Traveling wave tube service life evaluation method and device fused with multi-source degradation data and medium

Abstract

The invention relates to the technical field of reliability evaluation and life prediction, in particular to a traveling wave tube life evaluation method and device fused with multi-source degradation data and a medium. In consideration of a large number of accelerated tests of a cathode carried out on the ground, test data accumulated in an accelerated degradation test process of applying current sensitive stress are sorted and analyzed, and a hierarchical Bayesian model is utilized, and a hierarchical Bayesian fusion framework is adopted to obtain a hierarchical Bayesian model; the performance degradation data of the cathode under different acceleration current conditions are fused, namely the performance degradation data under different test conditions are fused, the service life data and the sample capacity are expanded, the influence of the cathode emission current on the service life of the cathode at a certain working temperature is explored, and the service life of the cathode of the traveling wave tube is evaluated.

Description

Technical Field

[0001] The present invention relates to the technical field of reliability assessment and life prediction, and in particular to a traveling wave tube life assessment method, device and medium integrating multi-source degradation data. Background Art

[0002] Due to its wide bandwidth, high power, high gain and high efficiency, the traveling wave tube is widely used as the final power amplifier in navigation, communication data transmission, electronic countermeasures and other fields. The space traveling wave tube is one of the most core payload units of the satellite system. As a typical microwave high-power product, its life is generally determined by the cathode. In the past, temperature was usually used as an accelerated stress to study the influence of temperature on the cathode life. However, the life of the cathode is closely related to its operating temperature and emission current. When the emission current exceeds a certain range, the life of the cathode decreases seriously and does not meet the engineering requirements. In order to study the influence of the cathode emission current density on the cathode life, considering that the accelerated degradation test with current as the sensitive stress has been carried out on the ground, and a large amount of performance degradation data has been accumulated, if the degradation data under different accelerated stress conditions can be integrated, the accuracy of life assessment can be greatly improved. According to the working mechanism of the hot cathode, the main factors affecting the cathode life are the transmission rate of barium from the generation point to the cathode surface, the evaporation rate of barium and the cathode emission current density. Since the generation and evaporation rate of barium are related to temperature, considering the long cathode life, temperature is often selected as the sensitive stress for accelerated life test in engineering. As engineering tasks put forward higher requirements for the output performance of traveling wave tube amplifiers, such as higher power and higher gain, the cathode emission current density is closely related to the cathode life. Therefore, under certain working temperature conditions, it is very important to study the influence of cathode emission current density on cathode life. Although many studies on cathode life have been carried out in engineering, most of them are accelerated by temperature. The logarithm of the cathode life is a linear function of the inverse of temperature obtained by taking the logarithm of both sides of the Arrhenius equation. The cathode life is obtained by obtaining accelerated life test data based on more than two groups of different temperature acceleration factors, and then extrapolated to obtain the theoretical estimated life of the cathode under other working temperature conditions. However, this method only considers the influence of temperature on life, and does not consider the influence of emission current on cathode life.

[0003] It can be seen that the method of using the Arrhenius model to fit and extrapolate the life under normal working stress in the prior art examines the influence of temperature on the cathode life, and the cathode life is closely related to its operating temperature and cathode emission current density. Only studying the influence of temperature on the life is difficult to meet the actual engineering needs. Summary of the invention

[0004] In view of this, the present invention provides a traveling wave tube life assessment method, device and medium that integrate multi-source degradation data, which can integrate performance degradation data under different current acceleration test conditions and improve the accuracy of traveling wave tube life assessment.

[0005] To achieve the above object, the present invention provides a traveling wave tube life assessment method integrating multi-source degradation data, comprising:

[0006] A Wiener process model integrating random effects and current acceleration stress is constructed, in which cathode emission current is used as the sensitive stress for acceleration.

[0007] Using hierarchical Bayesian theory, a Bayesian fusion model framework is constructed to fuse multi-source performance degradation data; the performance degradation data is the emission current of the cathode of the traveling wave tube under different current acceleration stresses;

[0008] Based on the analysis of the Wiener process model integrating random effects and current acceleration stress and the Bayesian fusion model framework, a likelihood function integrating the performance degradation data under different acceleration stress conditions is constructed;

[0009] Obtain the initial prior distribution of the model hyperparameters, and use Bayesian theory to integrate the current prior distribution with the current likelihood function to obtain the current joint posterior distribution of the model;

[0010] The unknown hyperparameters of the current level are obtained, and the current joint posterior distribution is used as the prior distribution of the next level. The Bayesian theory is used to integrate the prior distribution of the next level with the corresponding likelihood function to obtain the joint posterior distribution after model fusion. Among them, the cathode emission current under different current acceleration stresses is integrated multiple times according to the experimental progress, and then a Bayesian update is performed to obtain the unknown parameters in the posterior distribution of this level, and finally the life evaluation of the traveling wave tube under specified working conditions is completed.

[0011] Among them, the prior distribution, in engineering, uses uniform distribution or conjugate distribution as the uninformative prior overall distribution of the unknown parameters of the model.

[0012] Among them, the MCMC method is used to solve the unknown parameters of the posterior distribution, and the method based on kernel density estimation is used to update the unknown parameters.

[0013] Among them, a Wiener process model integrating random effects and current acceleration stress is constructed, and the specific implementation steps are as follows:

[0014] When the performance degradation of the cathode of the traveling wave tube can be described by a nonlinear Wiener process, its life distribution obeys an inverse Gaussian distribution; the drift coefficient in the Wiener process is used to characterize the degradation rate of the cathode of the traveling wave tube under different stresses. When the applied sensitive stress is current, the acceleration model uses an inverse power law model to characterize the acceleration effect of current on the performance degradation of the cathode. It is assumed that the drift coefficient is in the following form:

[0015]

[0016] In the formula, S k To accelerate stress, is the current acceleration stress S k The function of k is the stress S k The drift coefficient corresponding to the product below, b and η are unknown parameters;

[0017] Assume that the coefficient η related to the drift coefficient follows a normal distribution, that is, Taking into account the random effects between products due to the use environment, raw materials, processing technology and manufacturing methods, the Wiener process model integrating accelerated stress and random effects is expressed as:

[0018]

[0019] Y(t) is the performance measurement value at time t, λ and σ are the drift coefficient and diffusion coefficient of the Wiener process, which respectively reflect the degradation rate and performance fluctuation of the product performance, B(t) is the standard Brownian motion, are the mean and variance of η respectively.

[0020] Among them, the likelihood function is constructed by analyzing the characteristics of the cathode emission current collected during the accelerated degradation test of the traveling wave tube cathode current, which is specifically:

[0021]

[0022] In the formula, Δy ij =Y(t ij )-Y(t i,j-1 ), Δt ij =t ij -t i,j-1 ,λ i =(λ1,λ2,…,λ n ) represents the drift coefficient of the sample Wiener process under the corresponding working conditions, f(λ i |μ η ,σ η ,σ,b) ​​represents the probability density function corresponding to the model drift coefficient, f(Δy ij |λΔtij ,σ 2 Δt ij ) represents the probability density function that obeys the Wiener process.

[0023] Among them, the likelihood function is integrated with the prior distribution, and the expression of the posterior distribution is obtained as follows:

[0024] P(μ η ,σ η ,σ,bY)∝π(θ)L(Yμ η ,σ η ,σ,b)

[0025] Where π(θ) is the unknown hyperparameter μ in the model η ,σ η ,The prior distribution of σ,b;

[0026] The unknown parameters are solved by the MCMC method to obtain the unknown parameters in the posterior distribution of this level, and the current joint posterior distribution is used as the prior distribution of the next level, that is, the new current prior distribution;

[0027] According to the test progress, the degradation data under different current acceleration stresses are further integrated, and then Bayesian update is performed to obtain the current likelihood function;

[0028] Integrate the new current prior distribution with the current likelihood function to obtain the current joint posterior distribution of the model; reuse the MCMC method to solve the unknown parameters, obtain the unknown parameters in the posterior distribution of this level, and use the current joint posterior distribution as the prior distribution of the next level, that is, the new current prior distribution, until all multi-source performance degradation data under accelerated stress are integrated;

[0029] Among them, by continuously adding performance degradation data under different current acceleration stresses, the estimated values ​​of the hyperparameters in the posterior distribution are continuously updated. The specific implementation process is as follows:

[0030]

[0031] In the formula, Δy ij (S1) represents the incremental performance degradation from stress 1, π(θ) = π(μ η )π(σ η )π(σ)π(b) represents the joint prior distribution of model parameters quantified based on expert experience or historical data, represents the posterior distribution of the model that combines the prior information with the performance degradation increment under stress 1; Δy ij (S2) represents the performance degradation increment under stress 2, is the posterior distribution of the model updated according to the additional performance degradation data under stress 2; Δyij (S3) represents the performance degradation increment under stress 3, The posterior distribution of the model is updated again according to the performance degradation data under the additional stress 3.

[0032] The present invention also provides an electronic device, comprising a processor and a memory for storing executable instructions of the processor; the processor is used to read the executable instructions from the memory and execute the instructions to implement the traveling wave tube life assessment method integrating multi-source degradation data described in the present invention.

[0033] The present invention also provides a computer-readable storage medium, wherein the storage medium stores a computer program, and the computer program is used to execute the traveling wave tube life assessment method for fusing multi-source degradation data of the present invention.

[0034] Beneficial effects:

[0035] 1. In order to study the influence of cathode emission current density on cathode life, the method of the present invention takes into account that accelerated degradation tests with current as sensitive stress have been carried out on the ground, and a large amount of performance degradation data has been accumulated. The degradation data under different accelerated stress conditions are fused to greatly improve the accuracy of life assessment. Specifically, taking into account a large number of accelerated tests on the cathode carried out on the ground, the test data accumulated during the accelerated degradation test with current sensitive stress applied are sorted and analyzed, and the hierarchical Bayesian model and the hierarchical Bayesian fusion framework are used to fuse the cathode performance degradation data under different accelerating current conditions, that is, to fuse the performance degradation data under different test conditions, expand the life data and sample capacity, explore the influence of cathode emission current on cathode life under a certain operating temperature, and evaluate the cathode life of the traveling wave tube.

[0036] 2. In the method of the present invention, the hierarchical Bayesian theory is used to construct a Bayesian fusion model framework that integrates multi-source performance degradation data; by analyzing the characteristics of the performance degradation data collected during the accelerated degradation test of the cathode current of the traveling wave tube, a likelihood function is constructed, and then the prior distribution of the model is quantified, and then the joint posterior distribution of the model is obtained by using the Bayesian theory. By continuously appending performance degradation data under different accelerated stresses and continuously updating the estimated values ​​of the hyperparameters in the posterior distribution, a hierarchical Bayesian model that integrates multi-source performance degradation data is constructed, thereby achieving the purpose of integrating multi-source performance degradation data under different accelerated stresses.

[0037] 3. In the method of the present invention, the probability density function and cumulative distribution function of the product life are obtained according to the total probability formula. After considering the current acceleration stress and random effects, the reliability function of the traveling wave tube cathode under normal working current is obtained, and finally the reliability function of the product under normal working stress is obtained.

[0038] 4. The method of the present invention uses the performance degradation data of the traveling wave tube cathode current accelerated degradation test conducted on the ground to explore the influence of the emission current on the cathode life, filling the gap of only using temperature for acceleration in engineering. First, the influencing factors of hot cathode failure, including current and temperature, are analyzed, and accelerated degradation test data with current as sensitive stress is carried out on the ground; secondly, the Wiener process model integrating current acceleration and random effects is constructed using the collected multi-source degradation data; then, the hierarchical Bayesian fusion framework is constructed using Bayesian theory to fuse the performance degradation data under different accelerated stresses and expand the sample size; finally, the present invention can update the posterior distribution of unknown parameters in the model according to the data addition and update situation, and further update the life assessment result of the traveling wave tube cathode, that is, after the product life assessment result is obtained based on the existing data, when new data is available, the life assessment result can be updated.

[0039] 5. In the medium of the present invention, taking into account a large number of accelerated tests carried out on the ground on the cathode, the test data accumulated during the accelerated degradation test with current sensitive stress applied are sorted and analyzed, and the hierarchical Bayesian model and the hierarchical Bayesian fusion framework are used to fuse the cathode performance degradation data under different accelerating current conditions, that is, to fuse the performance degradation data under different test conditions, expand the life data and sample capacity, explore the influence of the cathode emission current on the cathode life at a certain operating temperature, and evaluate the cathode life of the traveling wave tube.

[0040] 6. The medium of the present invention uses hierarchical Bayesian theory to construct a Bayesian fusion model framework that integrates multi-source performance degradation data; by analyzing the characteristics of the performance degradation data collected during the accelerated degradation test of the cathode current of the traveling wave tube, a likelihood function is constructed, and then the prior distribution of the model is quantified, and then the joint posterior distribution of the model is obtained using Bayesian theory. By continuously adding performance degradation data under different accelerated stresses, the estimated values ​​of the hyperparameters in the posterior distribution are continuously updated, and a hierarchical Bayesian model that integrates multi-source performance degradation data is constructed to achieve the purpose of integrating multi-source performance degradation data under different accelerated stresses. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 The figure is a flow chart of lifespan assessment using Bayesian theorem according to an embodiment of the present invention.

[0042] Figure 2 Schematic diagram of the degradation trajectory of the cathode emission current under current stress acceleration in an embodiment of the present invention.

[0043] Figure 3 Schematic diagram of updating the posterior distribution in an embodiment of the present invention.

[0044] Figure 44 is a reliability distribution diagram in an embodiment of the present invention.

[0045] Figure 5 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0046] The present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0047] There are many factors that affect the life of the cathode. In order to better explore the degradation mechanism of cathode performance and fully integrate the data in the cathode accelerated degradation test process, the present invention proposes a traveling wave tube life assessment method that integrates multi-source degradation data, and integrates the performance degradation data under different current accelerated test conditions to achieve the purpose of improving the accuracy of traveling wave tube life assessment. The degradation characteristics of cathode performance under current accelerated stress are described by constructing a Wiener process model with integrated random effects; a fusion framework of multi-source degradation data is constructed based on hierarchical Bayesian theory, the purpose is to integrate the cathode performance degradation data under different current accelerated stress conditions; under the fusion framework, according to engineering experience, the initial prior distribution of the model hyperparameters is obtained, and then by introducing the acceleration model and random effects, the Bayesian theorem is used to obtain the posterior distribution containing unknown parameters, and then the MCMC method is used to solve the unknown parameters, and the current joint posterior distribution is used as the prior distribution of the next level, and the degradation data under different current accelerated stresses are integrated multiple times according to the test situation, and then the Bayesian update is performed to obtain the unknown parameters in the posterior distribution of this level, and finally the life assessment of the traveling wave tube under specified working conditions is completed.

[0048] The embodiment of the present invention takes the cathode used in a domestic traveling wave tube as a sample to illustrate the specific implementation of the present invention, which is mainly divided into the following steps:

[0049] Step 1: Using the basic theory of Wiener process, a Wiener process model integrating random effects and current acceleration stress is constructed, wherein the Wiener process model uses cathode emission current as sensitive stress for acceleration;

[0050] This step constructs a Wiener process model integrating random effects and current acceleration stress by combing the performance degradation data collected during the accelerated degradation test of the traveling wave tube cathode with current applied as the sensitive stress, based on the analysis of the Wiener process theory and the current acceleration model. The specific implementation steps are as follows:

[0051] When the performance degradation of the cathode of a traveling wave tube can be described by a nonlinear Wiener process, its life distribution obeys an inverse Gaussian distribution. The drift coefficient in the Wiener process is used to characterize the degradation rate of the cathode of the traveling wave tube under different stresses. When the applied sensitive stress is current, the acceleration model uses an inverse power law model to characterize the acceleration effect of current on the performance degradation of the cathode. It is assumed that the drift coefficient can be in the following form:

[0052]

[0053] In the formula, S k To accelerate stress, is the current acceleration stress S k The function of k is the stress S k The drift coefficient corresponding to the product below, b and η are unknown parameters.

[0054] Assume that the coefficient η related to the drift coefficient follows a normal distribution, that is, Taking into account the random effects between products due to factors such as use environment, raw materials, processing technology and manufacturing methods, the Wiener process model integrating accelerated stress and random effects can be expressed as:

[0055]

[0056] Y(t) is the performance measurement value at time t, λ and σ are the drift coefficient and diffusion coefficient of the Wiener process, which respectively reflect the degradation rate and performance fluctuation of the product performance, B(t) is the standard Brownian motion, are the mean and variance of η respectively.

[0057] Step 2: Using the hierarchical Bayesian theory, a Bayesian fusion model framework for fusing multi-source performance degradation data is constructed, wherein the Bayesian fusion model is used to realize the fusion of multi-source performance degradation data under different acceleration stresses; the performance degradation data is the emission current of the cathode of the traveling wave tube under different current acceleration stresses collected;

[0058] Step 3: Based on the analysis of the Wiener process model and Bayesian fusion model framework of integrated random effects and current accelerated stress, a likelihood function integrating the performance degradation data under different accelerated stress conditions is constructed. Specifically, the likelihood function is constructed by analyzing the characteristics of the performance degradation data collected during the traveling wave tube cathode current accelerated degradation test, which is:

[0059]

[0060] In the formula, Δy ij =Y(t ij )-Y(t i,j-1), Δt ij =t ij -t i,j-1 ,λ i =(λ1,λ2,…,λ n ) represents the drift coefficient of the sample Wiener process under the corresponding working conditions, f(λ i |μ η ,σ η ,σ,b) ​​represents the probability density function corresponding to the model drift coefficient, f(Δy ij |λΔt ij ,σ 2 Δt ij ) represents the probability density function that obeys the Wiener process.

[0061] Among them, the reliability function of the product under normal working stress is obtained as follows:

[0062] According to the total probability formula, the probability density function and cumulative distribution function of the product life are obtained, and D is the failure threshold of product performance degradation.

[0063]

[0064] After considering the current acceleration stress and random effects, the reliability function of the cathode of the traveling wave tube under normal working current is:

[0065]

[0066] Where S0 is the normal working current stress.

[0067] Step 4: Under the fusion framework, according to engineering experience, the initial prior distribution of the model hyperparameters is quantitatively obtained, and the current prior distribution is integrated with the current likelihood function using Bayesian theory (to introduce the acceleration model and random effects) to obtain the current joint posterior distribution of the model, i.e., the posterior distribution containing the unknown parameters; wherein, the prior distribution, in engineering, uses uniform distribution or conjugate distribution as the uninformative prior overall distribution of the model's unknown parameters;

[0068] Specifically, the likelihood function constructed in step 3 is integrated with the prior distribution to obtain the expression of the posterior distribution:

[0069] P(μ η ,σ η ,σ,bY)∝π(θ)L(Yμ η ,σ η ,σ,b)(7)

[0070] Where π(θ) is the unknown hyperparameter μ in the model η ,σ η ,σ,b’s prior distribution.

[0071] Step 5: Use the MCMC method to solve the unknown parameters, obtain the unknown parameters in the posterior distribution of this level, and use the current joint posterior distribution as the prior distribution of the next level, that is, the new current prior distribution;

[0072] According to the test progress, the degradation data under different current acceleration stresses are further integrated, and then Bayesian update is performed to obtain the current likelihood function;

[0073] Integrate the new current prior distribution with the current likelihood function to obtain the current joint posterior distribution of the model; return to re-execute step 5 until all multi-source performance degradation data under accelerated stress are integrated;

[0074] This step continuously adds performance degradation data under different current acceleration stresses, thereby continuously updating the estimated values ​​of the hyperparameters in the posterior distribution. The implementation process can be described as follows:

[0075] P S1 (μ η ,σ η ,σ,bΔy ij (S1))∝π(θ)L S1 (Δy ij (S1)μ η ,σ η ,σ,b)

[0076] P S2 (μ η ,σ η ,σ,bΔy ij (S2))∝P S1 (μ η ,σ η ,σ,bΔy ij (S1))L S2 (Δy ij (S2)μ η ,σ η ,σ,b)(8)

[0077] P S3 (μ η ,σ η ,σ,bΔy ij (S3))∝P S2 (μ η ,σ η ,σ,bΔy ij (S2))L S3 (Δy ij (S3)μ η ,σ η ,σ,b)

[0078] In the formula, Δy ij (S1) represents the incremental performance degradation from stress 1, π(θ) = π(μ η )π(σ η )π(σ)π(b) represents the joint prior distribution of model parameters quantified based on expert experience or historical data, P S1 (μ η ,σ η ,σ,bΔy ij (S1)) represents the posterior distribution of the model that combines the prior information with the performance degradation increment under stress 1; Δy ij (S2) represents the performance degradation increment under stress 2, P S2 (μ η ,σ η ,σ,bΔy ij (S2)) is the model posterior distribution updated according to the additional performance degradation data under stress 2; Δy ij (S3) represents the performance degradation increment under stress 3, P S3 (μ η ,σ η ,σ,bΔy ij (S3)) is the model posterior distribution updated again based on the performance degradation data under the additional stress 3.

[0079] Step 6: Use the MCMC method to estimate the unknown hyperparameters in the current joint posterior distribution of the model, and carry out the reliability assessment of the traveling wave tube cathode based on the parameter estimation results. Use formula (5) to obtain the life estimation result of the traveling wave tube under current acceleration, and complete the life assessment of the traveling wave tube under specified working conditions.

[0080] Among them, in the process of solving the unknown parameters of the posterior distribution using the MCMC method, the method based on kernel density estimation can be used to update the unknown parameters.

[0081] Figure 1 The figure is a flow chart of lifespan assessment using Bayesian theorem according to an embodiment of the present invention. Figure 2 Schematic diagram of the degradation trajectory of the cathode emission current under current stress acceleration in an embodiment of the present invention.

[0082] The following example of a domestic traveling wave tube is used to further illustrate the method of the present invention: a cathode current accelerated degradation test was carried out on the ground, and the performance degradation data under different stresses were fully integrated according to the constructed hierarchical Bayesian model. The accelerated stress was S1 = 2.5A / cm 2 and S2=2A / cm 2 , the normal working stress is S3 = 1.5A / cm 2During the test, the cathode emission current was recorded every 1000 hours. The degradation trajectory of the cathode emission current under the current stress acceleration condition is as follows: Figure 2 shown.

[0083] When only S1=2.5A / cm 2 When the acceleration performance degradation data under η , σ η , posterior distribution of σ and b. Add S2 = 2A / cm 2 After the accelerated degradation data under S1 = 2.5A / cm 2 The posterior distribution obtained under the above conditions is used as the prior distribution of the model, and the first update of the posterior parameters of the model is obtained. Similarly, when S3 = 1.5A / cm is added 2 After the normal degradation data under the condition of , the model posterior parameters after the first update are used as the prior distribution of this calculation to obtain the second update of the model posterior parameters. The probability density function diagram of the posterior distribution of the model parameters obtained by the three calculations is shown in Figure 3 The parameter estimates are shown in Table 1.

[0084] Table 1 Posterior distribution estimates of parameter updates

[0085]

[0086]

[0087] The failure threshold of the TWT cathode performance degradation increment under normal transmission current is 12.92 mA. According to the additional parameter estimation results, the reliability function distribution diagram of the TWT cathode is obtained, as shown in Figure 4 shown.

[0088] The multi-source heterogeneous data fusion method was used to obtain the cathode working temperature T = 1100 ° C and emission current J = 1.5 A / cm 2 Under the conditions of , the average cathode life is 189290 hours, or 21.6 years.

[0089] The embodiment of the present application also provides an electronic device, Figure 5The structure of the electronic device provided by the embodiment of the present invention is shown. For example, the electronic device 50 may include a processor 51, a memory 52 and a transmission device 53. The processor 51 is used to execute the traveling wave tube life assessment method for integrating multi-source degradation data mentioned in the above embodiment, wherein the processor and the memory may be connected by a bus or other means, taking the bus connection as an example. The transmission device may be connected to the processor and the memory by wire or wirelessly. The memory, as a non-transient computer-readable storage medium, may be used to store non-transient software programs, non-transient computer executable programs and modules, such as program instructions / modules corresponding to the traveling wave tube life assessment method for integrating multi-source degradation data in the embodiment of the present application. The processor executes various functional applications and data processing of the processor by running the non-transient software programs, instructions and modules stored in the memory, that is, the traveling wave tube life assessment method for integrating multi-source degradation data in the above method embodiment is implemented. The memory may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required by at least one function; the data storage area may store data created by the processor, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage devices. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof. The one or more modules are stored in the memory, and when executed by the processor, the traveling wave tube life assessment method for fusing multi-source degradation data in the embodiment is executed.

[0090] As another aspect, the present application also provides a computer-readable storage medium, which may be a computer-readable storage medium included in the device described in the above embodiment; or a computer-readable storage medium that exists independently and is not assembled into a device. The computer-readable storage medium may be a tangible storage medium, such as a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a floppy disk, a hard disk, a removable storage disk, a CD-ROM, or any other form of storage medium known in the technical field. The computer-readable storage medium stores one or more programs, and the programs are used by one or more processors to execute the traveling wave tube life assessment method for fusing multi-source degradation data described in the present application.

[0091] In summary, the above are only preferred embodiments of the present invention and are not intended to limit the protection scope of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A traveling wave tube life assessment method integrating multi-source degradation data, characterized in that: include: A Wiener process model integrating random effects and current acceleration stress is constructed, in which cathode emission current is used as the sensitive stress for acceleration. Using hierarchical Bayesian theory, a Bayesian fusion model framework is constructed to fuse multi-source performance degradation data; The performance degradation data is the emission current of the cathode of the traveling wave tube collected under different current acceleration stresses; Based on the analysis of the Wiener process model integrating random effects and current acceleration stress and the Bayesian fusion model framework, a likelihood function integrating the performance degradation data under different acceleration stress conditions is constructed; Obtain the initial prior distribution of the model hyperparameters, and use Bayesian theory to integrate the current prior distribution with the current likelihood function to obtain the current joint posterior distribution of the model; The unknown hyperparameters of the current level are obtained, and the current joint posterior distribution is used as the prior distribution of the next level. The Bayesian theory is used to integrate the prior distribution of the next level with the corresponding likelihood function to obtain the joint posterior distribution after model fusion. Among them, the cathode emission current under different current acceleration stresses is integrated multiple times according to the experimental progress, and then a Bayesian update is performed to obtain the unknown parameters in the posterior distribution of this level, and finally the life evaluation of the traveling wave tube under specified working conditions is completed.

2. The method according to claim 1, characterized in that The prior distribution, in engineering, uses uniform distribution or conjugate distribution as the uninformative prior overall distribution of the unknown parameters of the model.

3. The method according to claim 2, characterized in that ,The MCMC method is used to solve the unknown parameters of the posterior distribution, and the kernel density estimation-based method is used to update the unknown parameters.

4. The method according to any one of claims 1 to 3, characterized in that , a Wiener process model integrating random effects and current acceleration stress is constructed. The specific implementation steps are as follows: When the performance degradation of the cathode of the traveling wave tube can be described by a nonlinear Wiener process, its life distribution obeys an inverse Gaussian distribution; the drift coefficient in the Wiener process is used to characterize the degradation rate of the cathode of the traveling wave tube under different stresses. When the applied sensitive stress is current, the acceleration model uses an inverse power law model to characterize the acceleration effect of current on the performance degradation of the cathode. It is assumed that the drift coefficient is in the following form: In the formula, S k To accelerate stress, is the current acceleration stress S k The function of k is the stress S k The drift coefficient corresponding to the product below, b and η are unknown parameters; Assume that the coefficient η related to the drift coefficient follows a normal distribution, that is, Taking into account the random effects between products due to the use environment, raw materials, processing technology and manufacturing methods, the Wiener process model integrating accelerated stress and random effects is expressed as: Y(t) is the performance measurement value at time t, λ and σ are the drift coefficient and diffusion coefficient of the Wiener process, which respectively reflect the degradation rate and performance fluctuation of the product, B(t) is the standard Brownian motion, μ η , are the mean and variance of η respectively.

5. The method according to claim 4, characterized in that By analyzing the characteristics of the cathode emission current collected during the accelerated degradation test of the traveling wave tube cathode current, a likelihood function is constructed, which is: In the formula, Δy ij =Y(t ij )-Y(t i,j-1 ), Δt ij =t ij -t i,j-1 ,λ i =(λ1,λ2,…,λ n ) represents the drift coefficient of the sample Wiener process under the corresponding working conditions, f(λ i |μ η ,σ η ,σ,b) ​​represents the probability density function corresponding to the model drift coefficient, f(Δy ij |λΔt ij ,σ 2 Δt ij ) represents the probability density function that obeys the Wiener process.

6. The method according to claim 5, characterized in that Integrating the likelihood function with the prior distribution, the expression of the posterior distribution is: P(μ η ,s η ,σ,bY)∝π(θ)L(Yμ η ,s η ,s,b) Where π(θ) is the unknown hyperparameter μ in the model η ,σ η ,The prior distribution of σ,b; The unknown parameters are solved by the MCMC method to obtain the unknown parameters in the posterior distribution of this level, and the current joint posterior distribution is used as the prior distribution of the next level, that is, the new current prior distribution; According to the test progress, the degradation data under different current acceleration stresses are further integrated, and then Bayesian update is performed to obtain the current likelihood function; Integrate the new current prior distribution with the current likelihood function to obtain the current joint posterior distribution of the model; reuse the MCMC method to solve the unknown parameters, obtain the unknown parameters in the posterior distribution of this level, and use the current joint posterior distribution as the prior distribution of the next level, that is, the new current prior distribution, until all multi-source performance degradation data under accelerated stress are integrated; Among them, by continuously adding performance degradation data under different current acceleration stresses, the estimated values ​​of the hyperparameters in the posterior distribution are continuously updated. The specific implementation process is as follows: In the formula, Δy ij (S1) represents the incremental performance degradation from stress 1, π(θ) = π(μ η )π(σ η )π(σ)π(b) represents the joint prior distribution of model parameters quantified based on expert experience or historical data, represents the posterior distribution of the model that combines the prior information with the performance degradation increment under stress 1; Δy ij (S2) represents the performance degradation increment under stress 2, is the posterior distribution of the model updated according to the additional performance degradation data under stress 2; Δy ij (S3) represents the performance degradation increment under stress 3, The posterior distribution of the model is updated again according to the performance degradation data under the additional stress 3.

7. An electronic device, characterized in that: The electronic device includes a processor and a memory for storing executable instructions of the processor; the processor is used to read the executable instructions from the memory and execute the instructions to implement the traveling wave tube life assessment method integrating multi-source degradation data as described in any one of claims 1 to 6.

8. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and the computer program is used to execute the traveling wave tube life assessment method integrating multi-source degradation data as described in any one of claims 1 to 6.

Images