Crystal oscillator aging prediction method, system and device and storage medium

Abstract

The invention provides a crystal oscillator aging prediction method, system and device and a storage medium, and the method comprises the steps: obtaining an aging test data set, and building a comprehensive aging model based on the aging test data set; generating an initial joint model by combining the introduced improved temperature acceleration model with the comprehensive aging model; optimizing joint parameters of the initial joint model according to the aging test data set and a parameter optimization algorithm to generate joint optimization parameters; substituting the joint optimization parameters into the initial joint model to determine a joint model; and substituting the obtained aging data into the joint mode, predicting the aging behavior of the crystal oscillator under the current condition, and obtaining a long-term frequency deviation prediction result. According to the method, the accuracy of aging prediction of the crystal oscillator is remarkably improved.

Description

Technical Field

[0001] This application relates to the field of crystal oscillators, specifically to a method, system, device, and storage medium for predicting the aging of crystal oscillators. Background Technology

[0002] Quartz crystal oscillators are critical frequency control components in electronic devices, and their frequency stability directly affects the performance of the entire system. Due to internal physicochemical changes, the output frequency of the crystal oscillator will systematically drift over time; this phenomenon is called aging. Frequency drift caused by aging is a key parameter for evaluating the long-term reliability of crystal oscillators and will directly affect the long-term operating performance of communication systems, navigation equipment, and precision measuring instruments.

[0003] Currently, the aging models used in the industry for research and prediction of crystal oscillator aging are overly simplistic. Many existing technologies use simple linear or single logarithmic models to describe the aging process, failing to fully consider the complexity of the aging mechanism, or only considering the linear relationship between aging and the logarithm of time, which can lead to significant deviations in actual long-term predictions. Therefore, there is an urgent need in this field for a crystal oscillator aging prediction and warranty period assessment method that can accurately describe the multi-mechanism composite aging process, reliably correlate temperature acceleration effects, and is easy to apply in engineering. Summary of the Invention

[0004] To address the problem that the current technology is not accurate enough in predicting the aging degree of crystal oscillators in practical applications, this application provides a method, system, device and storage medium for predicting the aging of crystal oscillators.

[0005] The first aspect of this application provides a method for predicting the aging of a crystal oscillator, including: A comprehensive aging model is established based on the acquired aging test dataset; An initial joint model is generated by combining an improved temperature acceleration model with a comprehensive aging model. The joint parameters of the initial joint model are optimized based on the aging test dataset and the parameter optimization algorithm to generate joint optimized parameters; Substitute the joint optimization parameters into the initial joint model to determine the joint model; By substituting the acquired aging data into the joint model, the long-term frequency deviation prediction results are obtained.

[0006] In a possible implementation, the method further includes: The time from the start of use of the crystal oscillator to its frequency drift exceeding the frequency tolerance limit is calculated by combining the preset frequency tolerance limit of the crystal oscillator with the long-term frequency deviation prediction results.

[0007] In a possible implementation, the method further includes: Based on the long-term frequency deviation prediction results and at least one preset specific time interval, the average aging rate of the crystal oscillator in each interval is calculated. The long-term stability of the crystal oscillator is determined based on the average aging rate.

[0008] In a possible implementation, establishing a comprehensive aging model based on the acquired aging test dataset includes: The comprehensive aging model is determined based on the first aging mechanism parameters, the first time, the second aging mechanism parameters, the second time, the linear drift coefficient, the aging time, the initial frequency deviation, and the random fluctuation term.

[0009] In a possible implementation, the initial frequency deviation is obtained in the following ways: The initial temperature of each temperature group sample is obtained, and the functional relationship between the initial frequency deviation and the initial temperature is established by linear fitting or polynomial fitting to determine the initial frequency deviation.

[0010] In a possible implementation, optimizing the joint parameters of the initial joint model based on the aging test dataset and the parameter optimization algorithm to generate joint optimized parameters includes: The joint parameters of the initial joint model are initially estimated using the least squares method to obtain the initial values ​​of the initial joint model parameters; The initial values ​​are iteratively adjusted using the simplex optimization algorithm until the numerical value of the constructed objective function converges to a preset threshold range, thereby generating joint optimization parameters.

[0011] In a possible implementation, the step of substituting the acquired aging data into the joint mode to predict the long-term frequency deviation prediction result includes: The normal operating temperature of the crystal oscillator is obtained and substituted into the joint model of the comprehensive aging model and the temperature acceleration model to calculate the relative frequency deviation of the crystal oscillator at the normal operating temperature within the preset aging time. Based on the calculated relative frequency deviations corresponding to different aging times, a long-term aging prediction curve is generated to show how the relative frequency deviation changes with aging time. This prediction curve is the long-term frequency deviation prediction result.

[0012] A second aspect of this application provides a crystal oscillator aging prediction system, comprising: The first building module is used to establish a comprehensive aging model based on the acquired aging test dataset; The second building block is used to generate an initial joint model by combining the improved temperature acceleration model with the integrated aging model. The parameter calculation module is used to optimize the joint parameters of the initial joint model based on the aging test dataset and the parameter optimization algorithm, and generate joint optimized parameters. The optimization module is used to substitute the joint optimization parameters into the initial joint model to determine the joint model; The prediction module is used to input the acquired aging data into the joint model to predict the long-term frequency deviation.

[0013] A third aspect of this application provides a computer device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method described above.

[0014] A fourth aspect of this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0015] As can be seen from the above technical solution, this application provides a comprehensive aging model with multiple time logarithmic terms combined with linear drift terms. Each sub-term corresponds to a type of aging mechanism. The first time logarithmic term corresponds to stress relaxation, the second time logarithmic term corresponds to electrode changes, and the linear term corresponds to long-term changes. This application ensures that the aging model can realistically reproduce the actual aging process of the crystal oscillator by corresponding the aging mechanism to the aging model sub-terms one-to-one, avoiding the inherent bias of a single model using a single trend to fit multiple aging mechanisms. In addition, different aging mechanisms have vastly different sensitivities to temperature. This application introduces an improved Arrhenius model, which configures activation energy parameters separately for each type of aging mechanism, and establishes a quantitative relationship between the aging rate and temperature for each type of aging mechanism, ensuring that the aging behavior at different temperatures can be seamlessly connected. Finally, this application sets the initial joint model with joint optimization parameters obtained through parameter optimization algorithms. The resulting joint model covers multiple mechanisms, accurately correlates with temperature, and the joint model parameters have been verified by multiple temperature data. The extrapolation results can realistically reflect the long-term aging trend of the crystal oscillator, significantly improving the accuracy of crystal oscillator aging prediction. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic flowchart of a crystal oscillator aging prediction method according to an embodiment of this application.

[0018] Figure 2 This is a schematic diagram of the aging prediction results in the embodiments of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] Based on this, this application provides an implementation method for predicting the aging of a crystal oscillator, such as... Figure 1 As shown, the method includes: S101. Based on the acquired aging test dataset, a comprehensive aging model is established. S102, by introducing an improved temperature acceleration model and combining it with a comprehensive aging model, an initial joint model is generated; S103, optimize the joint parameters of the initial joint model based on the aging test dataset and parameter optimization algorithm to generate joint optimized parameters; S104, Substitute the joint optimization parameters into the initial joint model to determine the joint model; S105, the acquired aging data is substituted into the joint model to predict the long-term frequency deviation prediction result.

[0021] It should be noted that this application conducts long-term aging tests on the same batch of crystal oscillator samples under at least two different temperature conditions, records the output frequency values ​​of the samples at each time point, and generates an aging test dataset containing relative frequency deviations based on the recorded output frequency values. The specific execution steps include: selecting crystal oscillator samples from the same production batch with the same nominal frequency, dividing the samples into at least two groups, with no fewer than 15 samples in each group; placing the grouped samples in constant temperature chambers with different temperature settings, the temperature settings of the constant temperature chambers being between 70℃ and 125℃, and the temperature settings of each group being different, with the aging test lasting for no less than 6 months; using a quasi-second pulse signal as a frequency reference, measuring the output frequency value of each group of samples at the current test time at the same time point every day; calculating the relative frequency deviation of each sample at the corresponding test time point according to the nominal frequency of the crystal oscillator, the calculation method being: the difference between the output frequency value of the sample at that time point and the nominal frequency, divided by the nominal frequency, and integrating the relative frequency deviations of all samples at each time point to form the aging test dataset. It should be noted that the frequency measurement process meets the following requirements: a high-precision frequency counter is used for frequency measurement, and the measurement accuracy of the frequency counter is not less than 1×10⁻⁶. -12 Frequency data acquisition is automated to avoid the influence of human error on the measurement results. The data presented in this application are for illustrative purposes only and are not intended to be limiting.

[0022] It should be noted that selecting samples from the same batch in this application avoids interference from individual differences between samples from different production batches on the aging data, ensuring that the test data only reflects the effects of temperature and time on aging. For example, samples are initially screened before testing to remove samples whose initial frequency deviation exceeds the preset acceptable range, thus preventing samples with abnormal initial conditions from affecting the validity of subsequent data. It should also be noted that the aging test dataset in this application refers to a set containing the relative frequency deviations of samples at various temperature conditions and time points, where the relative frequency deviation is generally used to intuitively reflect the degree of frequency drift.

[0023] Crystal oscillator aging is typically the result of multiple mechanisms, including quartz plate stress relaxation, electrode quality changes, and long-term physicochemical changes. This application's comprehensive aging model incorporates multiple time-logarithmic terms combined with linear drift terms. Each term corresponds to a specific aging mechanism: the first time-logarithmic term corresponds to stress relaxation, the second time-logarithmic term to electrode changes, and the linear term to long-term changes. By establishing a one-to-one correspondence between aging mechanisms and model components, this application ensures the aging model accurately reflects the actual aging process of the crystal oscillator, avoiding the inherent bias of a single model using a single trend to fit multiple aging mechanisms. Furthermore, different aging mechanisms exhibit vastly different sensitivities to temperature. This application introduces an improved Arrhenius model, configuring activation energy parameters separately for each aging mechanism and establishing a quantitative relationship between the aging rate and temperature for each mechanism. This ensures unbiased connection between aging behaviors at different temperatures. Finally, this application uses joint optimization parameters obtained through a parameter optimization algorithm to set the initial joint model. The resulting joint model covers multiple mechanisms, accurately correlates with temperature, and its parameters have been validated with multiple temperature data. The extrapolation results accurately reflect the long-term aging trend of the crystal oscillator, significantly improving the accuracy of crystal oscillator aging prediction.

[0024] In one embodiment that can be implemented in this application, the method further includes: The time from the start of use of the crystal oscillator to its frequency drift exceeding the frequency tolerance limit is calculated by combining the preset frequency tolerance limit of the crystal oscillator with the long-term frequency deviation prediction results.

[0025] It should be noted that the theoretical warranty period of the product is the time from the start of use of the crystal oscillator until its frequency drift exceeds the frequency tolerance limit.

[0026] In this embodiment, the frequency tolerance limit refers to the maximum frequency drift range allowed during the use of the crystal oscillator. Its value is determined by the product specifications or user requirements. For example, the frequency tolerance limit of ordinary consumer-grade crystal oscillators is ±5ppm, and the frequency tolerance limit of crystal oscillators used in high-precision navigation equipment is ±1ppm. The core logic of the warranty period calculation is to find the maximum time in the long-term aging prediction curve where the absolute value of the relative frequency deviation is ≤ the frequency tolerance limit. Exceeding this maximum time point means that the frequency drift of the crystal oscillator will exceed the allowable range and cannot meet the usage requirements.

[0027] It should be noted that the theoretical warranty period calculated in this application is based on the model prediction results. In actual applications, appropriate adjustments may be made based on engineering experience, such as reserving a safety margin of 10% to 20%. However, the model prediction results provide a scientific quantitative basis for the determination of the warranty period. For example, if the crystal oscillator is used in a scenario with temperature fluctuations, such as the temperature of an automotive crystal oscillator changing with the environment, the fluctuating temperature curve can be substituted into the joint model to calculate the warranty period under dynamic temperatures. It is not limited to a constant normal temperature, and this application does not impose any restrictions on this.

[0028] In one embodiment that can be implemented in this application, the method further includes: S201, based on the long-term frequency deviation prediction results and at least one preset specific time interval, calculate the average aging rate of the crystal oscillator in each interval. S202, determine the long-term stability of the crystal oscillator based on the average aging rate.

[0029] In this embodiment, the average aging rate is calculated by (relative frequency deviation at the end of the time interval - relative frequency deviation at the beginning of the time interval) ÷ ​​(end time - beginning time). This value reflects the average rate of frequency drift within the interval. The smaller the absolute value of the average aging rate, the more stable the crystal oscillator's frequency is within that interval. For example, two consecutive time intervals are typically selected for comparison: the early interval (1-5 years) and the later interval (5-10 years). If the absolute value of the average aging rate in the later interval is less than the average aging rate in the early interval, and the trend is gradual (e.g., the difference in average aging rate between the two intervals is less than 0.1 ppm / year), it indicates that the crystal oscillator has good long-term stability.

[0030] It should be noted that stability assessment focuses not only on the magnitude of the average aging rate, but also on the trend of its change. Even if the average aging rate in a certain range is small, if it shows a significant increasing trend, such as an increase of 0.2 ppm / year, it indicates that there is a risk to the long-term stability of the crystal oscillator.

[0031] This application selects early and late timeframes to compare the stability of crystal oscillators at different stages. By clearly defining the calculation method for the average aging rate, this application ensures the consistency of evaluation indicators. Subsequently, the stability of the crystal oscillator is evaluated based on the magnitude and trend of the average aging rate, which is the most intuitive evaluation method. It should be noted that the results of the stability evaluation can serve as a basis for product grading. For example, crystal oscillators with an absolute average aging rate of less than 0.5 ppm / year are rated as Grade 1 products, and crystal oscillators with an absolute average aging rate of less than 1 ppm / year but greater than 0.5 ppm / year are rated as Grade 2 products.

[0032] In one embodiment of this application, establishing a comprehensive aging model based on the acquired aging test dataset includes: The comprehensive aging model is determined based on the first aging mechanism parameters, the first time, the second aging mechanism parameters, the second time, the linear drift coefficient, the aging time, the initial frequency deviation, and the random fluctuation term.

[0033] This step in the application establishes a comprehensive aging model based on the aging test dataset. The comprehensive aging model includes a time logarithmic term describing different aging mechanisms, a linear drift term reflecting long-term stable change trends, a temperature-related initial frequency deviation term, and a random fluctuation term, which are used to characterize the multi-mechanism composite aging behavior of the crystal oscillator.

[0034] It should be noted that the expression for determining the comprehensive aging model is as follows: ;in, , The relevant parameters for the first aging mechanism, , For the relevant parameters of the second aging mechanism, The linear drift coefficient is... For temperature Related initial frequency deviation, For random fluctuations, Absolute temperature The aging time is represented by the first aging mechanism, which corresponds to the stress relaxation process of the quartz plate inside the crystal oscillator. The second aging mechanism corresponds to the electrode mass change process. The linear drift term corresponds to the long-term stable physicochemical change trend.

[0035] Specifically, in the embodiments of this application, the core design idea of ​​the comprehensive aging model is to cover multiple aging mechanisms, which is different from the simplified models of a single mechanism in the prior art. This application uses a combination of multiple time logarithmic terms and linear drift terms to correspond to different physicochemical processes inside the crystal oscillator. For example, the first time logarithmic term corresponds to the stress relaxation process of the quartz plate, which is characterized by a logarithmic frequency drift in the early stage of aging and gradually slows down over time. The second time logarithmic term corresponds to the mass change process of the electrode, such as the migration or oxidation of electrode atoms, whose rate is more consistent with the logarithmic relationship of time. The linear drift term corresponds to the long-term stable changes of the packaging environment or internal materials, such as the slow leakage of trace gases in the package, which is characterized by a stable linear frequency drift. It should be noted that the random fluctuation term is used in the model of this application to accommodate measurement errors or accidental factors, such as small frequency jumps caused by short-term voltage fluctuations. Its value is usually much smaller than other terms and will not affect the judgment of the overall aging trend.

[0036] For example, if the test environment is highly stable, such as a test environment where the constant temperature accuracy is controlled within ±0.1℃ and the power supply voltage fluctuation is less than 0.1%, the effect of random fluctuation on the aging of the crystal oscillator can be ignored, and the model can be simplified to a multi-time logarithmic term + a linear drift term + a temperature-related initial deviation term.

[0037] In one embodiment that can be implemented in this application, the method for obtaining the initial frequency deviation includes: The initial temperature of each temperature group sample is obtained, and the functional relationship between the initial frequency deviation and the initial temperature is established by linear fitting or polynomial fitting to determine the initial frequency deviation.

[0038] Specifically, this application establishes a functional relationship between the initial frequency deviation and temperature T by measuring the frequency deviation of samples in each temperature group at the initial moment of the aging test, i.e., at t=0, and using linear or polynomial fitting methods. or ,in The fitting coefficients are denoted as .

[0039] In this embodiment, the fitting method for the temperature-related initial frequency deviation is either linear fitting or polynomial fitting. In practical applications, the specific fitting method is selected based on the trend of the relationship between the initial frequency deviation and temperature. For example, if the measured data shows that the initial frequency deviation and temperature have an approximately linear relationship (e.g., the initial frequency deviation increases by 0.1 ppm for every 10°C increase in temperature), linear fitting is used based on the current acquisition scenario. If the initial frequency deviation and temperature have a non-linear relationship (e.g., linear at low temperatures and non-linear at high temperatures), then polynomial fitting is used. During the fitting process, the first fitting coefficient, the second fitting coefficient, and the third fitting coefficient are constants calculated using the least squares method, and their values ​​reflect the correlation strength between the initial frequency deviation and temperature.

[0040] It should be noted that the initial frequency deviation refers to the frequency deviation when the aging time is 0. The relationship between the initial frequency deviation and temperature is mainly determined by the temperature characteristics of the crystal oscillator, such as the temperature coefficient of the quartz plate. The purpose of fitting in the current step is to accurately substitute the initial deviation at different temperatures in the subsequent model calculation.

[0041] This application is not limited to this. It should be noted that, for example, if the current test temperature range is 70℃~90℃, the test temperature range is narrow, only 20℃, and the nonlinearity between the initial frequency deviation and the temperature is not obvious, then the fitting method with the mean value of the initial deviation at all temperatures can be used, which will result in higher accuracy.

[0042] The temperature acceleration model introduced in this application includes the following steps: A modified Arrhenius model is adopted as the temperature acceleration model, with the expression: ;in, X is the aging rate parameter at temperature T, X0 is the aging rate parameter at the reference temperature, and E is the aging rate parameter at temperature T. Here, k represents the activation energy of the aging mechanism, and k is the Boltzmann constant, specifically 8.617 × 10⁻⁶. -5 eV / K, where T is absolute temperature; independent activation energy parameters E are configured for different aging mechanisms in the comprehensive aging model. 1. E 2. E 3, corresponding to the first aging mechanism, the second aging mechanism, and the linear drift process, respectively.

[0043] Specifically, in this application's embodiments, the improved Arrhenius model expression is first clearly defined, where the Boltzmann constant is a known physical constant and does not require estimation; secondly, this application configures independent activation energy parameters for different aging mechanisms. Since the temperature sensitivity of different mechanisms varies significantly, independent configuration ensures the accuracy of the temperature acceleration model. It should be noted that the improved Arrhenius model differs from the traditional model in that it configures independent activation energies for multiple mechanisms. Traditional models typically use a single activation energy, which cannot adapt to multi-mechanism scenarios.

[0044] For example, if the relationship between the rate and temperature of a certain type of aging mechanism does not conform to the Arrhenius model, other temperature-accelerated models can be substituted. For instance, the Arrhenius model is not applicable to some gas leakage processes, so the Eyring model can be selected instead. The selected model only needs to be able to quantify the relationship between rate and temperature, and this application does not impose any restrictions on it.

[0045] In one embodiment of this application, optimizing the joint parameters of the initial joint model based on the aging test dataset and the parameter optimization algorithm to generate joint optimized parameters includes: S301, The joint parameters of the initial joint model are initially estimated using the least squares method to obtain the initial values ​​of the initial joint model parameters; S302, the initial value is iteratively adjusted using the simplex optimization algorithm until the value of the constructed objective function converges to a preset threshold range, thereby generating joint optimization parameters.

[0046] The specific steps for generating joint optimization parameters include: constructing the objective function: ,in The measured relative frequency deviation in the aging test dataset. The predicted values ​​are obtained by combining the aging model and the temperature acceleration model. Initial parameters are estimated using the least squares method to obtain initial values ​​for the joint model parameters. Starting from these initial values, the simplex optimization algorithm is used for iterative optimization, adjusting the model parameters until the objective function converges to a preset threshold. The converged parameters are output as the joint optimization parameters, including… .

[0047] In this embodiment, the constructed objective function is the core of parameter optimization. Its minimization ensures that the overall deviation between the model prediction and the measured data is minimized. The least squares method is used to quickly obtain the initial parameters, avoiding the subsequent optimization algorithm from getting stuck in a large-scale search. The simplex optimization algorithm is used to finely adjust the parameters. Its advantage is that it does not require gradient calculation and has good adaptability to complex nonlinear models. The output joint optimization parameters include all parameters of the integrated model and the temperature acceleration model, which is the basis for subsequent extrapolation.

[0048] It should be noted that the convergence threshold for parameter iterative optimization is set according to the test accuracy requirements. The smaller the convergence threshold, the higher the parameter accuracy, but the longer the calculation time.

[0049] In one embodiment of this application, the step of substituting the acquired aging data into the joint mode to predict the long-term frequency deviation prediction result includes: S401, Obtain the normal operating temperature of the crystal oscillator and substitute it into the joint model of the comprehensive aging model and the temperature acceleration model to calculate the relative frequency deviation of the crystal oscillator at the normal operating temperature within the preset aging time. S402, Based on the calculated relative frequency deviation corresponding to different aging times, a long-term aging prediction curve of the relative frequency deviation changing with aging time is generated, and this prediction curve is the long-term frequency deviation prediction result.

[0050] Specifically, determining the normal operating temperature is a prerequisite for extrapolation; the preset long-term time is usually 10 to 20 years, covering the design life of most crystal oscillators; and generating a long-term aging prediction curve is to visually demonstrate the aging trend.

[0051] It should be noted that during the extrapolation process, it is necessary to ensure that the normal operating temperature is within the effective range of the temperature acceleration model. For example, the acceleration test temperature is 70℃~125℃, and the normal operating temperature is 25℃. It is necessary to ensure that 25℃ is within the temperature range that the model can extrapolate, and to avoid extrapolating to a temperature that exceeds the model's adaptation range.

[0052] For example, extrapolating to the normal operating temperature of 25°C, predicting the aging behavior over the next 20 years, such as... Figure 2As shown, the long-term aging prediction curve under 25℃ conditions is displayed. Assuming that the frequency tolerance limit of this crystal oscillator model is ±5ppm, according to the predicted aging curve, the time from the start of use to the frequency deviation exceeding -5ppm is calculated to be 8.5 years. Therefore, the warranty period of this product is set at 8 years. The average aging rate of the product from the 1st to the 5th year is calculated to be -0.82ppm / year, and the average aging rate from the 5th to the 10th year is -0.43ppm / year, indicating that the product stability gradually improves over time.

[0053] It should be noted that the data involved in this application is only used as an example to illustrate the solution, and this application does not impose any limitations.

[0054] Furthermore, this application provides a crystal oscillator aging prediction system, the system comprising: The first model 501 is used to build a comprehensive aging model based on the acquired aging test dataset; The second building module 502 is used to generate an initial joint model by combining the introduced improved temperature acceleration model with the comprehensive aging model; The parameter calculation module 503 is used to optimize the joint parameters of the initial joint model based on the aging test dataset and the parameter optimization algorithm, and generate joint optimized parameters. Optimization module 504 is used to substitute the joint optimization parameters into the initial joint model to determine the joint model; The prediction module 505 is used to input the acquired aging data into the joint mode to predict the long-term frequency deviation prediction result.

[0055] From a hardware perspective, in order to provide an embodiment of the electronic device for implementing all or part of the crystal oscillator aging prediction method, the electronic device specifically includes the following: The device comprises a processor, memory, a communications interface, and a bus; wherein the processor, memory, and communications interface communicate with each other via the bus; the communications interface is used to realize information transmission between servers, devices, distributed message middleware cluster devices, various databases, and user terminals, etc.; the electronic device can be a desktop computer, tablet computer, or mobile terminal, etc., and this embodiment is not limited to these. In this embodiment, the electronic device can be implemented with reference to the embodiments of the aging prediction method and the aging prediction system in the embodiments, the contents of which are incorporated herein, and repeated details will not be described again.

[0056] Embodiments of the present invention also provide a computer-readable storage medium capable of implementing all steps of the aging prediction method in the above embodiments, wherein the execution subject can be a server. The computer-readable storage medium stores a computer program that, when executed by a processor, implements all steps of the aging prediction method in the above embodiments.

[0057] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0058] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (devices), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0059] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0060] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0061] Specific embodiments have been used to illustrate the principles and implementation methods of this invention. The descriptions of the embodiments above are only for the purpose of helping to understand the method and core ideas of this invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this invention. Therefore, the content of this specification should not be construed as a limitation of this invention.

Claims

1. A method for predicting the aging of a crystal oscillator, characterized in that, include: A comprehensive aging model is established based on the acquired aging test dataset; An initial joint model is generated by combining an improved temperature acceleration model with a comprehensive aging model. The joint parameters of the initial joint model are optimized based on the aging test dataset and the parameter optimization algorithm to generate joint optimized parameters; Substitute the joint optimization parameters into the initial joint model to determine the joint model; By substituting the acquired aging data into the joint model, the long-term frequency deviation prediction results are obtained.

2. The crystal oscillator aging prediction method according to claim 1, characterized in that, The method further includes: The time from the start of use of the crystal oscillator to its frequency drift exceeding the frequency tolerance limit is calculated by combining the preset frequency tolerance limit of the crystal oscillator with the long-term frequency deviation prediction results.

3. The method for predicting the aging of a crystal oscillator according to claim 1, characterized in that, The method further includes: Based on the long-term frequency deviation prediction results and at least one preset specific time interval, the average aging rate of the crystal oscillator in each interval is calculated. The long-term stability of the crystal oscillator is determined based on the average aging rate.

4. The method for predicting the aging of a crystal oscillator according to claim 1, characterized in that, The establishment of a comprehensive aging model based on the acquired aging test dataset includes: The comprehensive aging model is determined based on the first aging mechanism parameters, the first time, the second aging mechanism parameters, the second time, the linear drift coefficient, the aging time, the initial frequency deviation, and the random fluctuation term.

5. The method for predicting the aging of a crystal oscillator according to claim 4, characterized in that, The methods for obtaining the initial frequency deviation include: The initial temperature of each temperature group sample is obtained, and the functional relationship between the initial frequency deviation and the initial temperature is established by linear fitting or polynomial fitting to determine the initial frequency deviation.

6. The method for predicting the aging of a crystal oscillator according to claim 1, characterized in that, The step of optimizing the joint parameters of the initial joint model based on the aging test dataset and the parameter optimization algorithm to generate joint optimized parameters includes: The joint parameters of the initial joint model are initially estimated using the least squares method to obtain the initial values ​​of the initial joint model parameters; The initial values ​​are iteratively adjusted using the simplex optimization algorithm until the numerical value of the constructed objective function converges to a preset threshold range, thereby generating joint optimization parameters.

7. The method for predicting the aging of a crystal oscillator according to claim 1, characterized in that, The step of substituting the acquired aging data into the joint model to predict the long-term frequency deviation prediction result includes: The normal operating temperature of the crystal oscillator is obtained and substituted into the joint model of the comprehensive aging model and the temperature acceleration model to calculate the relative frequency deviation of the crystal oscillator at the normal operating temperature within the preset aging time. Based on the calculated relative frequency deviations corresponding to different aging times, a long-term aging prediction curve is generated to show how the relative frequency deviation changes with aging time. This prediction curve is the long-term frequency deviation prediction result.

8. A crystal oscillator aging prediction system, characterized in that, The system includes: The first building module is used to establish a comprehensive aging model based on the acquired aging test dataset; The second building block is used to generate an initial joint model by combining the improved temperature acceleration model with the integrated aging model. The parameter calculation module is used to optimize the joint parameters of the initial joint model based on the aging test dataset and the parameter optimization algorithm, and generate joint optimized parameters. The optimization module is used to substitute the joint optimization parameters into the initial joint model to determine the joint model; The prediction module is used to input the acquired aging data into the joint model to predict the long-term frequency deviation.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images