control system;

[0016] S3. Divide the key functional modules of the UAV flight control system and construct a reliability block diagram of the UAV flight control system;

[0017] S4. Based on the reliability model parameters of various types of components in the UAV flight control system, the key functional modules of the UAV flight control system, and the reliability block diagram structure of the UAV flight control system, the Monte Carlo sampling method is used to estimate the reliability of the UAV flight control system.

[0018] The beneficial effects of the present invention are as follows:

[0019] 1. Based on the component second screening data, by establishing a component second screening data matrix and combining the Monte Carlo sampling method, a reliability model from the component level to the system level is constructed and the reliability is estimated, thus making up for the defects of low data utilization rate, poor model correlation, high modeling complexity, and insufficient randomness processing ability in the prior art, and providing a scientific basis for the safety and stability of the UAV system.

[0020] 2. By using the component second screening data and the reliability block diagram model, the present invention can estimate the reliability of the UAV flight control system in the initial design stage, filling the gap in the reliability analysis ability in the design stage of the prior art.

[0021] Preferably, the component second screening data matrix in step S1 is expressed by the formula:

[0022]

[0023] where represents the i th type of component second screening data matrix, , represents the number of component models in the UAV flight control system;

[0024] is expressed by the formula:

[0025]

[0026] where represents the i th number of components in the j th second screening test of the th type of component, i represents the j th number of failed components in the th second screening test of the i th type of component, j represents the th total duration of the i th second screening test of the j th type of component, , represents the iThe quantity of second screening data for components of each model.

[0027] Preferably, step S2 specifically includes the following sub-steps:

[0028] S21. Calculate the equivalent test duration of the components of the UAV flight control system under the target operating temperature conditions according to the temperature accelerated life model and the second screening data matrix of the components.

[0029] S22. Convert the second screening data matrix of the components into a left censored data matrix and a right censored data matrix to obtain the censored data of all models of components in the UAV flight control system.

[0030] S23. Set the reliability model for each model of components in the UAV flight control system.

[0031] S24. Based on the censored data of all models of components, use the maximum likelihood estimation method to estimate the parameter estimation values of the reliability model for each model of components in the UAV flight control system.

[0032] Preferably, the calculation formula for the equivalent test duration of the components of the UAV flight control system under the target operating temperature conditions in step S21 is:

[0033]

[0034] Where, represents the equivalent test duration of the components of the UAV flight control system under the target operating temperature conditions, represents the activation energy, represents the Boltzmann constant, k= 8.617×10 −5  eV / K, represents the target operating temperature, represents the natural base.

[0035] Preferably, step S22 specifically includes the following steps:

[0036] For the i th second screening data of the j th model of components, add to the left censored data matrix and add to the right censored data matrix ; Process the second screening data from the 1st to the th in sequence to obtain the left censored data matrix and the right censored data matrix of the i th model of components, and its expression formula is: ​​

[0037]

[0038] Let , , for the left-censored data matrix of the i th model component and the right-censored data matrix perform simplification to obtain:

[0039]

[0040] wherein, corresponds to the equivalent test duration of the left-censored data in corresponds to the equivalent test duration of the right-censored data in

[0041] Repeat the above steps until the conversion of the censored data of all model components is completed to obtain the censored data of all model components in the UAV flight control system.

[0042] Preferably, according to the failure characteristics of each model component, setting the reliability model of each model component in the UAV flight control system specifically is:

[0043] If the failure characteristic of the component is random failure, the reliability model of each model component in the UAV flight control system is an exponential distribution, and its expression formula is:

[0044]

[0045] wherein, represents the cumulative failure distribution function of the exponential distribution, represents the reliability function of the exponential distribution, represents the natural base, represents time, represents the exponential distribution parameter, and there is ;

[0046] If the failure characteristic of the component is wear-out failure, the reliability model of each model component in the UAV flight control system is a lognormal distribution, and its expression formula is:

[0047]

[0048] wherein, represents the cumulative failure distribution function of the lognormal distribution, represents the reliability function of the lognormal distribution, represents the log mean, and there is , represents the log standard deviation, and there is , represents the standard normal distribution, represents the logarithmic function with the natural base as the base;

[0049] If the failure characteristics of the components are early failure, random failure or wear-out failure, the reliability model of each type of component in the UAV flight control system is the Weibull distribution, and its expression formula is:

[0050]

[0051] where, represents the cumulative failure distribution function of the Weibull distribution, represents the reliability function of the Weibull distribution, represents the shape parameter of the Weibull distribution, and there is , represents the scale parameter of the Weibull distribution, and there is .

[0052] Preferably, the reliability model parameters of each type of component in the UAV flight control system described in step S24 include the exponential distribution parameter , the logarithmic mean , the logarithmic standard deviation , the shape parameter of the Weibull distribution, and the scale parameter of the Weibull distribution;

[0053] The specific steps of step S24 include the following sub-steps:

[0054] S241. Establish the maximum likelihood function of the reliability model parameters of each type of component in the UAV flight control system ;

[0055] S242. Convert the maximum likelihood function into the logarithmic maximum likelihood function;

[0056] S243. Use the optimization method to obtain the estimated values of the reliability model parameters of each type of component with the maximum logarithmic maximum likelihood function as the goal.

[0057] Preferably, the expression formula of the maximum likelihood function is:

[0058]

[0059] where, represents the distribution parameter vector, represents the cumulative failure distribution function of the reliability model of the i th type of component, Represents the reliability function of the reliability model of the i th model of component, Represents the i th model of component's p th left-censored equivalent test time, Represents the i th model of component's q th right-censored equivalent test time;

[0060] The logarithmic maximum likelihood function The calculation formula is:

[0061]

[0062] Wherein, Represents the logarithmic function with the natural base as the base.

[0063] Preferably, the step S4 specifically includes the following sub-steps:

[0064] S41. According to the component list and bit number information of the key functional modules in the UAV flight control system, and the reliability model parameters of each model of components in the UAV flight control system, use the Monte Carlo sampling method for sampling to obtain the failure time matrix of all components of the key functional modules in the UAV flight control system, and its expression formula is:

[0065]

[0066] Wherein, Represents the failure time sampling result of the i 1 th component for the j 1 th Monte Carlo sampling, , , Represents the total number of components, Represents the number of Monte Carlo samplings;

[0067] S42. According to the failure time matrix of all components of the key functional modules in the UAV flight control system, select the minimum value of the failure times of all components of each key functional module as the failure time of the key functional module to obtain the failure time matrix of all key functional modules in the UAV flight control system, and its expression formula is:

[0068]

[0069] Wherein, Represents the The failure time results of the th Monte Carlo sampling for a key functional module, , indicating the total number of key functional modules;

[0070] S43. According to the reliability block diagram of the UAV flight control system, for the parallel structure in the reliability block diagram, select the maximum failure time among all modules or sub-structures in the parallel structure as the failure time of the parallel structure; for the series structure in the reliability block diagram, select the minimum failure time among all modules or sub-structures in the series structure as the failure time of the series structure, and calculate the failure time vector of the UAV flight control system , and its expression formula is:

[0071]

[0072] where, represents the failure time result of the th Monte Carlo sampling of the UAV flight control system;

[0073] S44. Sort the failure time vector of the UAV flight control system in ascending order to obtain the failure time order statistic of the UAV flight control system, and its expression formula is:

[0074]

[0075] where, represents the number of failure times less than the expected service time, , represents the number of failure times at the expected service time corresponding to the failure time of the UAV flight control system;

[0076] S45. According to the failure time order statistic of the UAV flight control system, find the number of failure times less than the expected service time, and then calculate the reliability of the UAV flight control system and the mean time to failure of the UAV flight control system to complete the reliability prediction of the UAV flight control system.

[0077] Preferably, the calculation formula for the reliability of the UAV flight control system described in step S45 is:

[0078]

[0079] where, represents the reliability of the UAV flight control system;

[0080] The calculation formula for the mean time to failure of the UAV flight control system is:

[0081]

[0082] Among them, represents the average failure time of the UAV flight control system.

[0083] The beneficial effects of the above preferred solution are as follows:

[0084] 1. The above preferred solution constructs a component secondary screening data matrix based on the component secondary screening data, and converts the screening data into an equivalent test duration through a temperature accelerated life model, solving the problem of low utilization rate of screening data in the prior art;

[0085] 2. By converting the censored data of components into a reliability distribution model and combining the logical relationship of the functional modules of the flight control system, a unified reliability block diagram model is established, solving the technical defect that the system-level analysis in the prior art fails to effectively combine component failure data;

[0086] 3. The reliability block diagram modeling method is applicable to complex systems, can clearly express the logical relationship of series, parallel and hybrid structures in the UAV flight control system, and improves the accuracy and operability of modeling. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] Figure 1 Shown is a flowchart of a method for predicting the reliability of a UAV flight control system based on component secondary screening data.

[0088] Figure 2 Shown is a schematic diagram of the reliability block diagram structure of the UAV flight control system. DETAILED DESCRIPTION

[0089] Now, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. It should be understood that the embodiments shown and described in the drawings are merely exemplary, intended to explain the principles and spirit of the present invention, and not to limit the scope of the present invention.

[0090] Example:

[0091] As Figure 1 shown, a method for predicting the reliability of a UAV flight control system based on component secondary screening data includes the following steps:

[0092] S1. Collect the component list of the UAV flight control system, including the quantity of all components such as the flight control main board, attitude sensor, servo drive unit, power control unit, etc., and establish a component secondary screening data matrix of the UAV flight control system;

[0093] S2. According to the component secondary screening data matrix, estimate the parameters of the component reliability model to obtain the parameters of the reliability model of each type of component in the UAV flight control system;

[0094] S3. Divide the key functional modules of the UAV flight control system and construct the reliability block diagram of the UAV flight control system;

[0095] S4. Based on the reliability model parameters of various types of components in the UAV flight control system, the key functional modules of the UAV flight control system, and the structure of the reliability block diagram of the UAV flight control system, use the Monte Carlo sampling method to predict the reliability of the UAV flight control system.

[0096] In this embodiment, the component second screening data matrix in step S1 is expressed by the formula:

[0097]

[0098] where represents the second screening data matrix of the i th type of component, , represents the number of component types in the UAV flight control system;

[0099] is expressed by the formula:

[0100]

[0101] where represents the number of components in the i th second screening test of the j th type of component, represents the number of failed components in the i th second screening test of the j th type of component, represents the total duration of the i th second screening test of the j th type of component, represents the temperature condition of the i th second screening test of the j th type of component, , represents the number of second screening data of the i th type of component; the component second screening data matrix of the UAV flight control system is shown in Table 1.

[0102] Table 1 Component Second Screening Data Matrix of UAV Flight Control System

[0103]

[0104]

[0105]

[0106] In this embodiment, step S2 specifically includes the following sub-steps:

[0107] S21. Calculate the equivalent test duration of the components of the UAV flight control system under the target use temperature conditions according to the temperature acceleration life model and the component secondary screening data matrix;

[0108] S22. Convert the component secondary screening data matrix into a left-censored data matrix and a right-censored data matrix to obtain the censored data of all component models in the UAV flight control system;

[0109] S23. Set the reliability models of the components of each model in the UAV flight control system;

[0110] S24. Based on the censored data of all component models, use the maximum likelihood estimation method to estimate the parameter estimation values of the reliability models of the components of each model in the UAV flight control system.

[0111] In this embodiment, the Arrhenius temperature acceleration life model is adopted, and its expression formula is:

[0112]

[0113] Wherein, represents the degradation reaction rate, represents the life proportionality factor constant under calibration conditions, represents the activation energy, with the unit of eV, represents the Boltzmann constant, k= 8.617×10 −5  eV / K, represents the temperature, with the unit of K, represents the natural logarithm base;

[0114] Therefore, the calculation formula for the equivalent test duration of the components of the UAV flight control system under the target use temperature conditions in step S21 is:

[0115]

[0116] Wherein, represents the equivalent test duration of the components of the UAV flight control system under the target use temperature conditions, represents the Boltzmann constant, k= 8.617×10 −5  eV / K, represents the target use temperature; the equivalent test duration of the components of the UAV flight control system under the target use temperature conditions is shown in Table 2.

[0117] Table 2 Equivalent test duration under target use temperature conditions

[0118]

[0119]

[0120]

[0121] In this embodiment, step S22 specifically includes the following steps:

[0122] For the i th second screening data of the j th model component, add in , add in ; Process the second screening data from the first to the th second screening data in sequence, and the left censored data matrix i and the right censored data matrix of the th model component can be obtained. The expression formula is:

[0123]

[0124] Let , , simplify the left censored data matrix i and the right censored data matrix of the th model component, and get:

[0125]

[0126] Among them, corresponds to the equivalent test duration of the left censored data in ; corresponds to the equivalent test duration of the right censored data in ;

[0127] Repeat the above steps until the conversion of the censored data of all model components is completed, and the censored data of all model components in the UAV flight control system are obtained, as shown in Table 3.

[0128] Table 3 Censored data of all model components in the UAV flight control system

[0129]

[0130] In this embodiment, in step S23, setting the reliability model of each model component in the UAV flight control system according to the failure characteristics of each model component is specifically:

[0131] If the failure characteristics of components are random failures, the failure rate is independent of time, and the reliability models of various types of components in the UAV flight control system are exponential distributions, and their expression formulas are:

[0132]

[0133] Among them, represents the cumulative failure distribution function of the exponential distribution, represents the reliability function of the exponential distribution, represents the natural base, represents time, represents the exponential distribution parameter, and there is ;

[0134] If the failure characteristics of components are wear-out failures, the reliability models of various types of components in the UAV flight control system are lognormal distributions, and their expression formulas are:

[0135]

[0136] Among them, represents the cumulative failure distribution function of the lognormal distribution, represents the reliability function of the lognormal distribution, represents the log mean, and there is , represents the log standard deviation, and there is , represents the standard normal distribution, represents the logarithmic function with the natural base ;

[0137] If the failure characteristics of components are early failures, random failures or wear-out failures, the reliability models of various types of components in the UAV flight control system are Weibull distributions, and their expression formulas are:

[0138]

[0139] Among them, represents the cumulative failure distribution function of the Weibull distribution, represents the reliability function of the Weibull distribution, represents the shape parameter of the Weibull distribution, and there is , represents the scale parameter of the Weibull distribution, and there is .

[0140] In this embodiment, the reliability model parameters of various types of components in the UAV flight control system described in step S24 include the exponential distribution parameter , the log mean , the log standard deviation , the shape parameter of the Weibull distribution and the scale parameter of the Weibull distribution ;

[0141] The step S24 specifically includes the following sub-steps:

[0142] S241. Establish the maximum likelihood function of the reliability model parameters of each type of component in the UAV flight control system , and its expression formula is:

[0143]

[0144] Among them, represents the distribution parameter vector (such as the exponential distribution parameter , the logarithmic mean , the logarithmic standard deviation , the shape parameter of the Weibull distribution and the scale parameter of the Weibull distribution ), represents the cumulative failure distribution function of the reliability model of the i th type of component, represents the reliability function of the reliability model of the i th type of component, represents the i th left-censored equivalent test time of the p th type of component, represents the i th right-censored equivalent test time of the q th type of component;

[0145] S242. Convert the maximum likelihood function to the log-likelihood function , and its conversion formula is:

[0146]

[0147] Among them, represents the logarithmic function with the natural base ;

[0148] S243. Using optimization methods, such as the least squares method, etc., with the maximum log-likelihood function as the objective, obtain the estimated values of the reliability model parameters of each type of component; the estimated values of the reliability model parameters of each type of component in the UAV flight control system are shown in Table 4.

[0149] Table 4 Estimated values of the reliability model parameters of each type of component in the UAV flight control system

[0150]

[0151] In this embodiment, in step S3, key functional modules are divided according to the functions and roles of each module of the UAV flight control system, such as a control module (flight control main board and sensor interface unit), an execution module (servo drive unit and power control unit), a perception module (attitude sensor unit and GPS unit), a communication module (wireless transmission unit and signal relay unit), and a power supply module (battery management unit and power supply unit), etc. The attitude sensor unit is a triple-module redundant attitude sensor;

[0152] According to the working logic of the UAV flight control system and the functional coupling relationship of each module, a reliability block diagram of the UAV flight control system is determined. The structure of the reliability block diagram of the UAV flight control system is as Figure 2 shown, where the flight control main board, the triple-module redundant attitude sensor, the servo drive unit, and the power control unit are in series, and the triple-module redundant attitude sensor is in a parallel structure including three attitude sensors, thus constituting the reliability block diagram structure of the UAV flight control system with a hybrid structure.

[0153] In this embodiment, step S4 specifically includes the following sub-steps:

[0154] S41. According to the component list and location number information of the key functional modules in the UAV flight control system, as well as the reliability model parameters of each type of component in the UAV flight control system, the Monte Carlo sampling method is used for sampling to obtain the failure time matrix of all components of the key functional modules in the UAV flight control system , and its expression formula is:

[0155]

[0156] where represents the total number of components, and there is , represents the number of Monte Carlo samplings, and there is represents the failure time sampling result of the i 1 rd component for the j 1 th Monte Carlo sampling, , ; The random sampling results of the failure times of all components of the key functional modules in the UAV flight control system can be obtained, as shown in Table 5.

[0157] Table 5 Random sampling results of component failure times in the UAV flight control system (hours)

[0158]

[0159]

[0160]

[0161] S42. According to the failure time matrix of all components of the key functional modules in the UAV flight control system , select the minimum failure time of all components of each key functional module as the failure time of the key functional module, and obtain the failure time matrix of all key functional modules in the UAV flight control system , and its expression formula is:

[0162]

[0163] where represents the failure time result of the th Monte Carlo sampling of the th key functional module, , represents the total number of key functional modules; the failure times of all key functional modules in the UAV flight control system are shown in Table 6;

[0164] Table 6 Failure times of all key functional modules in the UAV flight control system

[0165]

[0166] S43. According to the reliability block diagram of the UAV flight control system, for the parallel structure in the reliability block diagram, select the maximum failure time of all modules or sub-structures in the parallel structure as the failure time of the parallel structure; for the series structure in the reliability block diagram, select the minimum failure time of all modules or sub-structures in the series structure as the failure time of the series structure, and calculate the failure time vector of the UAV flight control system , and its expression formula is:

[0167]

[0168] where represents the failure time result of the th Monte Carlo sampling of the UAV flight control system;

[0169] S44. Sort the failure time vector of the UAV flight control system in ascending order to obtain the failure time order statistic of the UAV flight control system, and its expression formula is:

[0170]

[0171] where represents the number of failure times less than the expected service time, , The number of failure times indicating the expected usage time The corresponding failure time of the UAV flight control system;

[0172] For the parallel structure in the reliability block diagram, in this example, the maximum value of the failure times of the first attitude sensor, the second attitude sensor, and the third attitude sensor in the parallel structure is selected as the failure time of the triple modular redundant attitude sensor parallel structure. The failure times of the triple modular redundant attitude sensor are shown in Table 7.

[0173] Table 7 Failure times of the triple modular redundant attitude sensor (hours)

[0174]

[0175] For the series structure in the reliability block diagram, the minimum value of the failure times of the flight control main board, the triple modular redundant attitude sensor, the servo drive unit, and the power control unit in the series structure is selected as the failure time of this series structure, that is, the failure time of the UAV flight control system, as shown in Table 8. The order statistics of the failure times of the UAV flight control system are shown in Table 9.

[0176] Table 8 Failure times of the UAV flight control system (hours)

[0177]

[0178] Table 9 Order statistics of the failure times of the UAV flight control system (hours)

[0179]

[0180] S45. According to the order statistics of the failure times of the UAV flight control system, for the reliability prediction of the expected usage time hours, find the number of failure times less than the expected usage time , and then calculate the reliability of the UAV flight control system and the mean time to failure of the UAV flight control system to complete the reliability prediction of the UAV flight control system; the reliability of the UAV flight control system is:

[0181] ;

[0182] The mean time to failure of the UAV flight control system is:

[0183] .

[0184] In the initial stage of the design of the UAV flight control system, based on the data of the second screening of components, a system-level reliability model is constructed and reliability prediction is carried out, so as to make up for the defects of low data utilization rate, poor model correlation, high modeling complexity and insufficient randomness processing ability in the prior art, and provide a scientific basis for the safety and stability of the UAV system.

[0185] Those of ordinary skill in the art will recognize that the embodiments described herein are for the purpose of assisting the reader in understanding the principles of the present invention, and it should be understood that the scope of protection of the present invention is not limited to such specific statements and embodiments. Those of ordinary skill in the art can make various other specific deformations and combinations that do not depart from the essence of the present invention based on the technical revelations disclosed in the present invention, and these deformations and combinations are still within the scope of protection of the present invention.

Claims

1. A method for predicting the reliability of a UAV flight control system based on secondary screening data of components, characterized in that: The following steps are involved: S1. Establish the second screening data matrix of components of the UAV flight control system; S2. Estimating the reliability model parameters of components according to the second screening data matrix of components, and obtaining the reliability model parameters of each type of components in the UAV flight control system; S3. Divide the key functional modules of the UAV flight control system and construct a reliability block diagram of the UAV flight control system; S4. Based on the reliability model parameters of various types of components in the UAV flight control system, the key functional modules of the UAV flight control system and the reliability block diagram structure of the UAV flight control system, the reliability of the UAV flight control system is predicted using the Monte Carlo sampling method; The second screening data matrix of components in step S1 The expression formula is: in, Indicates i The second screening data matrix of the model components, , Indicates the number of component models of the drone flight control system; The expression formula is: in, Indicates i The first part of the component j Number of test components for the second screen, Indicates i The first part of the component j Number of failed components in the second-screen test, Indicates i The first part of the component j Total duration of the second screening test, Indicates i The first part of the component j Temperature conditions for the second sieve test: , Indicates i The number of second screening data of each model of components; The step S2 specifically includes the following sub-steps: S21. Calculate the equivalent test duration of the components of the UAV flight control system under the target operating temperature conditions based on the temperature accelerated life model and the second screening data matrix of the components; S22. Convert the component second screening data matrix into a left-censored data matrix and a right-censored data matrix to obtain the censored data of all types of components in the UAV flight control system; S23. Set reliability models for various types of components in the UAV flight control system; S24. Based on the missing data of all types of components, the maximum likelihood estimation method is used to estimate the estimated values ​​of the reliability model parameters of each type of component in the UAV flight control system.

2. The reliability prediction method of the UAV flight control system based on the second screening data of components according to claim 1 is characterized in that: The calculation formula for the equivalent test time of the components of the UAV flight control system under the target operating temperature conditions in step S21 is: in, Indicates the equivalent test time of the components of the UAV flight control system under the target operating temperature conditions. represents the activation energy, represents the Boltzmann constant, k= 8.617×10 −5  eV / K, Indicates the target operating temperature, Represents the natural base.

3. The reliability prediction method of the UAV flight control system based on the second screening data of components according to claim 2 is characterized in that: The step S22 specifically includes the following steps: For i The first part of the component j Two-screen data, in the left-censored data matrix Increase indivual , in the right-censored data matrix Increase indivual ; Sequentially filter the data from the first to the second The second screening data is processed to obtain i Left-censored data matrix of components of different models and the right-censored data matrix , and its expression formula is: make , , for i Left-censored data matrix of components of different models and the right-censored data matrix Simplifying, we get: in, correspond Equivalent test duration for left-center censored data; correspond Equivalent test duration for middle and right censored data; Repeat the above steps until the component missing data conversion is completed for all types of components, and the component missing data of all types in the UAV flight control system is obtained.

4. The reliability prediction method of UAV flight control system based on component secondary screening data according to claim 1 is characterized in that: In step S23, according to the missing data of all types of components, the reliability model of each type of component in the UAV flight control system is set, specifically: If the failure characteristics of components are random failures, the reliability model of each type of components in the UAV flight control system is an exponential distribution, and its expression formula is: in, represents the exponentially distributed cumulative failure distribution function, represents the reliability function of the exponential distribution, represents the natural base, Indicates time, represents the exponential distribution parameter, ; If the failure characteristic of the component is wear-out failure, the reliability model of each type of component in the UAV flight control system is a log-normal distribution, and its expression formula is: in, represents the cumulative failure distribution function of the lognormal distribution, represents the reliability function of the lognormal distribution, represents the logarithmic mean, , represents the logarithmic standard deviation, , represents the standard normal distribution, Represents natural base Logarithmic function with base ; If the failure characteristics of components are early failure, random failure or wear-out failure, the reliability model of each type of component in the UAV flight control system is Weibull distribution, and its expression formula is: in, represents the cumulative failure distribution function of the Weibull distribution, represents the reliability function of the Weibull distribution, represents the shape parameter of the Weibull distribution, , represents the scale parameter of the Weibull distribution, .

5. The method for predicting the reliability of a UAV flight control system based on component secondary screening data according to claim 4 is characterized in that: The reliability model parameters of each type of components in the UAV flight control system in step S24 include exponential distribution parameters , log mean , logarithmic standard deviation , shape parameter of the Weibull distribution and the scale parameter of the Weibull distribution ; The step S24 specifically includes the following sub-steps: S241. Establish the maximum likelihood function of the reliability model parameters of various types of components in the UAV flight control system ; S242. The maximum likelihood function Convert to log maximum likelihood function; S243. Using the optimization method, the maximum logarithmic maximum likelihood function The goal is to obtain the estimated values ​​of the reliability model parameters of various types of components.

6. The reliability prediction method of UAV flight control system based on component secondary screening data according to claim 5 is characterized in that: The maximum likelihood function The expression formula is: in, represents the distribution parameter vector, Indicates i The cumulative failure distribution function of the reliability model of the components of each model, Indicates i The reliability function of the reliability model of the component model is Indicates i Model Component No. p left-censored equivalent test time, Indicates i Model Component No. q right-censored equivalent test time; The logarithmic maximum likelihood function The calculation formula is: in, Represents natural base The logarithmic function with base .

7. The reliability prediction method of UAV flight control system based on component secondary screening data according to claim 1 is characterized in that: The step S4 specifically includes the following sub-steps: S41. Based on the component list and position number information of the key functional modules in the UAV flight control system, as well as the reliability model parameters of various types of components in the UAV flight control system, the Monte Carlo sampling method is used to obtain the failure time matrix of all components in the key functional modules in the UAV flight control system. , and its expression formula is: in, Indicates i 1 component j The failure time sampling results of 1 Monte Carlo sampling, , , Indicates the total number of components. represents the number of Monte Carlo sampling; S42. Based on the failure time matrix of all components of key functional modules in the UAV flight control system , select the minimum failure time of all components of each key functional module as the failure time of the key functional module, and obtain the failure time matrix of all key functional modules in the UAV flight control system , and its expression formula is: in, Indicates Key functional modules The failure time results of the Monte Carlo sampling, , Indicates the total number of key functional modules; S43. According to the reliability block diagram of the UAV flight control system, for the parallel structure in the reliability block diagram, the maximum failure time of all modules or substructures in the parallel structure is selected as the failure time of the parallel structure; for the series structure in the reliability structure block diagram, the minimum failure time of all modules or substructures in the series structure is selected as the failure time of the series structure, and the failure time vector of the UAV flight control system is calculated. , and its expression formula is: in, Indicates the UAV flight control system Failure time results for Monte Carlo sampling; S44. The failure time vector of the UAV flight control system Sorting in ascending order, we get the failure time sequence statistics of the UAV flight control system, which is expressed as: in, represents the number of failure times that are less than the expected service time, , The number of expiration times that represent the expected usage time The corresponding failure time of the UAV flight control system; S45. Based on the failure time sequence statistics of the UAV flight control system, find the number of failure times that are less than the expected use time , and then calculate the reliability of the UAV flight control system and the average failure time of the UAV flight control system to complete the reliability prediction of the UAV flight control system.

8. The method for predicting the reliability of a UAV flight control system based on component secondary screening data according to claim 7 is characterized in that: The calculation formula of the reliability of the UAV flight control system in step S45 is: in, Indicates the reliability of the UAV flight control system; The calculation formula for the average failure time of the UAV flight control system is: in, Indicates the average failure time of the UAV flight control system.

Images