Aerospace BDR module life prediction method based on Arrhenius and Wiener models
By employing a hybrid lifetime prediction method based on the Arrhenius and Wiener models, the efficiency and accuracy issues of lifetime assessment for aerospace BDR modules are addressed, achieving efficient and accurate lifetime prediction. This method is applicable to the reliability design and lifetime assessment of aerospace BDR modules.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHANGHAI INST OF SPACE POWER SOURCES
- Filing Date
- 2025-12-18
- Publication Date
- 2026-04-21
AI Technical Summary
Existing technologies cannot quickly and accurately assess the lifespan of aerospace BDR modules, and conventional tests are time-consuming and costly, making it difficult to meet the requirements of aerospace missions for long lifespan and high reliability.
A hybrid lifetime prediction method based on the Arrhenius and Wiener models was adopted. Through step-accelerated degradation experiments, conversion efficiency was selected as a lifetime characteristic parameter. A hybrid model was established by combining temperature stress and stochastic processes to predict lifetime.
It achieves efficient and accurate BDR module life prediction, shortens the test cycle, reduces sample requirements and resource consumption, and has a prediction error of less than 9%, meeting the accuracy requirements of aerospace engineering.
Smart Images
Figure CN121902385A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of aerospace power reliability technology, specifically to a method for predicting the lifetime of aerospace BDR modules based on the Arrhenius and Wiener models. Background Technology
[0002] As the core power conversion component of aerospace power controllers, the BDR module directly determines the power supply stability and service life of the spacecraft's power subsystem. With the development of aerospace missions towards longer lifespans and higher reliability, stringent requirements have been placed on the quantitative assessment of the BDR module's lifespan.
[0003] However, BDR modules suffer from two major technical challenges under normal operating conditions: First, their performance degradation rate is extremely slow, with a natural lifespan exceeding ten years. Obtaining complete degradation data through conventional testing is time-consuming and inefficient. Second, conventional accelerated testing requires a large number of samples to ensure evaluation accuracy, resulting in high testing costs and resource consumption. Furthermore, the degradation process of BDR modules is dominated by temperature stress and involves the synergistic effects of multiple components, making it difficult for a single model to accurately characterize its degradation patterns and lifespan features.
[0004] Existing lifetime prediction methods cannot meet the needs of aerospace engineering for rapid and accurate assessment of BDR module lifetime. Therefore, there is an urgent need to propose a BDR module lifetime prediction method that balances experimental efficiency and prediction accuracy. Summary of the Invention
[0005] The purpose of this invention is to overcome the shortcomings of existing aerospace BDR module life prediction methods, which are mostly aimed at single components or general electronic devices and lack an integrated prediction scheme specifically for aerospace BDR modules, resulting in poor model adaptability, large prediction errors, and long test cycles.
[0006] To achieve the above objectives, this invention provides a method for predicting the lifetime of aerospace BDR modules based on the Arrhenius and Wiener models, comprising the following steps: Step S1, determine the lifetime characteristic parameters of the BDR module: perform equivalent analysis on the power conversion main circuit of the BDR module, screen out the core parameters that can reflect circuit performance degradation and are easy to monitor, and use the conversion efficiency as the lifetime characteristic parameter. The conversion efficiency calculation formula is as follows: (1-1) in, Let be the conversion efficiency at time t. , Let be the output voltage and output current at time t, respectively. , The input voltage and input current at time t are respectively; the conversion efficiency relative to the initial value is set. A 20% degradation rate is the failure threshold, i.e., the failure criterion is: (1-2) Step S2, design a step-accelerated degradation test scheme: using temperature as the accelerating stress, adopt a step-stress loading method, place several BDR module samples in a temperature chamber, conduct accelerated degradation tests under different temperature stress levels, and periodically collect parameter signals of input voltage, output voltage, input current, and output current of the BDR module samples; Step S3, Experimental Data Preprocessing: Perform feature analysis on the parameter signals collected in Step S2, extract degradation features, remove outlier data, and obtain a standardized degradation dataset. (1-3) Step S4, establish a hybrid lifetime prediction model: The relationship between temperature stress and the degradation rate of the BDR module is described based on the Arrhenius model, and the performance degradation evolution process of the module is characterized based on the Wiener stochastic process. The two models are combined to construct a degradation model for the BDR module. The expression for the Arrhenius model is: (1-4) in, For temperature stress The degradation rate is given by the model constants a and b. Let be the i-th step temperature stress, and 273.15 be the thermodynamic temperature conversion coefficient; the expression for the Wiener stochastic process degradation model is: (1-5) in, Let μ be the performance degradation at time t, μ be the drift parameter, and σ be the diffusion parameter. For standard Brownian motion; the expression for the hybrid lifetime prediction model is: (1-6) Where k is the order of the step stress, Let be the degradation rate under the i-th level temperature stress. , Let be the start and end times of the i-th stress level, respectively. It is a random fluctuation term; Step S5, Model parameter identification: Import the preprocessed experimental data into the hybrid lifetime prediction model and solve for the unknown parameters in the hybrid lifetime prediction model; Step S6, Actual Lifetime Extrapolation: Based on the hybrid lifetime prediction model identified in Step S5, the actual lifetime of the BDR module is extrapolated, using the following formula: (1-7); in, The target temperature stress.
[0007] Optionally, in step S2, the number of BDR module samples is greater than or equal to 6.
[0008] Optionally, in step S2, the different temperature stresses refer to those between -55℃ and +125℃.
[0009] Optionally, in step S3, the parameter signal includes any one or more of the core lifetime characteristic parameter signal, basic acquisition parameter signal, and degradation analysis parameter signal.
[0010] Optionally, in step S3, the degradation characteristics include any one or more of the following: mean, peak-to-peak value, root mean square, standard deviation, and conversion efficiency.
[0011] Optionally, in step S5, the unknown parameters in the hybrid lifetime prediction model are solved using the least squares method.
[0012] Optionally, in step S5, the MATLAB least squares toolbox is used to identify the model parameters and solve for the constants a and b of the Arrhenius model and the drift parameter μ and diffusion parameter σ of the Wiener process.
[0013] Optionally, in step S6, the BDR module is defined as having an actual lifespan under typical operating conditions at 25°C.
[0014] Compared to the prior art, the beneficial effects of the present invention include at least the following: This invention adopts a step-accelerated degradation test scheme, which has high test efficiency, greatly shortens the acquisition cycle of degradation data of BDR module, and solves the problem of excessive time consumption in conventional tests.
[0015] This invention establishes a hybrid lifetime prediction model by co-modeling the Arrhenius model and the Wiener model. It can achieve high-precision lifetime prediction with only 6 BDR module samples, which reduces the sample requirement and lowers the experimental cost and resource consumption.
[0016] (3) The hybrid life prediction model established in this invention takes into account both the dominant role of temperature stress and the random characteristics of the degradation process. The extrapolated life at 25℃ has an error of only 8.879% compared with the design life, which is highly accurate and meets the accuracy requirements of aerospace engineering.
[0017] (4) Based on the structural characteristics and working environment of the BDR module, the life characteristic parameters are scientifically selected and easy to monitor. The model formula is simple and practical and can be directly applied to the reliability design and life assessment of aerospace BDR modules. Attached Figure Description
[0018] Figure 1 This is a flowchart illustrating a method for predicting the lifetime of aerospace BDR modules based on the Arrhenius and Wiener models according to the present invention.
[0019] Figure 2 This is a stress profile diagram of the BDR module in the accelerated degradation test of the present invention.
[0020] Figure 3 The temperature chamber (a) and the interior of the temperature chamber (b) are shown in the BDR power board accelerated degradation test of the present invention.
[0021] Figure 4 The following is a dynamic response diagram of the BDR single module load 10A jump bus of the present invention; (a) is a load of 1500W~500W, (b) is a load of 500W~1500W.
[0022] Figure 5 The following is a waveform diagram of the step temperature output of the present invention; (a) shows Vin, Vo, Iin, and Io, and (b) shows the corresponding statistical characteristics, which are the mean, median, and standard deviation.
[0023] Figure 6 The following is an analysis of the step temperature output conversion efficiency of the present invention: (a) represents Pin, Pout, and Efficiency, and (b) represents the corresponding statistical characteristics, which are the mean, median, and standard deviation.
[0024] Figure 7 This is a preprocessed image of the accelerated degradation test results of the BDR power board according to the present invention. Detailed Implementation
[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. It should be noted that the drawings are in a very simplified form and use non-precise proportions, only used to facilitate and clearly illustrate the purpose of the embodiments of the present invention. To make the objectives, features and advantages of the present invention more apparent and understandable, please refer to the drawings. It should be understood that the structures, proportions, sizes, etc., shown in the accompanying drawings are only used to complement the content disclosed in the specification, for those skilled in the art to understand and read, and are not intended to limit the implementation conditions of the present invention. Therefore, they have no substantial technical significance. Any modification to the structure, change in the proportional relationship or adjustment of the size, without affecting the effects and objectives that the present invention can produce, should still fall within the scope of the technical content disclosed in the present invention.
[0026] like Figure 1As shown, this invention provides a method for predicting the lifetime of aerospace BDR modules based on the Arrhenius and Wiener models, comprising the following steps: Step S1: Determine the lifetime characteristic parameters of the BDR module: Perform equivalent analysis on the power conversion main circuit of the BDR module, and model the core components such as power MOSFETs, electrolytic capacitors, and diodes equivalently to derive expressions for performance parameters such as output ripple voltage, ripple coefficient, and conversion efficiency. Select core parameters that reflect circuit performance degradation and are easy to monitor. Considering that conversion efficiency has advantages such as clear physical meaning, convenient monitoring (only requiring the acquisition of input and output voltage and current), and significant degradation characteristics, conversion efficiency is chosen as the core lifetime characteristic parameter of the BDR module. The conversion efficiency calculation formula is: (1-1) in, Let be the conversion efficiency at time t. , Let be the output voltage and output current at time t, respectively. , The input voltage and input current at time t are respectively; the conversion efficiency relative to the initial value is set. A 20% degradation rate is the failure threshold, i.e., the failure criterion is: (1-2).
[0027] Step S2: Design a step-accelerated degradation test scheme: Using temperature as the accelerating stress (consistent with the main failure induction factors of BDR modules in aerospace environments), a step-stress loading method is adopted, and multiple sets of gradient temperature stress profiles are designed. Several BDR module samples are placed in the temperature chamber of the Saibao Laboratory of the Fifth Research Institute of the Ministry of Industry and Information Technology, and accelerated degradation tests are conducted under different temperature stress levels. The input voltage (Vin), output voltage (Vo), input current (Iin), and output current (Io) parameter signals of the BDR module samples are collected at regular intervals, the conversion efficiency is calculated, and the statistical characteristics of the parameter signals, such as mean, peak-to-peak value, root mean square, and standard deviation, are analyzed. The test period is from February 2021 to October 2021.
[0028] In some embodiments, the number of BDR module samples is greater than or equal to 6.
[0029] Step S3, Experimental Data Preprocessing: Perform feature analysis on the parameter signals collected in Step S2, remove abnormal data caused by equipment errors and sudden interference, calculate the conversion efficiency value at each time point, plot the conversion efficiency-time variation curve, extract the degradation trend term of conversion efficiency, and obtain a standardized degradation dataset. (1-3) Step S4, establish a hybrid lifetime prediction model: The relationship between temperature stress and the degradation rate of the BDR module is described based on the Arrhenius model, and the performance degradation evolution process of the module is characterized based on the Wiener stochastic process. The two models are combined to construct a degradation model for the BDR module. The expression for the Arrhenius model is: (1-4) in, For temperature stress The degradation rate is given by the model constants a and b. Let be the i-th step temperature stress, and 273.15 be the thermodynamic temperature conversion coefficient; the expression for the Wiener stochastic process degradation model is: (1-5) in, Let μ be the performance degradation at time t, μ be the drift parameter, and σ be the diffusion parameter. The model represents standard Brownian motion. Combining the stress loading characteristics of step acceleration tests, the Arrhenius model and the Wiener stochastic process are fused to obtain a hybrid lifetime prediction model suitable for the BDR module. This model can simultaneously reflect the influence of temperature stress and the stochastic characteristics of the degradation process. The expression for the hybrid lifetime prediction model is: (1-6) Where k is the order of the step stress, Let be the degradation rate under the i-th level temperature stress. , Let be the start and end times of the i-th stress level, respectively. It is a random fluctuation term; Step S5, Model Parameter Identification: The preprocessed degradation data is imported into the hybrid lifetime prediction model. The unknown parameters of the model are solved using the MATLAB least squares toolbox to obtain the constants a and b of the Arrhenius model and the drift parameter μ and diffusion parameter σ of the Wiener process, thus completing the model calibration.
[0030] Step S6, Actual Lifetime Extrapolation: Based on the hybrid lifetime prediction model identified in Step S5, the actual lifetime of the BDR module is extrapolated, using the following formula: (1-7).
[0031] In some embodiments, the calibrated hybrid lifetime prediction model is extrapolated to a typical operating condition of 25°C, and the degradation rate λ at that temperature is obtained by substituting it into the Arrhenius model. Combined with the failure threshold criterion of the Wiener process, the actual lifetime of the BDR module at 25°C is calculated.
[0032] Example like Figures 2-7 As shown, six self-made aerospace BDR power boards were selected as BDR module samples. The packaging of the BDR module samples had no special requirements, the operating temperature range was -55~+125℃, and the main function was to realize power conversion.
[0033] like Figure 4 As shown, the dynamic response performance (such as output voltage stability and current regulation speed) of a single BDR module under extreme load changes was first verified to ensure that the basic functions of the module were qualified before subsequent accelerated degradation tests were carried out. Six self-made aerospace BDR power boards were subsequently used as BDR module samples to reduce random errors and improve the accuracy of model parameter identification through multi-sample data.
[0034] like Figure 5 As shown, the correlation between temperature and parameter stability is intuitively presented, verifying the premise that temperature dominates degradation, and providing reliable raw data for conversion efficiency calculation and model stochastic term design.
[0035] like Figure 6 As shown, Figure 6 (a) intuitively reflects the power loss law and conversion efficiency degradation trend of BDR module under different temperature stress levels - the higher the temperature stress, the faster the conversion efficiency decreases, which verifies that temperature is the dominant stress for module degradation. Figure 6 (b) represents the corresponding statistical characteristics, which are the mean, median, and standard deviation, demonstrating the stability of the parameters: as the temperature increases, the standard deviation of the conversion efficiency increases, indicating that high temperature will exacerbate the fluctuation of module performance, providing experimental basis for the introduction of random terms in the model.
[0036] The experiment was conducted at the Saibao Laboratory (Guangzhou) of the Fifth Research Institute of the Ministry of Industry and Information Technology, using a high-precision temperature chamber (non-static-proof equipment). A stepped temperature stress profile was designed, with temperature gradients including multiple levels such as 25℃, 45℃, 65℃, 85℃, 105℃, and 125℃. Each stress level lasted for a certain period, with the total experimental period from February 2021 to October 2021. Data acquisition utilized monitoring equipment such as the ST100, collecting input voltage Vin, output voltage Vo, input current Iin, and output current Io for each sample at fixed intervals. The sampling frequency was 100KHz, with each acquisition lasting 100 seconds. Statistical characteristics such as mean, median, and standard deviation were recorded.
[0037] The data preprocessing stage adopts The criteria eliminate outliers in voltage and current data to ensure data validity. According to the formula... Calculate the conversion efficiency at each time point. Plot the conversion efficiency-time variation curves for the six BDR module samples, extract the degradation trend term for each curve, and obtain a standardized degradation dataset.
[0038] Then, a mixture model is constructed. The expression for the Arrhenius model is: in, For temperature stress The degradation rate is given by the model constants a and b. denoted as i-th step temperature stress (unit: °C), and 273.15 is the thermodynamic temperature conversion coefficient.
[0039] The expression for the Wiener stochastic process degradation model is: Where X(t) is the conversion efficiency degradation at time t, μ is the drift parameter, σ is the diffusion parameter, and B(t) is the standard Brownian motion.
[0040] By combining the stress loading characteristics of step acceleration tests, the Arrhenius model and the Wiener stochastic process are fused to obtain a hybrid degradation model suitable for the BDR module, which can simultaneously reflect the influence of temperature stress and the stochastic characteristics of the degradation process. The hybrid degradation model corresponding to this embodiment is as follows: The preprocessed degraded data was imported into MATLAB, and the parameters were solved using the least squares toolbox, yielding a = 0.00061 and b = 945.6. Therefore, the actual lifespan (in years) of the BDR power board at 25℃ was extrapolated as follows: Based on accelerated degradation test data, the actual lifespan of the BDR module at 25℃ is extrapolated to be 10.8879 years, which is 8.879% less than the designed lifespan of 10 years.
[0041] like Figure 7 As shown in the figure, the abnormal data in the voltage and current acquisition process were eliminated by the 3σ criterion, and the degree of degradation of the conversion efficiency relative to the initial value was standardized. The curve clearly shows the consistent degradation trend of the six samples, which proves that the preprocessed data has high validity and provides reliable input for parameter identification of the hybrid model.
[0042] In summary, this invention provides a life prediction method for aerospace BDR modules based on the Arrhenius and Wiener models, comprising the following steps: Step S1, determining the life characteristic parameters of the BDR module; Step S2, designing a step-accelerated degradation test scheme; Step S3, preprocessing the test data from Step S2; Step S4, establishing a hybrid life prediction model; Step S5, importing the preprocessed test data into the hybrid life prediction model for model parameter identification; Step S6, extrapolating the actual life of the BDR module based on the identified hybrid life prediction model. This invention achieves actual life prediction of BDR modules under typical operating conditions at 25℃, with a prediction error of less than 9%, meeting the development requirements of high reliability and long life of aerospace electronic products, and achieving high test results with low sample consumption.
[0043] Although the present invention has been described in detail through the preferred embodiments above, it should be understood that the above description should not be considered as a limitation of the present invention. Various modifications and substitutions to the present invention will be apparent to those skilled in the art after reading the above description. Therefore, the scope of protection of the present invention should be defined by the appended claims.
Claims
1. A method for predicting the lifetime of aerospace BDR modules based on the Arrhenius and Wiener models, characterized in that, Includes the following steps: Step S1, determine the lifetime characteristic parameters of the BDR module: perform equivalent analysis on the power conversion main circuit of the BDR module, screen out the core parameters that can reflect circuit performance degradation and are easy to monitor, and use the conversion efficiency as the lifetime characteristic parameter. The conversion efficiency calculation formula is as follows: (1-1) in, Let be the conversion efficiency at time t. , Let be the output voltage and output current at time t, respectively. , The input voltage and input current at time t are respectively; the conversion efficiency relative to the initial value is set. A 20% degradation rate is the failure threshold, i.e., the failure criterion is: (1-2) Step S2, design a step-accelerated degradation test scheme: using temperature as the accelerating stress, adopt a step-stress loading method, place several BDR module samples in a temperature chamber, conduct accelerated degradation tests under different temperature stress levels, and periodically collect parameter signals of input voltage, output voltage, input current, and output current of the BDR module samples; Step S3, Experimental Data Preprocessing: Perform feature analysis on the parameter signals collected in Step S2, extract degradation features, remove outlier data, and obtain a standardized degradation dataset. (1-3) Step S4, establish a hybrid lifetime prediction model: The relationship between temperature stress and the degradation rate of the BDR module is described based on the Arrhenius model, and the performance degradation evolution process of the module is characterized based on the Wiener stochastic process. The two models are combined to construct a degradation model for the BDR module. The expression for the Arrhenius model is: (1-4) in, For temperature stress The degradation rate is given by the model constants a and b. Let be the i-th step temperature stress, and 273.15 be the thermodynamic temperature conversion coefficient; the expression for the Wiener stochastic process degradation model is: (1-5) in, Let μ be the performance degradation at time t, μ be the drift parameter, and σ be the diffusion parameter. For standard Brownian motion; the expression for the hybrid lifetime prediction model is: (1-6) Where k is the order of the step stress, Let be the degradation rate under the i-th level temperature stress. , Let be the start and end times of the i-th stress level, respectively. It is a random fluctuation term; Step S5, Model parameter identification: Import the preprocessed experimental data into the hybrid lifetime prediction model and solve for the unknown parameters in the hybrid lifetime prediction model; Step S6, Actual Lifetime Extrapolation: Based on the hybrid lifetime prediction model identified in Step S5, the actual lifetime of the BDR module is extrapolated, using the following formula: (1-7)； in, The target temperature stress.
2. The aerospace BDR module lifetime prediction method as described in claim 1, characterized in that, In step S2, the number of BDR module samples is greater than or equal to 6.
3. The aerospace BDR module lifetime prediction method as described in claim 1, characterized in that, In step S2, the different temperature stresses refer to those between -55℃ and +125℃.
4. The aerospace BDR module lifetime prediction method as described in claim 1, characterized in that, In step S3, the parameter signal includes any one or more of the core lifetime characteristic parameter signal, basic acquisition parameter signal, and degradation analysis parameter signal.
5. The aerospace BDR module lifetime prediction method as described in claim 1, characterized in that, In step S3, the degradation characteristics include any one or more of the following: mean, peak-to-peak value, root mean square, standard deviation, and conversion efficiency.
6. The aerospace BDR module lifetime prediction method as described in claim 1, characterized in that, In step S5, the unknown parameters in the hybrid lifetime prediction model are solved using the least squares method.
7. The aerospace BDR module lifetime prediction method as described in claim 6, characterized in that, In step S5, the MATLAB least squares toolbox is used to identify the model parameters and solve for the constants a and b of the Arrhenius model and the drift parameter μ and diffusion parameter σ of the Wiener process.
8. The aerospace BDR module lifetime prediction method as described in claim 1, characterized in that, In step S6, the BDR module is defined as having an actual lifespan under typical operating conditions at 25°C.