Product information determination method, device, electronic device and medium

Abstract

The present disclosure relates to the technical field of product design, and in particular, provides a method, device, electronic device and medium for determining product information, the method comprising: obtaining test data of a sample product; constructing a first model matrix between the mean of a quality parameter and a product parameter based on the test data, and constructing a second model matrix between a preset quantile of the logarithm of life data and the product parameter; processing the first model matrix and the second model matrix based on a quasi-unrelated regression algorithm to obtain model regression coefficients of a first linear regression model and a second linear regression model, and substituting the model regression coefficients into the first linear regression model and the second linear regression model to obtain a first target linear regression model and a second target linear regression model; optimizing the first target linear regression model and the second target linear regression model simultaneously based on a preset optimization algorithm to obtain target parameter values ​​of the product parameters. The present disclosure can improve the accuracy and reliability of the determined product information.

Description

Technical Field

[0001] The present disclosure relates to the technical field of product design, and in particular to a method, device, electronic device and medium for determining product information. Background Art

[0002] With the continuous advancement of industrial technology and the intensification of market competition, in order to improve customer satisfaction and reduce product warranty costs, companies are paying more and more attention to improving product quality. Usually, the quality of a product can be measured by combining product quality parameters and life data.

[0003] In the related technology, the quality parameters and life data of products can be measured through experimental design in industrial experiments, and based on the measured quality parameters and life data, product information such as product parameter values ​​that affect product quality and life data can be determined to improve product quality.

[0004] However, the product information determination scheme provided in the related art has the problem that the experimental design is not a completely randomized experiment, and the correlation between quality parameters and life data is not considered. As a result, the product information determined in the product quality optimization process will not meet the actual product quality optimization needs, resulting in poor reliability of the obtained product information. Summary of the invention

[0005] The present disclosure is proposed in view of the above problems. The present disclosure provides a method, device, electronic device and medium for determining product information, which can improve the accuracy and reliability of the determined product information.

[0006] According to one aspect of the present disclosure, a method for determining product information is provided, comprising:

[0007] Acquire test data of a sample product, wherein the test data includes a plurality of sample quality parameter values, a plurality of sample life data and sample parameter values ​​of a product parameter of the sample product;

[0008] Based on the test data, construct a first model matrix of a first linear regression model between a mean value of a quality parameter and the product parameter, and construct a second model matrix of a second linear regression model between a preset quantile of the logarithm of the life data and the product parameter;

[0009] Processing the first model matrix and the second model matrix based on a quasi-unrelated regression algorithm to obtain model regression coefficients of the first linear regression model and the second linear regression model, and substituting the model regression coefficients into the first linear regression model and the second linear regression model to obtain a first target linear regression model and a second target linear regression model;

[0010] Based on a preset optimization algorithm, the first target linear regression model and the second target linear regression model are optimized simultaneously within the parameter value range of the product parameter to obtain the target parameter value of the product parameter.

[0011] According to another aspect of the present disclosure, there is provided a device for determining product information, comprising:

[0012] An acquisition module is configured to acquire test data of a sample product, wherein the test data includes a plurality of sample quality parameter values, a plurality of sample life data and sample parameter values ​​of a product parameter of the sample product;

[0013] A construction module is configured to construct a first model matrix of a first linear regression model between a mean value of a quality parameter and the product parameter based on the test data, and to construct a second model matrix of a second linear regression model between a preset quantile of a logarithm of life data and the product parameter;

[0014] a determination module configured to process the first model matrix and the second model matrix based on a quasi-unrelated regression algorithm to obtain model regression coefficients of the first linear regression model and the second linear regression model, and substitute the model regression coefficients into the first linear regression model and the second linear regression model to obtain a first target linear regression model and a second target linear regression model;

[0015] The optimization module is configured to simultaneously optimize the first target linear regression model and the second target linear regression model within the parameter value range of the product parameter based on a preset optimization algorithm to obtain a target parameter value of the product parameter.

[0016] According to another aspect of the present disclosure, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the above-mentioned product information determination method.

[0017] According to another aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned product information determination method is implemented.

[0018] The product information determination method, device, electronic device and medium provided by the embodiments of the present disclosure, on the one hand, after obtaining the test data, can construct a linear regression model between the preset quantiles describing the mean of the quality parameter and the logarithm of the life data and the product parameters, respectively, so as to eliminate the correlation between the quality parameter values ​​of multiple samples obtained on the same test bench, and eliminate the correlation between the life data of multiple samples obtained on the same test bench, solve the sub-sampling problem caused by non-completely randomized experiments, and improve the matching degree between the determined product information and the actual product quality optimization needs; on the other hand, the model regression coefficients of the first linear regression model used to determine the product quality parameter value and the second linear regression model used to determine the life data can be estimated while considering the correlation between the quality parameters and the life data, so as to make the product information determined based on the first target linear regression model and the second target linear regression model finally obtained more reliable.

[0019] It is to be understood that both the foregoing general description and the following detailed description are exemplary, and are intended to provide further explanation of the technology as claimed. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The above and other purposes, features and advantages of the present disclosure will become more apparent by describing the embodiments of the present disclosure in more detail in conjunction with the accompanying drawings. The accompanying drawings are used to provide a further understanding of the embodiments of the present disclosure and constitute a part of the specification. Together with the embodiments of the present disclosure, they are used to explain the present disclosure and do not constitute a limitation of the present disclosure. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0021] Figure 1 It is a schematic diagram of an implementation scenario of a product information determination solution according to an embodiment of the present disclosure.

[0022] Figure 2 is a flowchart illustrating a method for determining product information according to an embodiment of the present disclosure.

[0023] Figure 3 is a schematic diagram of fitting diagnosis of the mean value of the quality parameter according to an embodiment of the present disclosure.

[0024] Figure 4 is a diagram illustrating the diagnostic fit of the quantiles of the logarithm of life data according to an embodiment of the present disclosure.

[0025] Figure 5 It is a schematic diagram illustrating the results of obtaining target parameter values ​​of thickness, adjustment time and width based on a genetic algorithm in an embodiment of the present disclosure.

[0026] Figure 6 is a block diagram illustrating a product information determination device according to an embodiment of the present disclosure.

[0027] Figure 7 is a schematic diagram illustrating a computer program product according to an embodiment of the present disclosure.

[0028] Figure 8 is a hardware block diagram illustrating an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical solution and advantages of the present disclosure more obvious, the exemplary embodiments according to the present disclosure will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all the embodiments of the present disclosure, and it should be understood that the present disclosure is not limited to the exemplary embodiments described here.

[0030] For industrial products, the quality parameters and life data of the products can usually be used to measure the quality of the products, and the quality parameters and life data of the products are usually affected by the product parameters. Therefore, determining the appropriate product parameter values ​​to improve the quality of the products has become the key point of quality improvement of industrial products. Among them, product parameters are usually regular parameters of the products, such as length, thickness and / or component concentration in the material.

[0031] In the related art, Design of Experiment is widely used to determine the parameter values ​​of products. In the process of improving the quality of products through experimental design, randomization is one of the basic principles of experimental design. However, due to the selection of some experimental schemes in the experimental design (for example, sample processing schemes such as subsampling, block or clustering), the experimental process presents a non-completely randomized design. For example, engineers may apply the treatment combination to each test bench instead of a single test unit on each test bench, so that the experiment contains subsampling, resulting in the experimental process being non-completely randomized, which will make the product information determined in the process of product quality optimization not meet the actual product quality optimization needs. In addition, there is usually a correlation between product quality parameters and life data. However, in the process of improving the design of product quality, the related technology ignores the characteristics of the correlation between quality parameters and life data, resulting in poor reliability of the obtained product information.

[0032] In order to solve the above problems, the present disclosure provides a product information determination solution, such as Figure 1 As shown, Figure 1 FIG. 1 is a schematic diagram showing an implementation scenario of a product information determination solution provided by an exemplary embodiment of the present disclosure. Figure 1 As shown, the implementation scenario 100 includes a test bench 101 and a product information determination device 102, wherein the test bench 101 is used to measure test data of a sample product, and the product information determination device 102 can be a computer, a notebook or a server, etc.

[0033] A communication link may be established between the test bench 101 and the product information determination device 102 to implement the product information determination solution provided in the embodiment of the present disclosure.

[0034] Figure 2 A flowchart of a product information determination method according to an exemplary embodiment of the present disclosure is shown. The method can be applied to a product information determination device, such as Figure 2 As shown, the method of the embodiment of the present disclosure may include:

[0035] Step S201, obtaining test data of sample products;

[0036] The test data includes a plurality of sample quality parameter values ​​of the sample product, a plurality of sample life data and sample parameter values ​​of the product parameters;

[0037] Step S202, based on the test data, construct a first model matrix of a first linear regression model between the mean of the quality parameter and the product parameter, and construct a second model matrix of a second linear regression model between the preset quantile of the logarithm of the life data and the product parameter;

[0038] Step S203, processing the first model matrix and the second model matrix based on a quasi-unrelated regression algorithm to obtain model regression coefficients of the first linear regression model and the second linear regression model, and substituting the model regression coefficients into the first linear regression model and the second linear regression model to obtain a first target linear regression model and a second target linear regression model;

[0039] Step S204, optimizing the first target linear regression model and the second target linear regression model simultaneously within the parameter value range of the product parameter based on a preset optimization algorithm to obtain a target parameter value of the product parameter;

[0040] In summary, the product information determination method provided by the embodiment of the present disclosure, on the one hand, after obtaining the test data, can construct a linear regression model between the preset quantiles describing the mean of the quality parameter and the logarithm of the life data and the product parameters, respectively, which can eliminate the correlation between the quality parameter values ​​of multiple samples obtained on the same test bench, and eliminate the correlation between the life data of multiple samples obtained on the same test bench, solve the sub-sampling problem caused by non-completely randomized experiments, and improve the matching degree between the determined product information and the actual product quality optimization needs; on the other hand, considering the correlation between the quality parameters and the life data, the model regression coefficients of the first linear regression model used to determine the product quality parameter value and the second linear regression model used to determine the life data can be estimated, so that the product information determined based on the first target linear regression model and the second target linear regression model finally obtained are more reliable.

[0041] The following Figure 2The specific implementation methods of each step in the embodiment shown are described in detail:

[0042] In step S201 , the product information determination device may obtain test data of a sample product.

[0043] In the embodiments of the present disclosure, the sample product is a product that needs quality improvement, and the test data includes multiple sample quality parameter values ​​of the sample product, multiple sample life data, and sample parameter values ​​of product parameters, wherein the life data is used to characterize the length of time the product is normally used, and is often used to characterize product reliability, the quality parameters are used to characterize the extent to which the product meets the requirements, and the product parameters are used to characterize the attribute characteristics of the product. The quality parameters and product parameters can be determined based on the type of product, and the embodiments of the present disclosure are not limited to this. For example, if the product is an electronic product, the quality parameters may be the degree of wear, the degree of battery loss and / or the standby time, etc.; the product parameters may be the product size (such as length, width and thickness, etc.), memory size and system version, etc.

[0044] It should be noted that in the embodiments of the present disclosure, testers can obtain test data through multiple test benches, wherein each test bench has multiple observation units, and each observation unit corresponds to a sample product. During the testing process, testers can apply test conditions on each test bench to test and obtain test data, and transmit the monitored test data to the product information determination device. The test conditions can be determined based on actual needs, and the embodiments of the present disclosure do not limit this. For example, if the sample product is a mobile phone, the test conditions may be conditions such as continuously keeping the screen on, high temperature or low temperature, etc.

[0045] In an optional implementation, the process of the product information determination device acquiring the test data of the sample product may include: obtaining the test data of the sample product in response to receiving the test data of the sample product sent by the test bench.

[0046] In step S202, the product information determination device can construct a first model matrix of a first linear regression model between the mean of quality parameters and product parameters based on the test data, and construct a second model matrix of a second linear regression model between preset quantiles of the logarithm of life data and product parameters.

[0047] In the embodiment of the present disclosure, the first linear regression model is a relationship model between the mean of the quality parameters of the product and the product parameters, and the second linear regression model is a relationship model between the preset quantiles of the logarithm of the product's life data and the product parameters, wherein the preset quantiles can be determined based on actual needs, and the embodiment of the present disclosure does not limit this. For example, the preset quantiles can be percentiles.

[0048] Among them, the first linear regression model is:

[0049] ;(Formula 1)

[0050] In formula 1, is the mean value of the quality parameter, For the k The parameter value of each product parameter, is the random error of the quality parameter value, Model regression coefficients for the first linear regression model.

[0051] The second linear regression model is:

[0052] ;(Formula 2)

[0053] In formula 2, is the logarithm of life data p Quantile, is the parameter value of the sth product parameter, is the random error of the life data, The model regression coefficients of the second linear regression model, where p The quantile can be determined based on actual needs, and the embodiments of the present disclosure do not limit this. For example, p Quantiles can be percentiles.

[0054] In an optional embodiment, the multiple sample quality parameter values ​​are multiple sample quality parameter values ​​obtained respectively on multiple test benches, and the process of the product information determination device constructing the first model matrix of the first linear regression model between the quality parameter mean and the product parameter based on the test data may include: determining the mean of the multiple sample quality parameter values ​​associated with each test bench to obtain the sample quality parameter mean; then, substituting the sample parameter values ​​of the product parameters and the sample quality parameter mean associated with each test bench into the first linear regression model to obtain the first linear regression model associated with each test bench; further, combining the first linear regression models associated with each test bench to obtain the first model matrix of the first linear regression model. The first model matrix of the first linear regression model can be constructed by using the multiple sample quality parameter values ​​respectively associated with the multiple test benches and the sample parameter values ​​of the product parameters, so as to improve the precision or accuracy of the model regression coefficients of the determined first linear regression model.

[0055] For example, if there are m test units (i.e., test benches), each test unit has n observation units, the sample data obtained by the product information determination device may include the sample quality parameter value ,in, represents the sample quality parameter value obtained in the jth observation unit of the i-th test unit, where , ; Then, the product information determination device can determine the mean of multiple sample quality parameter values ​​associated with each test bench, and the sample quality parameter mean is:

[0056] ;(Formula 3)

[0057] In formula 3, For the The mean value of sample quality parameters associated with each test bench;

[0058] Furthermore, the product information determination device may substitute the sample parameter values ​​of the product parameters and the mean value of the sample quality parameters associated with each test bench into the first linear regression model to obtain the first linear regression model associated with each test bench:

[0059] ;(Formula 4)

[0060] In formula 4, For the The vector form of the mean value of the sample quality parameter associated with the test bench, For the The first sample product on the test bench k Sample parameter values ​​for product parameters, For the The random errors in the quality parameter values ​​in the first linear regression model associated with the test rigs.

[0061] Finally, the product information determination device can combine the first linear regression model associated with each test bench to obtain a first model matrix of the first linear regression model:

[0062] ;(Formula 5)

[0063] In formula 5, is the sample quality parameter mean matrix, is the parameter value matrix of product parameters, is the model regression coefficient matrix of the first linear regression model, is the random error matrix of the quality parameter values.

[0064] In an optional embodiment, the plurality of sample life data includes a plurality of sample life data respectively obtained from a plurality of test benches, the plurality of sample life data obeys a Weibull distribution, and the process of constructing a second model matrix of a second linear regression model between preset quantiles of the logarithm of the life data and product parameters by the product information determination device based on the test data may include: determining the preset quantiles of the logarithm of the sample life data associated with each test bench based on the plurality of sample life data respectively obtained from the plurality of test benches; then, substituting the sample parameter values ​​of the product parameters and the preset quantiles of the logarithm of the sample life data associated with each test bench into the second linear regression model to obtain the second linear regression model associated with each test bench; further, combining the second linear regression models associated with each test bench to obtain the second model matrix of the second linear regression model. The second model matrix of the second linear regression model can be constructed by using the plurality of sample life data respectively associated with the plurality of test benches and the sample parameter values ​​of the product parameters, so as to improve the reliability of the model regression coefficient of the determined second linear regression model.

[0065] It should be noted that in the embodiments of the present disclosure, the test bench is a device for collecting life data, and different test benches can simulate different test conditions, such as environmental factors such as temperature and humidity. Failure mechanism refers to the failure mode and cause of a product during use. Different failure mechanisms may include material fatigue, wear, corrosion, etc. During the collection of life data, it is ensured that the failure mechanism of the product on the test bench is consistent. It can be assumed that the life data obtained on different test benches have the same statistical characteristics, which is helpful for data comparison and analysis, thereby improving the effectiveness of the test.

[0066] In an optional embodiment, the process of determining the preset quantile of the logarithm of the sample life data associated with each test bench based on the multiple sample life data respectively obtained from the multiple test benches by the product information determination device may include: determining the sample minimum extreme value distribution function, cumulative sum distribution function and probability density function of the logarithm of the sample life data based on the multiple sample life data respectively obtained from the multiple test benches; then, based on the cumulative sum distribution function and the probability density function, the sample location parameter value and the sample scale parameter value associated with each test bench, wherein the sample scale parameter value associated with each test bench is the same; further, substituting the standard distribution function of the sample minimum extreme value distribution function, and the location parameter value and the sample scale parameter value associated with each test bench into the preset quantile of the logarithm of the life data, and the relationship equation between the standard distribution function of the minimum extreme value distribution function, the location parameter and the scale parameter, to obtain the preset quantile of the logarithm of the sample life data associated with each test bench. Since the failure mechanism of the products on the test benches is consistent, the sample scale parameter value associated with each test bench is also the same, thereby improving the effectiveness of the preset quantile of the logarithm of the sample life data associated with each test bench.

[0067] It should be noted that, in the embodiment of the present disclosure, if there are m test units (i.e., test benches), and each test unit has n observation units, then the sample life data obtained from the jth observation unit in the i-th test bench is t ij , then since the life data of multiple samples obey the Weibull distribution, that is , is the scale parameter of the Weibull distribution for life data, is the shape parameter of the Weibull distribution for the life data, which also means that the logarithm of the sample life data follows the smallest extreme value distribution, where the logarithm of the sample life data is y=log(t), and the smallest extreme value distribution is , is the location parameter of the smallest extreme value distribution for the logarithm of the life data, is the scale parameter of the smallest extreme value distribution for the logarithm of life data.

[0068] The cumulative sum distribution function is:

[0069] ;(Formula 6)

[0070] In Formula 6, For the The first test bench j The logarithm of the sample life data obtained from observation units, , For the i The logarithm of the life data associated with the test benches follows the location parameter of the smallest extreme value distribution, For the i The logarithm of the life data associated with the test rigs follows the scale parameter of the smallest extreme value distribution.

[0071] The probability density function is:

[0072] ;(Formula 7)

[0073] It should also be noted that, in the embodiment of the present disclosure, the product information determination device is based on the cumulative sum distribution function and the probability density function, and the process of determining the sample position parameter values ​​and the sample scale parameter values ​​associated with each test bench can be implemented based on maximum likelihood estimation; wherein the sample position parameter values ​​determined based on the maximum likelihood estimation are the sample position parameter values ​​whose logarithm of the life data associated with each test bench obeys the minimum extreme value distribution; the sample scale parameter values ​​determined based on the maximum likelihood estimation are the sample scale parameter values ​​whose logarithm of the life data associated with each test bench obeys the minimum extreme value distribution.

[0074] Among them, when the preset quantile is p In the case of quantiles, the relationship between the preset quantile of the logarithm of life data and the standard distribution function of the smallest extreme value distribution function, the location parameter and the scale parameter is:

[0075] ;(Formula 8)

[0076] In formula 8, For the i The logarithm of the sample life data associated with the test benches p Quantile, is the standard distribution function with the smallest extreme value distribution function.

[0077] It can be understood that after determining the preset quantile of the logarithm of the sample life data associated with each test bench, the product information determination device substitutes the sample parameter value of the product parameter and the preset quantile of the logarithm of the sample life data associated with each test bench into the second linear regression model, and the second linear regression model associated with each test bench is obtained as follows:

[0078] ;(Formula 9)

[0079] In formula 9, For the i The first sample product on the test bench s Sample parameter values ​​for product parameters, For the i The random errors of the life data in the second linear regression model associated with the test rigs.

[0080] Furthermore, the product information determines a second linear regression model associated with the equipment combination and each test bench, and the second model matrix of the second linear regression model is obtained as follows:

[0081] ;(Formula 10)

[0082] In formula 10, is the random error matrix of the sample life data to the quality parameter value p The quantile matrix, is the parameter value matrix of product parameters, is the model regression coefficient matrix of the second linear regression model, is the random error matrix of the life data.

[0083] It is understandable that in the embodiment of the present disclosure, if it is necessary to determine product information under different failure mechanisms, the tester can obtain test data and determine the preset quantile of the logarithm of sample life data associated with each test bench when the failure mechanisms of multiple test benches are different. This process is similar to the process of obtaining test data and determining the preset quantile of the logarithm of sample life data associated with each test bench when the failure mechanisms of multiple test benches are the same, and the embodiment of the present disclosure will not be described in detail.

[0084] In step S203, the product information determination device can process the first model matrix and the second model matrix based on the quasi-unrelated regression algorithm to obtain the model regression coefficients of the first linear regression model and the second linear regression model, and substitute the model regression coefficients into the first linear regression model and the second linear regression model to obtain the first target linear regression model and the second target linear regression model.

[0085] In an optional embodiment, the product information determination device processes the first model matrix and the second model matrix based on a quasi-unrelated regression algorithm to obtain the model regression coefficients of the first linear regression model and the second linear regression model, including: determining the random error matrix of the quality parameter values ​​in the first model matrix and the variance-covariance matrix between the random errors of the life data in the second model matrix based on the least squares method; then, merging the first model matrix and the second model matrix into a linear regression model matrix to be processed; further, processing the linear regression model matrix to be processed based on a quasi-unrelated regression algorithm to obtain an estimated matrix of model regression coefficients and the covariance of the estimated matrix of model regression coefficients, wherein the estimated matrix of model regression coefficients and the covariance of the estimated matrix of model regression coefficients are respectively related to the variance-covariance matrix; wherein, if the estimated matrix of model regression coefficients is tested based on the covariance of the estimated matrix of model regression coefficients, the data in the estimated matrix of model regression coefficients are determined as the model regression coefficients of the first linear regression model and the second linear regression model. The model regression coefficients of the second linear regression model can be determined by combining the random error matrix of the quality parameter values ​​in the first model matrix determined based on the least squares method and the variance covariance matrix between the random errors of the life data in the second model matrix, thereby improving the reliability of the model regression coefficients of the determined second linear regression model.

[0086] Among them, the random error matrix of the quality parameter values ​​in the first model matrix and the variance covariance matrix between the random errors of the life data in the second model matrix are:

[0087] ;(Formula 11)

[0088] In formula 11 , , , and m is the number of test units, k is the number of product parameters in the first linear regression model, s is the number of product parameters in the second linear regression model, is the least squares regression residual matrix of the quality parameter values, is the transposed matrix of the least squares regression residual matrix of the quality parameter values, is the least squares regression residual matrix of life data, is the transposed matrix of the least squares regression residual matrix of life data; where, , is the transposed matrix of the parameter value matrix of the product parameters in the first model matrix, is the transposed matrix of the parameter value matrix of the product parameters in the second model matrix.

[0089] The linear regression model matrix to be processed is:

[0090] ;(Formula 12)

[0091] In formula 12, is the mean value of the sample quality parameter in the first model matrix, and the random error matrix of the sample life data to the quality parameter value in the second model matrix p The matrix of quantiles, is a matrix composed of the parameter values ​​of the product parameters in the first model matrix and the second model matrix, is a matrix composed of the model regression coefficients in the first model matrix and the model regression coefficients in the second model matrix, is a matrix composed of the random errors of the quality parameter values ​​in the first model matrix and the random errors of the life data in the second model matrix.

[0092] It should be noted that the estimated matrix of the model regression coefficient determined based on the seemingly unrelated regression algorithm is: The best linear unbiased estimate of :

[0093] ;(Formula 13)

[0094] In formula 13, is the Kronecker product, is the random error matrix of the quality parameter values ​​and the inverse of the variance-covariance matrix between the random errors of the life data.

[0095] The covariance of the estimated matrix of the model regression coefficients is:

[0096] ;(Formula 14)

[0097] It should be noted that in the embodiment of the present disclosure, based on the covariance of the estimated matrix of the model regression coefficient, the process of testing the estimated matrix of the model regression coefficient can be implemented based on the t-test. Among them, the statistic obtained by the t-test is:

[0098] ;(Formula 15)

[0099] In formula 15, is the statistic obtained by the t-test, yes Corresponds to diagonal elements of .

[0100] In an optional factual method, if the test of the estimated matrix of the model regression coefficients based on the covariance of the estimated matrix of the model regression coefficients fails, the product information determination device can repeat the above process of determining the estimated matrix of the model regression coefficients and the covariance of the estimated matrix of the model regression coefficients until the test of the estimated matrix of the model regression coefficients based on the covariance of the estimated matrix of the model regression coefficients passes, thereby obtaining the model regression coefficients of the first linear regression model and the second linear regression model.

[0101] In step S204, the product information determination device simultaneously optimizes the first target linear regression model and the second target linear regression model within the parameter value range of the product parameter based on a preset optimization algorithm to obtain the target parameter value of the product parameter.

[0102] In the embodiments of the present disclosure, the parameter value range of the product parameter can be determined based on actual needs, and the embodiments of the present disclosure are not limited to this.

[0103] In an optional embodiment, the process of obtaining the target parameter value of the product parameter by the product information determination device by optimizing the first target linear regression model and the second target linear regression model simultaneously within the parameter value interval of the product parameter based on a preset optimization algorithm may include: optimizing the first target linear regression model and the second target linear regression model simultaneously within the parameter value interval of the product parameter based on a genetic algorithm to obtain the target parameter value of the product parameter. The target parameter value may be determined based on the genetic algorithm within the parameter value interval of the product parameter so as to obtain a more accurate and reliable target parameter value.

[0104] It should be noted that in the embodiments of the present disclosure, the preset optimization algorithm may also be a particle swarm optimization algorithm, or a whale algorithm, etc., and the present disclosure does not limit this.

[0105] In an optional implementation, after determining the target parameter value of the product parameter, the estimated quality parameter value of the target product is determined based on the target parameter value of the product parameter and the first target linear regression model, wherein the parameter value of the product parameter of the target product is the target parameter value; and, based on the target parameter value of the product parameter and the second target linear regression model, the estimated life data of the target product is determined; further, the estimated quality parameter value and the estimated life data are determined as the quality information of the target product. After determining the target parameter value of the product parameter, the quality parameter value and life data of the target product can be estimated to facilitate understanding of the quality information of the target product after quality improvement.

[0106] For example, if the product that needs to determine product information is product A, the quality parameter of product A is volume, and the product parameters of product A include thickness, adjustment time and width. The relevant information of the product parameters is shown in Table 1. In order to facilitate data analysis and processing, the parameter value range of the product parameters is encoded. The encoded product parameters The respective parameter value ranges are:

[0107]

[0108] in, is the thickness of the product, Adjustment time for the product, Is the product width.

[0109] Table 1

[0110]

[0111] Optionally, assuming that there are 17 test benches, the acquired test data can be as shown in Table 2. Then, the mean of the volume and the decile of the logarithm of the life data can be determined based on the test data, and the mean of the volume and the decile of the logarithm of the life of the equipment can be obtained, as shown in Table 3 below; and the first model matrix and the second model matrix are constructed based on the data in Table 3.

[0112] Furthermore, by combining the least square method and the seemingly unrelated regression algorithm, the estimated matrix of the regression coefficients of the determined model is shown in Table 4 In Table 4 , the t value is the statistic obtained by the t test; Figure 3 and Figure 4 The figure shows the fitting diagnostic diagram of the mean of the quality parameter and the fitting diagnostic diagram of the quantile of the logarithm of the life data for each test bench during the operation of the seemingly unrelated regression algorithm, where: Figure 3 The a part in is the residual between the actual observed value of the volume and the model predicted value, which is used to check the fit of the model; Figure 3 Part b in the figure is the volume test value associated with each test rig, which helps to visualize the impact of different test rigs on the volume, thereby evaluating the treatment effect and the quality of model fitting; Figure 3 The c part of is the distribution of the residual of the volume, where Figure 3 When the points in part c of are distributed along the diagonal, it indicates that the residuals conform to the normal distribution. Figure 3 The d part in is the residual of the volume, which conforms to the normal distribution assumption; Figure 3 The e part in is the white noise characteristic of the residual sequence of the volume. There is no significant pattern in the figure, and the white noise probability is relatively uniform under different lag periods, so the residual sequence is white noise.

[0113] Table 2

[0114]

[0115] Figure 4 Part a in is the residual between the actual observed value of the quantile of the logarithm of life data and the model predicted value, which is used to check the fit of the model; Figure 4 Part b in is the quantile test value of the logarithm of life data associated with each test rig, which helps to visualize the impact of different test rigs on the quantile of the logarithm of life data, thereby evaluating the treatment effect and the quality of model fitting; Figure 4 The c in is the distribution of residuals, where Figure 4 When the points in part c of the equation are distributed along the diagonal, it indicates that the residuals of the quantiles of the logarithm of life data conform to the normal distribution. Figure 4 The d part in is the residual of the quantile of the logarithm of life data, which conforms to the normal distribution assumption; Figure 4 The e part in is the white noise characteristic of the residual sequence of the quantiles of the logarithm of life data. There is no significant pattern in the figure, and the white noise probability is relatively uniform under different lag periods, so the residual sequence is white noise.

[0116] Table 3

[0117]

[0118] Next, after determining the model regression coefficient based on the quasi-unrelated regression algorithm, the first target linear regression model for determining the volume can be obtained as follows:

[0119] ;(Formula 16)

[0120] In formula 16, It is the interaction term of product thickness and width in product parameters.

[0121] And, the second objective linear regression model for determining the quantile model of the logarithm of the life response is obtained as:

[0122] ;(Formula 17)

[0123] Table 4

[0124]

[0125] Finally, based on the genetic algorithm, the first target linear regression model in Formula 16 and the second target linear regression model in Formula 17 are optimized simultaneously within the parameter value range of thickness, adjustment time and width to obtain the target parameter values ​​of thickness, adjustment time and width, as shown in Table 5. It can be understood that when the thickness, adjustment time and width of the product are as shown in Table 5, the product is the target product after quality update.

[0126] Table 5

[0127]

[0128] Among them, Figure 5 As shown, Figure 5 This is a schematic diagram of the results of obtaining target parameter values ​​of thickness, adjustment time and width based on genetic algorithm, where: Figure 5 Part a in the figure shows the average distance between individuals, which indicates the average distance between individuals in the genetic algorithm population. It usually refers to the distance in the genotype or solution space, reflecting the diversity of the population. Figure 5 Part b in the GA is the stopping criterion, which indicates the stopping condition or standard of the GA, such as reaching the maximum number of iterations or reaching a specific fitness threshold. It helps determine when the GA terminates and ensures that the GA stops after meeting specific conditions, thereby avoiding over-calculation or premature stopping. Stall(T) and Stall(G) are the stopping conditions used in the GA to determine whether the GA tends to stagnate in terms of the objective function value and overall fitness. Stall(T) indicates the stagnation condition of the objective function. If the change (improvement) of the objective function value is less than a preset threshold (usually a very small positive number, which can be determined based on actual needs) in a certain number of consecutive generations (iterations), the algorithm is considered to be stagnant in terms of the objective function value and may have found a local optimal solution. Stall(G) indicates the stagnation condition of the overall fitness. If the change of the overall fitness (i.e., the average or maximum value of the fitness values ​​of all individuals in the population) is less than a preset threshold in a certain number of consecutive generations (iterations), the fitness of the algorithm is considered to be stagnant overall. Figure 5 The c part in is the maximum constraint, which is used to determine whether the individual violates the constraint conditions and helps understand how the algorithm handles the constraints. The maximum constraint is , which is less than the set Constraints of Figure 5 The d part in is the objective function value (best value) of the optimal solution in the group in each iteration and the average objective function value (average value) of all individuals in the group. Figure 5 The e part in represents the best individual solution found in the current iteration, which helps to intuitively understand the setting of the current best solution; Figure 5 The f part in it represents the score of the best individual, the score of the worst individual, and the average score of all individuals in the group in each iteration. The score of the best individual and the score of the worst individual are used to characterize the extreme performance in the group, and the average score provides the overall performance of the group, which helps to evaluate the optimization process and identify whether there is a risk of premature convergence or changes in group performance.

[0129] Furthermore, the target parameter values ​​of thickness, adjustment time and width in Table 5 can be substituted into the first target linear regression model in Formula 15 to obtain the estimated quality parameter value of the target product. is 80, and the target parameter values ​​of thickness, adjustment time and width in Table 5 are substituted into the second target linear regression model in Formula 16, and the quantile of the logarithm of the estimated life data of the target product is obtained to be 8.0135.

[0130] An exemplary embodiment of the present disclosure provides a product information determination device, which may be a chip of the product information determination device. Figure 6 FIG. 2 shows a schematic block diagram of functional modules of a device for determining product information according to an exemplary embodiment of the present disclosure. Figure 6 As shown, the product information determination device 600 includes:

[0131] An acquisition module 601 is configured to acquire test data of a sample product, wherein the test data includes a plurality of sample quality parameter values, a plurality of sample life data and sample parameter values ​​of a product parameter of the sample product;

[0132] A construction module 602 is configured to construct a first model matrix of a first linear regression model between a mean value of a quality parameter and the product parameter based on the test data, and to construct a second model matrix of a second linear regression model between a preset quantile of a logarithm of life data and the product parameter;

[0133] A determination module 603 is configured to process the first model matrix and the second model matrix based on a quasi-unrelated regression algorithm to obtain model regression coefficients of the first linear regression model and the second linear regression model, and substitute the model regression coefficients into the first linear regression model and the second linear regression model to obtain a first target linear regression model and a second target linear regression model;

[0134] The optimization module 604 is configured to simultaneously optimize the first target linear regression model and the second target linear regression model within the parameter value range of the product parameter based on a preset optimization algorithm to obtain a target parameter value of the product parameter.

[0135] Optionally, the multiple sample quality parameter values ​​are multiple sample quality parameter values ​​respectively obtained on multiple test benches, and the construction module 602 is configured as follows:

[0136] determining a mean of a plurality of sample quality parameter values ​​associated with each test bench to obtain a sample quality parameter mean;

[0137] Substituting the sample parameter values ​​of the product parameters and the mean value of the sample quality parameter associated with each test bench into the first linear regression model to obtain a first linear regression model associated with each test bench;

[0138] The first linear regression model associated with each test bench is combined to obtain a first model matrix of the first linear regression model, wherein the first linear regression model associated with each test bench is:

[0139] ;

[0140] in, For the i The vector form of the mean value of the sample quality parameter associated with the test bench, For the i The first sample product on the test bench k Sample parameter values ​​for product parameters, For the i The random errors in the values ​​of the quality parameters in the first linear regression model associated with the test rigs, The model regression coefficients of the first linear regression model.

[0141] Optionally, the plurality of sample life data includes a plurality of sample life data respectively obtained on a plurality of test benches, and the plurality of sample life data obeys a Weibull distribution. The construction module 602 is configured as follows:

[0142] Determining a preset quantile of the logarithm of the sample life data associated with each test bench based on the plurality of sample life data respectively acquired on the plurality of test benches;

[0143] Substituting the sample parameter values ​​of the product parameters and the preset quantiles of the logarithm of the sample life data associated with each test bench into the second linear regression model to obtain a second linear regression model associated with each test bench;

[0144] The second linear regression model associated with each test bench is combined to obtain a second model matrix of the second linear regression model, wherein the second linear regression model associated with each test bench is:

[0145] ;

[0146] in, For the i The logarithm of the sample life data associated with the test benches p Quantile For the i The first sample product on the test bench s Sample parameter values ​​for product parameters, For thei Random Errors in Lifetime Data from the Second Linear Regression Model for the Test Rigs The model regression coefficients of the second linear regression model.

[0147] Optionally, the construction module 602 is configured to:

[0148] Determine a sample minimum extreme value distribution function, a cumulative sum distribution function and a probability density function of the logarithm of the sample life data based on the plurality of sample life data respectively obtained on the plurality of test benches;

[0149] a sample location parameter value and a sample scale parameter value associated with each test bed based on the cumulative sum distribution function and the probability density function, wherein the sample scale parameter value associated with each test bed is the same;

[0150] Substitute the standard distribution function of the sample minimum extreme value distribution function, as well as the location parameter value and the sample scale parameter value associated with each test bench into the preset quantiles of the logarithm of the life data, and the relationship equation between the standard distribution function of the minimum extreme value distribution function, the location parameter and the scale parameter to obtain the preset quantiles of the logarithm of the sample life data associated with each test bench.

[0151] Optionally, the determining module 603 is configured to:

[0152] Determine, based on least squares processing, a random error matrix of quality parameter values ​​in the first model matrix and a variance-covariance matrix between random errors of life data in the second model matrix;

[0153] Merging the first model matrix and the second model matrix into a linear regression model matrix to be processed;

[0154] Processing the linear regression model matrix to be processed based on a quasi-unrelated regression algorithm to obtain an estimated matrix of model regression coefficients and a covariance of the estimated matrix of model regression coefficients, wherein the estimated matrix of model regression coefficients and the covariance of the estimated matrix of model regression coefficients are respectively related to the variance-covariance matrix;

[0155] If the estimated matrix of the model regression coefficients is tested based on the covariance of the estimated matrix of the model regression coefficients, the data in the estimated matrix of the model regression coefficients are determined as the model regression coefficients of the first linear regression model and the second linear regression model.

[0156] Optionally, the optimization module 604 is configured to:

[0157] Based on the genetic algorithm, the first target linear regression model and the second target linear regression model are optimized simultaneously within the parameter value range of the product parameter to obtain the target parameter value of the product parameter.

[0158] Optionally, the apparatus further includes an estimation model 605 configured to:

[0159] Determining an estimated quality parameter value of a target product based on the target parameter value of the product parameter and the first target linear regression model, wherein the parameter value of the product parameter of the target product is the target parameter value;

[0160] Determining estimated life data of the target product based on the target parameter value of the product parameter and the second target linear regression model;

[0161] The estimated quality parameter value and the estimated life data are determined as the quality information of the target product.

[0162] The exemplary embodiment of the present disclosure also provides an electronic device, comprising: at least one processor; and a memory connected to the at least one processor in communication. The memory stores a computer program that can be executed by the at least one processor, and the computer program is used to cause the electronic device to perform the method according to the embodiment of the present disclosure when executed by the at least one processor.

[0163] The exemplary embodiments of the present disclosure also provide a non-transitory computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor of a computer, is used to cause the computer to perform the method according to the embodiments of the present disclosure.

[0164] like Figure 7 As shown, the exemplary embodiment of the present disclosure further provides a computer program product 700, including a computer program 701, wherein when the computer program is executed by a processor of a computer, it is used to enable the computer to perform the method according to the embodiment of the present disclosure.

[0165] like Figure 8 As shown, the electronic device 800 includes a computing unit 801, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 802 or a computer program loaded from a storage unit 808 to a random access memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the electronic device 800 can also be stored. The computing unit 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0166] A plurality of components in the electronic device 800 are connected to the I / O interface 805, including: an input unit 806, an output unit 807, a storage unit 808, and a communication unit 809. The input unit 806 may be any type of device capable of inputting information to the electronic device 800, and the input unit 806 may receive input digital or character information, and generate key signal inputs related to user settings and / or function control of the electronic device. The output unit 807 may be any type of device capable of presenting information, and may include, but is not limited to, a display, a speaker, a video / audio output terminal, a vibrator, and / or a printer. The storage unit 808 may include, but is not limited to, a disk, an optical disk. The communication unit 809 allows the electronic device 800 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks, and may include, but is not limited to, a modem, a network card, an infrared communication device, a wireless communication transceiver, and / or a chipset, such as a Bluetooth™ device, a WiFi device, a WiMax device, a cellular communication device, and / or the like.

[0167] The computing unit 801 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 801 performs the various methods and processes described above. For example, in some embodiments, the method of the exemplary embodiment of the present disclosure may be implemented as a computer software program, which is tangibly included in a machine-readable medium, such as a storage unit 808. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 800 via the ROM 802 and / or the communication unit 809. In some embodiments, the computing unit 801 may be configured to perform the method of the exemplary embodiment of the present disclosure in any other appropriate manner (e.g., by means of firmware).

[0168] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0169] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0170] As used in this disclosure, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, apparatus, and / or device (e.g., disk, optical disk, memory, programmable logic device (PLD)) for providing machine instructions and / or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term "machine-readable signal" refers to any signal for providing machine instructions and / or data to a programmable processor.

[0171] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0172] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0173] A computer system may include clients and servers. Clients and servers are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship to each other.

[0174] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instruction is loaded and executed on a computer, the process or function described in the embodiment of the present disclosure is executed in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, a terminal, a user device or other programmable device. The computer program or instruction may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer program or instruction may be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server, data center, etc. that integrates one or more available media. The available medium may be a magnetic medium, for example, a floppy disk, a hard disk, a tape; it may also be an optical medium, for example, a digital video disc (DVD); it may also be a semiconductor medium, for example, a solid state drive (SSD).

[0175] Although the present disclosure has been described in conjunction with specific features and embodiments thereof, it is apparent that various modifications and combinations may be made thereto without departing from the spirit and scope of the present disclosure. Accordingly, this specification and the drawings are merely exemplary illustrations of the present disclosure as defined by the appended claims, and are deemed to have covered any and all modifications, variations, combinations or equivalents within the scope of the present disclosure. Obviously, those skilled in the art may make various modifications and variations to the present disclosure without departing from the spirit and scope of the present disclosure. Thus, if these modifications and variations of the present disclosure fall within the scope of the claims of the present disclosure and their equivalents, the present disclosure is also intended to include these modifications and variations.

Claims

1. A method for determining product information, characterized in that: include: Acquire test data of a sample product, wherein the test data includes a plurality of sample quality parameter values, a plurality of sample life data and sample parameter values ​​of a product parameter of the sample product; Based on the test data, construct a first model matrix of a first linear regression model between a mean value of a quality parameter and the product parameter, and construct a second model matrix of a second linear regression model between a preset quantile of the logarithm of the life data and the product parameter; Processing the first model matrix and the second model matrix based on a quasi-unrelated regression algorithm to obtain model regression coefficients of the first linear regression model and the second linear regression model, and substituting the model regression coefficients into the first linear regression model and the second linear regression model to obtain a first target linear regression model and a second target linear regression model; Based on a preset optimization algorithm, the first target linear regression model and the second target linear regression model are simultaneously optimized within the parameter value range of the product parameter to obtain a target parameter value of the product parameter; The multiple sample quality parameter values ​​are multiple sample quality parameter values ​​respectively obtained on multiple test benches, and the first model matrix of the first linear regression model between the quality parameter mean and the product parameter is constructed based on the test data, including: determining a mean of a plurality of sample quality parameter values ​​associated with each test bench to obtain a sample quality parameter mean; The plurality of sample life data includes a plurality of sample life data respectively obtained on the plurality of test benches, the plurality of sample life data obeys Weibull distribution, and based on the test data, a second model matrix of a second linear regression model between a preset quantile of the logarithm of the life data and the product parameter is constructed, including: Based on the plurality of sample life data respectively acquired on the plurality of test benches, a preset quantile of the logarithm of the sample life data associated with each test bench is determined.

2. The product information determination method according to claim 1, characterized in that: The first model matrix of the first linear regression model between the mean of the quality parameter and the product parameter is constructed based on the test data, and further includes: Substituting the sample parameter values ​​of the product parameters and the mean value of the sample quality parameter associated with each test bench into the first linear regression model to obtain a first linear regression model associated with each test bench; The first linear regression model associated with each test bench is combined to obtain a first model matrix of the first linear regression model, wherein the first linear regression model associated with each test bench is: ； in, For the The vector form of the mean value of the sample quality parameter associated with the test bench, For the The first sample product on the test bench k Sample parameter values ​​for product parameters, For the The random errors in the values ​​of the quality parameters in the first linear regression model associated with the test rigs, The model regression coefficients of the first linear regression model.

3. The product information determination method according to claim 1, characterized in that: The second model matrix of the second linear regression model between the preset quantiles of the logarithm of the life data and the product parameters is constructed based on the test data, and further includes: Substituting the sample parameter values ​​of the product parameters and the preset quantiles of the logarithm of the sample life data associated with each test bench into the second linear regression model to obtain a second linear regression model associated with each test bench; The second linear regression model associated with each test bench is combined to obtain a second model matrix of the second linear regression model, wherein the second linear regression model associated with each test bench is: ； in, For the The logarithm of the sample life data associated with the test benches Quantile, For the The first sample product on the test bench s Sample parameter values ​​for product parameters, For the The random errors in the life data in the second linear regression model associated with the test rigs, The model regression coefficients of the second linear regression model.

4. The product information determination method according to claim 3, characterized in that: The step of determining a preset quantile of a logarithm of the sample life data associated with each test bench based on the plurality of sample life data respectively acquired on the plurality of test benches comprises: Determine a sample minimum extreme value distribution function, a cumulative sum distribution function and a probability density function of the logarithm of the sample life data based on the plurality of sample life data respectively obtained on the plurality of test benches; Determining a sample location parameter value and a sample scale parameter value associated with each test bed based on the cumulative sum distribution function and the probability density function, wherein the sample scale parameter value associated with each test bed is the same; Substitute the standard distribution function of the sample minimum extreme value distribution function, as well as the location parameter value and the sample scale parameter value associated with each test bench into the preset quantiles of the logarithm of the life data, and the relationship equation between the standard distribution function of the minimum extreme value distribution function, the location parameter and the scale parameter to obtain the preset quantiles of the logarithm of the sample life data associated with each test bench.

5. The product information determination method according to claim 1, characterized in that: The step of processing the first model matrix and the second model matrix based on a quasi-unrelated regression algorithm to obtain model regression coefficients of the first linear regression model and the second linear regression model includes: Determine, based on least squares processing, a random error matrix of quality parameter values ​​in the first model matrix and a variance-covariance matrix between random errors of life data in the second model matrix; Merging the first model matrix and the second model matrix into a linear regression model matrix to be processed; Processing the linear regression model matrix to be processed based on a quasi-unrelated regression algorithm to obtain an estimated matrix of model regression coefficients and a covariance of the estimated matrix of model regression coefficients, wherein the estimated matrix of model regression coefficients and the covariance of the estimated matrix of model regression coefficients are respectively related to the variance-covariance matrix; If the estimated matrix of the model regression coefficients is tested based on the covariance of the estimated matrix of the model regression coefficients, the data in the estimated matrix of the model regression coefficients are determined as the model regression coefficients of the first linear regression model and the second linear regression model.

6. The product information determination method according to claim 1, characterized in that: The method of optimizing the first target linear regression model and the second target linear regression model simultaneously within the parameter value range of the product parameter based on a preset optimization algorithm to obtain the target parameter value of the product parameter includes: Based on the genetic algorithm, the first target linear regression model and the second target linear regression model are optimized simultaneously within the parameter value range of the product parameter to obtain the target parameter value of the product parameter.

7. The product information determination method according to claim 1, characterized in that: The method further comprises: Determining an estimated quality parameter value of a target product based on the target parameter value of the product parameter and the first target linear regression model, wherein the parameter value of the product parameter of the target product is the target parameter value; Determining estimated life data of the target product based on the target parameter value of the product parameter and the second target linear regression model; The estimated quality parameter value and the estimated life data are determined as the quality information of the target product.

8. A device for determining product information, characterized in that: include: An acquisition module is configured to acquire test data of a sample product, wherein the test data includes a plurality of sample quality parameter values, a plurality of sample life data and sample parameter values ​​of a product parameter of the sample product; A construction module is configured to construct a first model matrix of a first linear regression model between a mean value of a quality parameter and the product parameter based on the test data, and to construct a second model matrix of a second linear regression model between a preset quantile of a logarithm of life data and the product parameter; a determination module configured to process the first model matrix and the second model matrix based on a quasi-unrelated regression algorithm to obtain model regression coefficients of the first linear regression model and the second linear regression model, and substitute the model regression coefficients into the first linear regression model and the second linear regression model to obtain a first target linear regression model and a second target linear regression model; An optimization module is configured to optimize the first target linear regression model and the second target linear regression model simultaneously within a parameter value interval of the product parameter based on a preset optimization algorithm to obtain a target parameter value of the product parameter; The multiple sample quality parameter values ​​are multiple sample quality parameter values ​​respectively obtained on multiple test benches, and the construction module is configured as follows: determining a mean of a plurality of sample quality parameter values ​​associated with each test bench to obtain a sample quality parameter mean; The plurality of sample life data includes a plurality of sample life data respectively obtained on a plurality of test benches, and the plurality of sample life data obeys Weibull distribution. The building module is configured as follows: Based on the plurality of sample life data respectively acquired on the plurality of test benches, a preset quantile of the logarithm of the sample life data associated with each test bench is determined.

9. An electronic device comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the product information determination method according to any one of claims 1 to 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, the product information determination method according to any one of claims 1 to 7 is implemented.

Images