Product reliability evaluation method and system based on improved degradation distribution method

Abstract

The present invention discloses a product reliability assessment method and system based on an improved degradation distribution method, comprising: selecting a performance degradation amount, obtaining different product usage time / mileage and corresponding performance degradation data through testing; selecting an appropriate distribution, fitting the corresponding distribution parameters and mean values ​​with the variation trend of usage time / mileage, calculating the slope parameter and intercept parameter of the distribution parameter and mean fitting equation to obtain a complete fitting equation; introducing the slope parameter of the mean fitting equation and the distribution parameter fitting equation into the improved reliability calculation model to obtain a reliability calculation model that adapts to the current performance degradation variation trend, thereby performing product reliability assessment. By improving the reliability calculation model, the present invention can automatically select the corresponding reliability calculation model based on the calculation conditions, avoiding the situation where manual model selection errors occur.

Description

Technical Field

[0001] The present invention relates to the technical field of product performance degradation prediction, and in particular to a product reliability evaluation method and system based on an improved degradation quantity distribution method. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] Degradation is a physical or chemical process that causes changes in product performance. This change develops gradually over time, ultimately leading to product failure. Degradation data, which shows the degradation of product performance parameters over time, such as crack length or wear, is called degradation data. Reliability analysis based on performance degradation data is an effective method for product reliability assessment.

[0004] The reliability assessment method based on the degradation distribution method regards the sample performance degradation as a random variable, and the degradation data of each sample at the same time as a set of realizations of the random variable, and describes the sample degradation process from the perspective of degradation distribution.

[0005] Currently, when using the degradation distribution method to analyze product performance degradation, the plotted mean values ​​of performance parameters often show either a monotonically increasing trend (e.g., a gradual increase in crack length) or a monotonically decreasing trend (e.g., a gradual decrease in component size after wear). Reliability calculations require professionals to select different calculation models based on these two scenarios. However, in actual calculations and software development, incorrect calculation model selection often occurs, leading to erroneous calculation results, biased analysis conclusions, and poor decision-making. Summary of the Invention

[0006] In order to solve the above problems, the present invention proposes a product reliability assessment method and system based on an improved degradation distribution method. By improving the reliability calculation model, the corresponding reliability calculation model can be automatically selected according to the calculation conditions, avoiding the situation where the model is selected incorrectly by humans.

[0007] According to a first aspect of an embodiment of the present invention, a product reliability assessment method based on an improved degradation distribution method is provided, comprising:

[0008] Select the performance degradation amount and obtain the product's different usage time / mileage and its corresponding performance degradation data through testing;

[0009] Based on the different usage time / mileage of the product and its corresponding performance degradation data, an appropriate distribution is selected, and the corresponding distribution parameters and mean values ​​are fitted to the changing trends of usage time / mileage. The slope and intercept parameters of the distribution parameter and mean fitting equations are calculated to obtain a complete fitting equation.

[0010] The slope parameter and distribution parameter fitting equation of the mean fitting equation are introduced into the improved reliability calculation model to obtain a reliability calculation model that adapts to the current performance degradation trend, and then the product reliability assessment is carried out;

[0011] The improved reliability calculation model includes a slope parameter of the mean value fitting equation and a Boolean operation to determine whether the slope parameter of the mean value fitting equation satisfies a set condition.

[0012] If the selected performance degradation amount obeys the normal distribution, the improved reliability calculation model is as follows:

[0013]

[0014] Among them, Boolean() is a Boolean operation, and the () in the Boolean operation is a judgment condition. When the judgment condition is met, Boolean() = 1, otherwise Boolean() = 0; k is the slope parameter of the mean fitting equation, μ x (t) is the fitting straight line equation of the mean, σ x (t) is the fitting straight line equation of the standard deviation, Φ is the cumulative distribution function of the normal distribution, and D is the threshold value of the failure criterion.

[0015] If the selected performance degradation value obeys the Weibull distribution, the improved reliability calculation model is as follows:

[0016]

[0017] Among them, Boolean() is a Boolean operation, and the () in the Boolean operation is a judgment condition. When the judgment condition is met, Boolean() = 1, otherwise Boolean() = 0; k is the slope parameter of the mean fitting equation, m x (t) is the fitting straight line equation of the shape parameter, η x (t) is the fitting straight line equation of the scale parameter; exp{} is the cumulative distribution function of the Weil distribution, and D is the threshold value of the failure criterion.

[0018] According to a second aspect of an embodiment of the present invention, a product reliability evaluation system based on an improved degradation distribution method is provided, comprising:

[0019] The data acquisition module is used to select the performance degradation amount and obtain the different usage time of the product and its corresponding performance degradation amount data through testing;

[0020] The parameter fitting module is used to select an appropriate distribution based on the different usage time / mileage of the product and its corresponding performance degradation data, fit the corresponding distribution parameters and mean values ​​with the changing trends of usage time / mileage, calculate the slope parameter and intercept parameter of the distribution parameter and mean fitting equation, and obtain a complete fitting equation;

[0021] The reliability assessment module is used to incorporate the slope parameter and distribution parameter fitting equation of the mean fitting equation into the improved reliability calculation model to obtain a reliability calculation model that adapts to the current performance degradation trend, thereby conducting product reliability assessment.

[0022] The improved reliability calculation model includes a slope parameter of the mean value fitting equation and a Boolean operation to determine whether the slope parameter of the mean value fitting equation satisfies a set condition.

[0023] According to a third aspect of an embodiment of the present invention, a terminal device is provided, which includes a processor and a memory, the processor being used to implement various instructions; the memory being used to store multiple instructions, and the instructions being suitable for being loaded by the processor and executing the above-mentioned product reliability assessment method based on the improved degradation distribution method.

[0024] Compared with the prior art, the present invention has the following beneficial effects:

[0025] (1) The present invention introduces the mean slope parameter k on the basis of the original reliability calculation model, and adds a Boolean operation on whether the mean slope parameter meets the set conditions. It can automatically select a reliability calculation model that adapts to the current performance degradation trend, avoiding the situation where the model is selected incorrectly by humans.

[0026] Advantages of additional aspects of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 Flowchart of a product reliability assessment method based on an improved degradation distribution method in an embodiment of the present invention;

[0028] Figure 2 A scatter plot showing the change in the mean and standard deviation of performance degradation of a component with mileage in an embodiment of the present invention;

[0029] Figure 3 This is a scatter plot of the mean value of performance degradation, shape parameters, and scale parameters of a component as a function of mileage in an embodiment of the present invention. DETAILED DESCRIPTION

[0030] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used in the present invention have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0031] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0032] In the absence of conflict, the embodiments of the present invention and the features thereof may be combined with each other.

[0033] Example 1

[0034] Theoretically, the distribution of the same degradation amount for the same model of product will not change due to different measurement times or mileages. The specific distribution of the degradation amount can be determined by hypothesis testing based on the measured data of the degradation performance parameters or by engineering experience.

[0035] However, the plotted trend of the mean value of the performance parameter at each time / mileage is usually monotonically increasing or decreasing. Of course, in rare cases, the trend may be constant, indicating no degradation trend, and is not considered in this case.

[0036] In the prior art, if the plotted trend of the mean value of the performance parameter shows a monotonically increasing trend, for example, if the crack length gradually increases, the failure criterion is x ≥ D, where D is the threshold of the failure criterion (for example, the component crack x should be less than 5mm, and when x ≥ 5mm, the component is judged to have failed, where D = 5). The reliability calculation model is as follows:

[0037] ① If the selected performance degradation quantity obeys the normal distribution, the reliability calculation model is:

[0038]

[0039] Among them, R(t) is the reliability calculation model, Φ is the normal distribution cumulative distribution function, μ x (t) is the fitting straight line equation of the mean, σ x (t) is the equation of the fitted line with the standard deviation.

[0040] ② If the selected performance degradation quantity obeys the Weil distribution, the reliability calculation model is:

[0041]

[0042] Among them, R(t) is the reliability calculation model, m x (t) is the fitting straight line equation of the shape parameter, η x (t) is the fitting straight line equation of the scale parameter.

[0043] If the plotted trend of the mean value of the performance parameter shows a monotonically increasing trend, for example, the size of the component gradually decreases after wear, then the failure criterion is x≤D, where D is the threshold of the failure criterion (for example, the thickness x of the component after wear should be greater than 5mm. When x≤5mm, the component is judged to have worn to the limit failure, where D=5). The reliability calculation model is as follows:

[0044] ① If the selected performance degradation quantity obeys the normal distribution, the reliability calculation model is:

[0045]

[0046] Among them, R(t) is the reliability calculation model, Φ is the normal distribution cumulative distribution function, μ x (t) is the fitting straight line equation of the mean, σ x (t) is the equation of the fitted line with the standard deviation.

[0047] ② If the selected performance degradation quantity obeys the Weil distribution, the reliability calculation model is:

[0048]

[0049] Among them, R(t) is the reliability calculation model, m x (t) is the fitting straight line equation of the shape parameter, η x (t) is the fitting straight line equation of the scale parameter.

[0050] Therefore, when calculating reliability, it is necessary to select the corresponding reliability calculation model based on the trend of changes in the plotted points of the mean value of the performance parameters; if the model is selected incorrectly, the subsequent calculation results will be biased, affecting the decision-making.

[0051] Based on this, in one or more embodiments, a product reliability evaluation method based on an improved degradation distribution method is disclosed, combined with Figure 1 , specifically including the following process:

[0052] Step S101: Determine the failure threshold D, select the performance degradation amount, and obtain data on different product usage times / mileages and their corresponding performance degradation amounts through testing;

[0053] Specifically, the failure threshold D and the failure interval can generally be obtained from product technical documents or engineering experience, or can be set manually.

[0054] The performance degradation amount is selected according to actual needs, for example, the crack length or component wear amount can be selected.

[0055] A set number j of product samples are selected, and the performance degradation value corresponding to each product sample at time / mileage t is obtained, as shown in Table 1.

[0056] Table 1 Record of performance parameters of each sample of a test product as the mileage changes

[0057] Mileage\Sample Sample 1 Sample 2 …… Sample j <![CDATA[t1]]> <![CDATA[x 11 ]]> <![CDATA[x 12 ]]> …… <![CDATA[x 1j ]]> <![CDATA[t2]]> <![CDATA[x 21 ]]> <![CDATA[x 22 ]]> …… <![CDATA[x 2j ]]> …… …… …… …… …… <![CDATA[t i ]]> <![CDATA[x i1 ]]> <![CDATA[x i2 ]]> …… <![CDATA[x ij ]]>

[0058] Generally speaking, data can be obtained from product performance tests or regular inspections during product maintenance. The more samples and the more dimensional data, the more reliable the results.

[0059] Then, data cleaning is performed. The calculation model of this embodiment only supports analysis of numerical values ​​of equally spaced dimension measurements under identifiable samples of the same type of product, the same performance parameters, and the same failure threshold conditions. If the test samples are of different types and differ greatly, individual samples cannot be distinguished, performance parameters are inconsistent, failure threshold standards are not unified, measurement intervals are inconsistent, or measurement intervals and measurement values ​​are recorded as non-numerical types, the data needs to be corrected or eliminated.

[0060] After data cleaning is completed, the performance degradation of each sample at each measurement time / mileage t is calculated based on the recorded performance degradation data. i The mean μ i (i=1,2,…), that is, the mean values ​​of the performance degradation data at different times / mileages and their corresponding values ​​are obtained.

[0061] Step S102: performing parameter fitting based on different usage time / mileage of the product and the corresponding performance degradation data mean values ​​to obtain the slope parameter and intercept parameter of the fitting equation;

[0062] In this embodiment, the equation for fitting can be selected from the following equations as needed:

[0063] μ x (t) = b + kt

[0064] Or, ln(μ x (t))=b+kt

[0065] Or, μ x (t) = b + kln(t)

[0066] Among them, μ x (t) represents the mean value of performance degradation, x is the performance degradation, t is the running time / mileage, b is the intercept parameter, and k is the slope parameter of the fitting equation.

[0067] This embodiment substitutes the obtained sample mean of performance degradation and the corresponding measured time / mileage data into one of three fitting equations; uses the least squares method or maximum likelihood estimation method to calculate the slope parameter k and intercept parameter b of the fitting equation for the sample mean, and obtains an equation for how the sample mean changes with time / mileage.

[0068] Generally speaking, the equation for how the sample mean changes with mileage is a monotonic function. In this embodiment, the slope parameter k is used as a judgment parameter of the improved reliability calculation model.

[0069] In addition, other fitting parameters need to be calculated:

[0070] Assuming that the performance degradation follows a normal distribution, the standard deviation of the sample performance degradation and its fitting equation parameters need to be calculated based on the performance degradation data to obtain the equation σ for the sample standard deviation changing with time / mileage. x (t).

[0071] If it is assumed that the performance degradation follows the Weibull distribution, it is necessary to calculate the fitting model of the shape and scale parameters based on the performance degradation data and calculate its parameters to obtain the equation m for the change of shape and scale parameters with time / mileage x (t) and η x (t). The method for calculating the parameters of the fitting equation is the same as the method for calculating the parameters of the sample mean model.

[0072] Step S103: Substituting the calculated slope parameter into the improved reliability calculation model to obtain a reliability calculation model that adapts to the current performance degradation trend;

[0073] Among them, the improved reliability calculation model adds a slope parameter and adds a Boolean operation to check whether the slope parameter meets the set conditions.

[0074] In this embodiment, the failure threshold D and the sample mean μ x The parameter k of (t), and the equation μ for how the sample mean and sample standard deviation change with time / mileage x (t) and σ x (t) (normal distribution), or, the equation of shape and scale as a function of time / mileage m x (t) and η x(t) (Weibull distribution), and is substituted into the improved performance reliability model to obtain the function R′(t) of the product reliability changing with time / mileage without manual selection.

[0075] Specifically, the improved reliability calculation model is as follows:

[0076] If the selected performance degradation amount obeys the normal distribution, the improved reliability calculation model is as follows:

[0077]

[0078] Among them, Boolean() is a Boolean operation, and the () in the Boolean operation is a judgment condition. When the judgment condition is met, Boolean() = 1, otherwise Boolean() = 0; k is the mean slope parameter, μ x (t) is the fitting straight line equation of the mean, σ x (t) is the fitting straight line equation of the standard deviation; Φ is the cumulative distribution function of the normal distribution, and D is the threshold value of the failure criterion.

[0079] If the selected performance degradation follows the Weibull distribution, the improved reliability calculation model is as follows:

[0080]

[0081] Among them, Boolean() is a Boolean operation, and the () in the Boolean operation is a judgment condition. When the judgment condition is met, Boolean() = 1, otherwise Boolean() = 0; k is the slope parameter of the mean fitting equation, m x (t) is the fitting straight line equation of the shape parameter, η x (t) is the fitting straight line equation of the scale parameter; exp is the cumulative distribution function of the Weil distribution, and D is the threshold value of the failure criterion.

[0082] The improved reliability calculation model in this embodiment can directly obtain a reliability calculation model that adapts to the current performance degradation trend by introducing the slope parameter of the mean fitting equation, without manual selection. Taking the normal distribution as an example, assuming that when k = -2, that is, the judgment condition is met, then Boolean = 1, and the reliability function formula becomes:

[0083]

[0084] This result is consistent with the model when the original calculation model decreases monotonically.

[0085] The Weibull distribution can be obtained similarly:

[0086]

[0087] This result is also consistent with the original calculation model.

[0088] Assume that when k = 2, that is, the judgment condition is not met, then Boolean = 0, and the reliability function formula becomes:

[0089]

[0090] This result is consistent with the original calculation model when it increases monotonically.

[0091] The Weibull distribution can be obtained similarly:

[0092]

[0093] This result is also consistent with the original calculation model.

[0094] The above verification process also illustrates the effectiveness and accuracy of the improved model in this embodiment.

[0095] Step S104: using the obtained reliability calculation model to perform product reliability assessment.

[0096] The reliability calculation model of this embodiment is a function model in which reliability changes with time or mileage, that is, R(t), a function of R with respect to t; based on this function model, the reliability corresponding to any time or mileage can be calculated, or the corresponding time or mileage can be calculated through the reliability.

[0097] For example, in the improved reliability calculation model, by inputting time / mileage t and setting the failure criterion threshold D, the corresponding reliability value R can be calculated. R values ​​range from 0 to 1. A higher R value indicates a higher product reliability, meaning a greater probability of no failure. The resulting reliability value can also be compared with a user-defined reliability standard to determine whether the product meets the specified reliability standard at a certain time or mileage.

[0098] Alternatively, you can input a set reliability value standard into the improved reliability calculation model and calculate the corresponding time / mileage. For example, you can input a reliability standard of 0.9 into the model and calculate the corresponding time or mileage (assuming it is 100). This will show that under this model, when the product's usage time or mileage does not exceed 100, the product's reliability meets the standard.

[0099] Example verification

[0100] The verification process of the method of this embodiment is given below, which is as follows:

[0101] The collected and cleaned test data are shown in Table 2:

[0102] Table 2 Record of performance degradation of a component as a function of mileage

[0103]

[0104] The performance degradation amount is defined as x, and the failure threshold is D = 10. When x ≥ D, the component fails. The reliability calculation model of this parameter varying with mileage is calculated using the performance degradation amount distribution method.

[0105] (1) Assuming that the performance degradation amount follows a normal distribution

[0106] According to the performance degradation distribution method, assuming that the performance degradation follows a normal distribution, the mean and standard deviation of the degradation of each sample of the component at the corresponding mileage are calculated. The calculation results are shown in the following table:

[0107] Table 3

[0108] mileage Mean μ Standard deviation σ 250 0.462 0.140057 500 1.035 0.278361 750 1.674 0.390671 1000 2.178 0.49323 1250 2.72 0.641077 1500 3.298 0.73295 1750 3.786 0.874931 2000 4.326 1.075455 2250 4.81 1.179356 2500 5.363 1.289876 2750 5.872 1.381331 3000 6.401 1.541463 3250 6.908 1.680743 3500 7.413 1.868497 3750 7.917 1.940047 4000 8.534 2.071175

[0109] According to the calculation results, plot the points, observe the degradation law, and select the appropriate model for fitting, such as Figure 2 shown.

[0110] Then calculate the mean. The mean calculation results of this example data are as follows:

[0111] Table 4

[0112]

[0113]

[0114] Calculate the mean fitting parameters from Figure 2 It can be seen that the mean μ of the degradation amount shows a linear degradation trend, so the linear equation is selected:

[0115] μ x (t) = b + kt

[0116] Fitting is performed and the model parameters are calculated using the maximum likelihood method to obtain:

[0117] μ x (t) = 0.04235 + 0.002118t

[0118] Where, k = 0.002118.

[0119] Calculate other fitting parameters. Assuming that the performance degradation follows a normal distribution, in addition to calculating the fitting equation for the mean μ, it is also necessary to calculate the fitting equation for the standard deviation σ. The method is the same as that for the fitting equation for the mean μ, and we get:

[0120] σ x(t) = -0.00637 + 0.00052t

[0121] σ x The parameters in (t) do not participate in the calculation of this patent and are therefore ignored.

[0122] Determine the failure threshold D and interval. In this embodiment, the failure threshold D=10. When the performance degradation amount x≥10, the product fails, that is, when x<10, the product is reliable.

[0123] The failure threshold D, μ x (t), σ x (t) and k are directly imported into the improved reliability calculation model to obtain:

[0124]

[0125] Since k=0.002118>0, the condition in Boolean(0.002118<0) is not satisfied, so Boolean(0.002118<0)=0, from which we can get:

[0126]

[0127] The results show that the results are consistent with the original calculation model selected when the failure criterion is x≥D. The form is consistent, but the formula selection step is omitted, which reduces human intervention and realizes the control of the calculation model.

[0128] Assume that the user wants to know the reliability of the product when the mileage is 8000. Substituting t=8000 into R′(t), we can obtain R′(8000)=0.0463.

[0129] (2) Assuming that the performance degradation follows the Weibull distribution

[0130] According to the performance degradation distribution method, assuming that the performance degradation obeys the 2-parameter Weibull distribution, the mean μ, shape parameter m and scale parameter η of the degradation of each sample of the component at the corresponding mileage are calculated. The calculation results are shown in Table 5 below. The scatter plot of the parameters changing with mileage is shown in Figure 3 shown.

[0131] Table 5

[0132] mileage Mean μ Shape m Scale η 250 0.462 3.487 0.5137 500 1.035 4.226 1.141 750 1.674 4.728 1.83 1000 2.178 4.786 2.378 1250 2.72 4.586 2.978 1500 3.298 4.953 3.596 1750 3.786 4.764 4.139 2000 4.326 4.408 4.75 2250 4.81 4.399 5.278 2500 5.363 4.447 5.88 2750 5.872 4.578 6.429 3000 6.401 4.492 7.02 3250 6.908 4.492 7.58 3500 7.413 4.319 8.153 3750 7.917 4.41 8.691 4000 8.534 4.438 9.362

[0133] Then the mean calculation is performed. The mean calculation results of this example data are as follows:

[0134] Table 6

[0135] mileage Mean μ 250 0.462 500 1.035 750 1.674 1000 2.178 1250 2.72 1500 3.298 1750 3.786 2000 4.326 2250 4.81 2500 5.363 2750 5.872 3000 6.401 3250 6.908 3500 7.413 3750 7.917 4000 8.534

[0136] After getting the mean, perform mean fitting parameters. Figure 3 It can be seen that the mean μ of the degradation amount shows a linear degradation trend, so the linear equation is selected:

[0137] μ x (t) = b + kt

[0138] Fitting is performed. The model parameters are calculated using the maximum likelihood method and obtained as follows:

[0139] μ x (t) = 0.04235 + 0.002118t

[0140] Where, k = 0.002118.

[0141] Other fitting parameters need to be calculated. Assuming that the performance degradation follows a two-parameter Weibull distribution, the fitting equations for the shape parameter m and scale parameter η need to be calculated. The method is the same as that for the fitting equation for the mean μ, resulting in:

[0142] m x (t)=4.394+0.000035t

[0143] Similarly, we can get:

[0144] η x (t) = 0.03923 + 0.002326t

[0145] m x (t) and η x The parameters in (t) do not participate in the calculation of this patent and are therefore ignored.

[0146] Determine the failure threshold D and interval. In this embodiment, the failure threshold D=10. When the performance degradation amount x≥10, the product fails, that is, when x<10, the product is reliable.

[0147] Set the failure threshold D, m x (t), η x (t) and k are directly imported into the improved reliability calculation model in this patent to obtain:

[0148]

[0149] Since k=0.002118>0, the condition in Boolean(0.002118>0) is satisfied. Therefore,

[0150] Boolean(0.002118>0)=1, so we can get:

[0151]

[0152] The results show that the results are consistent with the original calculation model selected when the failure criterion is x≥D. The form is consistent, but the formula selection step is omitted, which reduces human intervention and realizes the control of the calculation model.

[0153] Assume that the user wants to know the reliability of the mileage at 8000. Substituting t=8000 into R′(t), we can obtain R′(8000)=0.0529, which is similar to the calculation result under normal distribution.

[0154] Example 2

[0155] In one or more embodiments, a product reliability assessment system based on an improved degradation distribution method is disclosed, comprising:

[0156] The data acquisition module is used to select the performance degradation amount and obtain the different usage time of the product and its corresponding performance degradation amount data through testing;

[0157] The parameter fitting module is used to select an appropriate distribution based on the different usage time / mileage of the product and its corresponding performance degradation data, fit the corresponding distribution parameters and mean values ​​with the changing trends of usage time / mileage, calculate the slope parameter and intercept parameter of the distribution parameter and mean fitting equation, and obtain a complete fitting equation;

[0158] The reliability assessment module is used to incorporate the slope parameter and distribution parameter fitting equation of the mean fitting equation into the improved reliability calculation model to obtain a reliability calculation model that adapts to the current performance degradation trend, thereby conducting product reliability assessment.

[0159] The improved reliability calculation model includes a slope parameter of the mean value fitting equation and a Boolean operation to determine whether the slope parameter of the mean value fitting equation satisfies a set condition.

[0160] As an optional implementation, when the reliability assessment module further performs product reliability assessment, the product operating time / mileage is input into the reliability calculation model to obtain the reliability value, and then determine the product reliability; or, the set product reliability value is input into the reliability calculation model to obtain the product operating time / mileage corresponding to the product reliability.

[0161] The specific implementation of each of the above modules has been described in Example 1 and will not be described in detail here.

[0162] Example 3

[0163] In one or more embodiments, a terminal device is disclosed, including a server. The server includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the product reliability assessment method based on the improved degradation distribution method described in Example 1 is implemented. For the sake of brevity, this description is omitted here.

[0164] It should be understood that in this embodiment, the processor may be a central processing unit (CPU), or may be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), off-the-shelf field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0165] The memory may include a read-only memory and a random access memory, and provides instructions and data to the processor. A portion of the memory may also include a non-volatile random access memory. For example, the memory may also store information about the device type.

[0166] During implementation, each step of the above method may be completed by an integrated logic circuit of hardware in a processor or by instructions in the form of software.

[0167] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A product reliability evaluation method based on an improved degradation distribution method, characterized in that: include: Select the performance degradation amount and obtain the product's different usage time / mileage and its corresponding performance degradation data through testing; Based on the different usage time / mileage of the product and its corresponding performance degradation data, an appropriate distribution is selected, and the corresponding distribution parameters and mean values ​​are fitted to the changing trends of usage time / mileage. The slope and intercept parameters of the distribution parameter and mean fitting equations are calculated to obtain a complete fitting equation. The slope parameter and distribution parameter fitting equation of the mean fitting equation are introduced into the improved reliability calculation model to obtain a reliability calculation model that adapts to the current performance degradation trend, and then the product reliability assessment is carried out; The improved reliability calculation model adds the slope parameter of the mean fitting equation and adds a Boolean operation to determine whether the slope parameter of the mean fitting equation meets the set conditions, specifically: If the selected performance degradation amount obeys the normal distribution, the improved reliability calculation model is as follows: Among them, Boolean() is a Boolean operation, and the () in the Boolean operation is a judgment condition. When the judgment condition is met, Boolean()=1, otherwise Boolean()=0; k is the slope parameter of the mean fitting equation, is the fitting straight line equation of the mean, is the fitting straight line equation of the standard deviation, Φ is the normal distribution cumulative distribution function, and D is the failure criterion; If the selected performance degradation follows the Weibull distribution, the improved reliability calculation model is as follows: is the fitting straight line equation of the shape parameter, is the fitting straight line equation of the scale parameter; {} is the cumulative distribution function of the Weil distribution.

2. The product reliability evaluation method based on the improved degradation distribution method according to claim 1, characterized in that: Through testing, we obtain data on the different usage times of the product and its corresponding performance degradation, specifically: A set number of product samples are selected, and the performance degradation data of each product sample at a set time is obtained. The performance degradation data of the product samples at each time are averaged as the average performance degradation data corresponding to each measurement time.

3. The product reliability evaluation method based on the improved degradation distribution method according to any one of claims 1 to 2, characterized in that: Based on the different usage time of the product and the corresponding performance degradation data mean, parameter fitting is performed. The specific fitting equation is: or, or, in, represents the mean value of performance degradation, x is the performance degradation, t is the running time / mileage, b is the intercept parameter, and k is the slope parameter of the fitting equation.

4. The product reliability evaluation method based on the improved degradation distribution method according to any one of claim 3, characterized in that: Estimate the mean using the least squares method or the maximum likelihood method The intercept parameter b and slope parameter k of .

5. The product reliability evaluation method based on the improved degradation distribution method according to claim 1, characterized in that: A reliability calculation model that adapts to the current performance degradation trend is obtained, and then product reliability assessment is performed, specifically: Inputting the product running time / mileage into the reliability calculation model to obtain a reliability value, thereby determining the product reliability; Alternatively, the set product reliability value is input into the reliability calculation model to obtain the product operating time / mileage corresponding to the product reliability.

6. A product reliability evaluation system based on an improved degradation distribution method, characterized in that: include: The data acquisition module is used to select the performance degradation amount and obtain the different usage time of the product and its corresponding performance degradation amount data through testing; The parameter fitting module is used to select an appropriate distribution based on the different usage time / mileage of the product and its corresponding performance degradation data, fit the corresponding distribution parameters and mean values ​​with the changing trends of usage time / mileage, calculate the slope parameter and intercept parameter of the distribution parameter and mean fitting equation, and obtain a complete fitting equation; The reliability assessment module is used to incorporate the slope parameter and distribution parameter fitting equation of the mean fitting equation into the improved reliability calculation model to obtain a reliability calculation model that adapts to the current performance degradation trend, thereby conducting product reliability assessment. The improved reliability calculation model adds the slope parameter of the mean fitting equation and adds a Boolean operation to determine whether the slope parameter of the mean fitting equation meets the set conditions, specifically: If the selected performance degradation amount obeys the normal distribution, the improved reliability calculation model is as follows: Among them, Boolean() is a Boolean operation, and the () in the Boolean operation is a judgment condition. When the judgment condition is met, Boolean()=1, otherwise Boolean()=0; k is the slope parameter of the mean fitting equation, is the fitting straight line equation of the mean, is the fitting straight line equation of the standard deviation, Φ is the normal distribution cumulative distribution function, and D is the failure criterion; If the selected performance degradation follows the Weibull distribution, the improved reliability calculation model is as follows: is the fitting straight line equation of the shape parameter, is the fitting straight line equation of the scale parameter; {} is the cumulative distribution function of the Weil distribution.

7. The product reliability evaluation system based on the improved degradation distribution method according to claim 6, characterized in that: When the reliability evaluation module performs reliability evaluation, the product operating time / mileage is input into the reliability calculation model to obtain a reliability value, thereby determining product reliability; Alternatively, the set product reliability value is input into the reliability calculation model to obtain the product operating time / mileage corresponding to the product reliability.

8. A terminal device comprising a processor and a memory, wherein the processor is used to implement various instructions; the memory is used to store multiple instructions, characterized in that: The instructions are suitable for being loaded by a processor and executing the product reliability assessment method based on the improved degradation distribution method according to any one of claims 1 to 5.

Images