CN107015875B - A method and device for evaluating the storage life of an electronic complete machine - Google Patents

Abstract

The invention discloses a method and a device for evaluating the storage life of an electronic complete machine. The method includes: acquiring performance degradation data and natural storage years of an electronic complete machine; constructing a degradation data trend model according to test samples in the performance degradation data, and according to the degradation data trend model and verification samples in the performance degradation data , obtain the predicted life of the complete electronic machine; obtain the acceleration factor according to the natural storage period and the pre-established acceleration model; obtain the characteristic life of the complete electronic machine according to the predicted life of the complete electronic machine and the acceleration factor. The invention analyzes the performance degradation data of the electronic complete machine, evaluates the life of the electronic complete machine according to the analysis results, then analyzes the acceleration factor of the electronic complete machine based on the storage life of the electronic complete machine, and then combines the estimated life and acceleration factor. Evaluating the characteristic life of an electronic complete machine has the advantage of high evaluation accuracy.

Description

A method and device for evaluating the storage life of an electronic complete machine

technical field

The invention relates to the technical field of electronic complete machines, in particular to a method and device for evaluating the storage life of an electronic complete machine.

Background technique

In the accelerated storage test of the electronic complete machine, due to the complex functions of the electronic complete machine product, there are situations where the product performance degradation law is complex, and the natural storage data and the accelerated storage test data coexist. This situation makes it difficult to evaluate the test results. The traditional accelerated test data evaluation method cannot process data for this situation, so that the acceleration factor or storage life of the product cannot be evaluated, and the test purpose cannot be achieved.

It is a common engineering problem that needs to be solved urgently. Using an effective accelerated test evaluation method can improve the effective use of data resources, thereby improving the accuracy of the evaluation results, and may even affect the research results and reduce the hidden dangers caused by inaccurate life evaluation results.

When predicting the trend of the product performance degradation law, the commonly used processing method is the regression analysis method. The regression analysis method uses linear functions, exponential functions, power functions, etc. to perform regression fitting on the performance degradation data of the product, and obtain the regression equation of the degradation trend. The degradation trend is then predicted, but for some nonlinear degradation data with complex laws, the regression analysis method is not very accurate, and even sometimes difficult to apply. The current research hotspot is the artificial neural network method, but in engineering applications, the artificial neural network method is not ideal for predicting the trend of product degradation, and its application needs to be further studied.

In the process of realizing the present invention, the inventor found that for the coexistence of natural storage data and accelerated storage test data, the current processing method is to first use the accelerated test data to evaluate the test results, obtain the storage life results of the product, and then simply Adding the natural storage time to the life results, and even sometimes ignoring the natural storage data when evaluating the acceleration factor, these processing methods will bias the evaluation results.

SUMMARY OF THE INVENTION

An object of the present invention is to solve the problem of errors in the evaluation results when evaluating the storage life of an electronic complete product in the prior art.

The present invention provides a method for evaluating the storage life of an electronic complete machine, including:

Obtain the performance degradation data and natural storage period of the electronic complete machine;

Build a degradation data trend model according to the test samples in the performance degradation data, and obtain the predicted life of the electronic complete machine according to the degradation data trend model and the verification samples in the performance degradation data;

Obtain the acceleration factor according to the natural storage period and the pre-established acceleration model;

The characteristic lifetime of the electronic complete unit is obtained according to the predicted lifetime of the electronic complete unit and the acceleration factor.

Optionally, the performance degradation data includes: multiple sets of product performance data and detection time sample data corresponding to the product performance data;

Correspondingly, the building a degradation data trend model according to the test samples in the performance degradation data includes:

dividing the performance degradation data into test samples and verification samples according to the detection time;

Use support vector machine to establish the initial degradation data trend model;

Taking the product performance data in the test sample as the input vector and the performance degradation data as the output vector, the initial degradation data trend model is trained by using the least squares support vector machine algorithm to construct the degradation data trend model.

Optionally, according to the degradation data trend model and the verification samples in the performance degradation data, obtaining the predicted life of the electronic complete machine includes:

According to the order of detection time, the prediction step is performed;

The predicting step includes:

Taking the detection time in the first group of sample data in the verification sample as input, and combining with the degradation data trend model, obtain the predicted value of the corresponding product performance parameter;

Determine whether the predicted value of the product performance parameter reaches the upper limit or lower limit of the product failure threshold;

If so, the detection time corresponding to the predicted value of the product performance parameter is taken as the predicted life of the electronic complete machine.

Optionally, if the predicted value of the product performance parameter does not reach the upper limit or the lower limit of the product failure threshold, update the degradation data trend model according to the first set of sample data in the verification sample;

delete the first group of sample data in the verification sample;

The predicting step is repeatedly performed until the predicted lifetime of the electronic complete machine is obtained.

Optionally, before obtaining the acceleration factor according to the natural storage period and the pre-established acceleration model, the method further includes:

According to the temperature stress data included in the performance degradation data, combined with the degradation data trend model, obtain the predicted lifespan of a plurality of test electronic complete machines under different temperature stresses;

The acceleration model is constructed according to the predicted life of multiple test electronic complete machines under different temperature stress and the natural storage life of each test electronic complete machine;

Correspondingly, obtaining the acceleration factor according to the natural storage period and the pre-established acceleration model includes:

Evaluate the parameters in the acceleration model according to the predicted life of a plurality of test electronic complete machines under different temperature stresses;

Acceleration factors of the electronic complete machine under different temperature stresses are obtained by calculating according to the acceleration model.

The invention provides a storage life evaluation device for an electronic complete machine, including:

The acquisition module is used to acquire the performance degradation data and natural storage period of the electronic whole machine;

an evaluation module, configured to construct a degradation data trend model according to the test samples in the performance degradation data, and obtain the predicted life of the electronic complete machine according to the degradation data trend model and the verification samples in the performance degradation data;

a processing module, configured to obtain an acceleration factor according to the natural storage period and a pre-established acceleration model;

An optimization module, configured to obtain the characteristic life of the electronic complete device according to the predicted life of the electronic complete device and the acceleration factor.

Optionally, the performance degradation data includes: multiple sets of product performance data and detection time sample data corresponding to the product performance data;

Correspondingly, the evaluation module is used to divide the performance degradation data into a test sample and a verification sample according to the detection time; use a support vector machine to establish an initial degradation data trend model; use the product performance data in the test sample is the input vector, the performance degradation data value is the output vector, and the initial degradation data trend model is trained by using the least squares support vector machine algorithm to construct the degradation data trend model.

Optionally, the evaluation module is further configured to perform the prediction step according to the order of detection time;

The predicting step includes: taking the detection time in the first group of sample data in the verification sample as an input, and combining with the degradation data trend model to obtain the predicted value of the corresponding product performance parameter; judging the value of the product performance parameter. Whether the predicted value reaches the upper limit or lower limit of the product failure threshold; if so, the detection time corresponding to the predicted value of the product performance parameter is taken as the predicted life of the electronic complete machine.

Optionally, the evaluation module is further configured to, if it is judged that the predicted value of the product performance parameter does not reach the upper limit or the lower limit of the product failure threshold, then according to the first group of sample data in the verification sample, the The degradation data trend model is updated; the first group of sample data in the verification sample is deleted; and the predicting step is repeatedly performed until the predicted life of the electronic complete machine is obtained.

Optionally, the device further includes: a modeling module;

The modeling module is configured to obtain the predicted life spans of multiple test electronic complete machines under different temperature stresses according to the temperature stress data included in the performance degradation data and in combination with the degradation data trend model; Build an accelerated model for the predicted life of the test electronic complete machine and the natural storage life of each test electronic complete machine;

Correspondingly, the processing module is used to evaluate the parameters in the acceleration model according to the predicted life of a plurality of test electronic complete machines under different temperature stresses; calculate and obtain the electronic complete machine under different temperature stresses according to the acceleration model acceleration factor.

It can be seen from the above technical solutions that the method and device for evaluating the storage life of an electronic complete machine proposed by the present invention analyze the performance degradation data of the electronic complete machine, and evaluate the life of the electronic complete machine according to the analysis results, and then based on the electronic complete machine. The storage life analyzes the acceleration factor of the complete electronic machine, and then combines the estimated life and the acceleration factor to evaluate the characteristic life of the electronic complete machine, which has the advantage of high evaluation accuracy. .

Description of drawings

The features and advantages of the present invention will be more clearly understood by reference to the accompanying drawings, which are schematic and should not be construed as limiting the invention in any way, in which:

FIG. 1 shows a schematic flowchart of a method for evaluating the storage life of an electronic complete machine provided by the present invention;

FIG. 2 shows a schematic flowchart of a prediction step provided by the present invention;

3a-3c show schematic diagrams of product test data under different stress levels provided by the present invention;

Figures 4a-4c show schematic diagrams of product degradation trend prediction curves under different stress levels provided by the present invention;

FIG. 5 shows a schematic structural diagram of a device for evaluating the storage life of an electronic complete machine provided by the present invention.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be described clearly and completely below with reference to the accompanying drawings. Obviously, the described embodiments are part of the implementation of the present invention. examples, but not all examples. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present invention.

FIG. 1 shows a schematic flowchart of a method for evaluating the storage life of an electronic complete machine according to an embodiment of the present invention. Referring to FIG. 1 , the method can be implemented by a processor, and specifically includes the following steps:

110. Obtain the performance degradation data and natural storage period of the electronic complete machine;

120. Construct a degradation data trend model according to the test samples in the performance degradation data, and obtain the predicted life of the electronic complete machine according to the degradation data trend model and the verification samples in the performance degradation data;

130. Obtain an acceleration factor according to the natural storage period and a pre-established acceleration model;

It should be noted that, before step 130, the method further includes: according to the temperature stress data included in the performance degradation data, combined with the degradation data trend model, to obtain the predicted life spans of multiple test electronic complete machines under different temperature stresses; The predicted life of multiple test electronic complete machines under temperature stress and the natural storage life of each test electronic complete machine build an accelerated model;

Correspondingly, step 130 specifically includes:

The parameters in the acceleration model are evaluated according to the predicted lifetimes of multiple test electronic complete machines under different temperature stresses; and the acceleration factors of the electronic complete machine under different temperature stresses are obtained by calculating according to the acceleration model.

140. Obtain the characteristic life of the complete electronic device according to the predicted life of the complete electronic device and the acceleration factor.

It can be seen that the embodiment of the present invention analyzes the performance degradation data of the electronic complete machine, evaluates the life of the electronic complete machine according to the analysis results, and then analyzes the acceleration factor of the electronic complete machine based on the storage life of the electronic complete machine, and then combines the evaluated results. The life and acceleration factors evaluate the characteristic life of the electronic complete machine, which has the advantage of high evaluation accuracy.

FIG. 2 shows a schematic flowchart of a prediction step provided by an embodiment of the present invention. The method can be implemented by a processor, and specifically includes the following steps:

210. Training data

The performance degradation data includes: multiple sets of product performance data and detection time sample data corresponding to the product performance data;

220. Establish a degradation trend model through the least squares support vector machine

dividing the performance degradation data into test samples and verification samples according to the detection time;

Use support vector machine to establish the initial degradation data trend model;

Taking the product performance data in the test sample as the input vector and the performance degradation data as the output vector, the initial degradation data trend model is trained by using the least squares support vector machine algorithm to construct the degradation data trend model.

230. Predict the performance degradation data value according to the degradation trend model

According to the order of detection time, the prediction step is performed;

The predicting step includes:

Taking the detection time in the first group of sample data in the verification sample as input, and combining with the degradation data trend model, obtain the predicted value of the corresponding product performance parameter;

240. Determine whether the performance degradation data value reaches the failure threshold, if so, execute step 250; if not, execute step 260;

250. Use the detection time corresponding to the predicted value of the product performance parameter as the predicted life of the electronic complete machine.

260. Update the degradation data trend model according to the first group of sample data in the verification sample; delete the first group of sample data in the verification sample;

The predicting step is repeatedly performed until the predicted lifetime of the electronic complete machine is obtained.

The present invention is further described in detail below in conjunction with the evaluation of the results of the accelerated storage test of a certain type of electronic complete machine:

9 units of a certain type of electronic complete machine were put into the accelerated storage test. All 9 units have a certain natural storage period. The test is carried out by applying constant stress. The test stress is temperature stress, and the stress level is divided into 3 grades, respectively At 80°C, 95°C and 110°C, 3 products were arranged for each stress level for testing. During the test, the performance parameters of the products were tested according to the specified test points, and the performance degradation data of 9 products were obtained, as shown in Figure 3a-Fig. 3c, the data has units omitted.

Step 1. Use support vector machine to establish a degradation data trend model:

Firstly, the trend model of degraded data is established by support vector machine, the detection time T=(t ₁ , t ₂ ,..., t _n ) corresponding to the performance degradation data is used as the input vector, and the performance degradation data value Y=(y ₁ , y _{2 )} ,…,y _n ) is the output vector, and the least squares support vector machine algorithm can be used to train the degraded data trend model:

In the formula, α and β support vector machine model parameters, ψ(*) is the kernel function, and the kernel function used here is the radial basis (RBF) kernel function.

The present invention completes the establishment of the above trend model through the least squares support vector machine toolbox in the MATLAB software. Degraded data trend model.

Step 2. Use the degradation trend model to predict product life:

Through the obtained degradation trend model f(t), the time t _n+1 corresponding to the predicted data is used as the input, and the predicted value y _n+1 of the product performance parameters can be obtained, that is, a set of predicted data {t _n+1 , y _n+1 }. Add this set of data to the original performance degradation data as new model training data, that is, the new model training data is T'=(t ₁ ,t ₂ ,...,t _n ,t _n+1 ) and Y'= (y ₁ ,y ₂ ,…,y _n ,y _n+1 ), a new degradation data trend model f'(t) can be obtained, and then the next set of predictions can be obtained through the new degradation data trend model f'(t) Data {t _n+2 ,y _n+2 }. In this way, the prediction model is continuously updated and the product performance data is predicted according to the above method. When the predicted product performance data {t _n+m , y _n+m } (m≥1) reaches the product failure threshold (the failure threshold is 0.639) , t _n+m is the predicted life of the product.

The degradation trend prediction curves of the products under various stress levels are shown in Figure 4a-4c, and the life prediction results of the products are shown in Table 1.

Table 1 Lifetime prediction results

Step 3. Combine natural storage data and accelerated storage test data:

There are r _i products under the temperature stress Si ( _i =1,2,...,k), and the predicted lifetimes of these products under the stress _{Si are P i1 ,} P _i2 ,...,P _iri (see Table 1). ), the natural storage years of the products are respectively _Qi1 , _Qi2 ,…, _Qiri (9 products are respectively 8 years, 8 years, 10 years, 10 years, 10 years, 10 years, 8 years, 8 years , 8 years), and set the acceleration factor of the accelerated temperature stress Si ( _i =1,2,…,k) relative to the normal temperature stress S0 as Ai, then the actual life of the product under the temperature stress Si should be:

_{Li ij} =P _ij +Q _ij /A _i (i=1,2,...,k; _j =1,2,...,ri ) (2)

Step 4. Accelerate model parameter evaluation:

There is the following acceleration model between the characteristic life θ _i of the product and the accelerated temperature stress Si _:

In the formula, a and b are the parameters to be estimated, and _Si is the accelerated temperature stress.

According to the acceleration model, the acceleration factor of the product under the accelerated stress level _Si relative to the normal stress level S ₀ can be obtained as:

It can be known that A _i is a function of b, then the product life L _ij in formula (2) is a function of b.

It is generally assumed that the life of complex electronic products obeys the exponential distribution. According to the parameter estimation method of the exponential distribution, the maximum likelihood estimation of the average life of the product under each stress level _Si is:

Since _{Li ij} is a function of b, so θ _i is also a function of b.

According to the temperature stress level of the k group and the average life {1/S _i ,lnθ _i }(i=1,2,...,k), using the formula (3), the estimated values of the parameters a and b can be obtained by the least square method:

To solve the above-mentioned transcendental equations, the present invention realizes the solving of the above-mentioned equations through MATLAB software programming, and the results are a=-6.17, b=5644.8.

Step 5. Evaluation of acceleration factor and storage life:

After the parameters a and b are obtained, the acceleration factor can be calculated according to formula (4), and the acceleration factor of a certain type of electronic machine under each accelerated temperature stress relative to normal temperature (25°C) can be obtained. The results are shown in Table 2. And according to formula (3), the storage characteristic life of the product at normal temperature (25°C) was calculated, and the result was 40.2 years.

Table 2 Acceleration factor results

It can be seen that the present invention has the following technical effects:

(1) When predicting the trend of linear or nonlinear degraded data, using support vector machine for modeling can make the trend of the predicted data consistent with the observed data, and the support vector machine is very convenient to use.

(2) For products that have been stored for a certain period of time, the acceleration factor is used to combine the natural use data with the accelerated test data, which can fully and reasonably utilize the natural storage data, improve the utilization rate of data resources, and make the evaluation results more accurate. high.

For the method implementation, for the sake of simple description, it is expressed as a series of action combinations, but those skilled in the art should know that the implementation of the present invention is not limited by the described action sequence, because according to the implementation of the present invention , certain steps may be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily necessary for the embodiments of the present invention.

FIG. 5 shows a schematic structural diagram of an apparatus for evaluating the storage life of an electronic complete machine according to an embodiment of the present invention. Referring to FIG. 5 , the apparatus includes: an acquisition module 510 , an evaluation module 520 , a processing module 530 and an optimization module 540 , wherein :

an acquisition module 510, used for acquiring performance degradation data and natural storage years of the electronic complete machine;

The evaluation module 520 is configured to construct a degradation data trend model according to the test samples in the performance degradation data, and obtain the predicted life of the electronic complete machine according to the degradation data trend model and the verification samples in the performance degradation data ;

a processing module 530, configured to obtain an acceleration factor according to the natural storage age and a pre-established acceleration model;

The optimization module 540 is configured to obtain the characteristic life of the electronic complete device according to the predicted life of the electronic complete device and the acceleration factor.

Each functional model in this embodiment is described in detail below:

The performance degradation data includes: multiple sets of product performance data and sample data of detection time corresponding to the product performance data;

The evaluation module 520 is configured to divide the performance degradation data into a test sample and a verification sample according to the detection time; use a support vector machine to establish an initial degradation data trend model; take the product performance data in the test sample as input vector, the performance degradation data value is an output vector, and the initial degradation data trend model is trained by using the least squares support vector machine algorithm to construct a degradation data trend model.

The evaluation module 520 is further configured to perform the prediction step according to the order of detection time;

The predicting step includes: taking the detection time in the first group of sample data in the verification sample as an input, and combining with the degradation data trend model to obtain the predicted value of the corresponding product performance parameter; judging the value of the product performance parameter. Whether the predicted value reaches the upper limit or lower limit of the product failure threshold; if so, the detection time corresponding to the predicted value of the product performance parameter is taken as the predicted life of the electronic complete machine.

The evaluation module 520 is further configured to analyze the trend of the degradation data according to the first set of sample data in the verification sample if it is determined that the predicted value of the product performance parameter does not reach the upper limit or the lower limit of the product failure threshold. The model is updated; the first group of sample data in the verification sample is deleted; and the prediction step is repeatedly performed until the predicted life of the electronic complete machine is obtained.

In a feasible example, the apparatus further includes: a modeling module;

The modeling module is configured to obtain the predicted life spans of multiple test electronic complete machines under different temperature stresses according to the temperature stress data included in the performance degradation data and in combination with the degradation data trend model; Build an accelerated model for the predicted life of the test electronic complete machine and the natural storage life of each test electronic complete machine;

Correspondingly, the processing module 530 is used to evaluate the parameters in the acceleration model according to the predicted life of a plurality of test electronic complete machines under different temperature stresses; calculate and obtain the electronic complete machine under different temperature stresses according to the acceleration model. acceleration factor.

It can be seen that the embodiment of the present invention analyzes the performance degradation data of the electronic complete machine, evaluates the life of the electronic complete machine according to the analysis results, and then analyzes the acceleration factor of the electronic complete machine based on the storage life of the electronic complete machine, and then combines the evaluated results. Lifetime and Acceleration Therefore, the characteristic lifetime of the electronic complete machine is evaluated, which has the advantage of high evaluation accuracy.

As for the apparatus implementation, since it is basically similar to the method implementation, the description is relatively simple, and for related parts, please refer to the partial description of the method implementation.

It should be noted that, in each component of the device of the present invention, the components are logically divided according to the functions to be implemented, but the present invention is not limited to this, and each component can be re-divided as required or a combination.

Various component embodiments of the present invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. In this device, the PC controls the device or device remotely through the Internet, and precisely controls each operation step of the device or device. The present invention can also be implemented as apparatus or apparatus programs (eg, computer programs and computer program products) for performing part or all of the methods described herein. The program implementing the present invention in this way can be stored on a computer-readable medium, and the files or documents generated by the program can be counted, generate data reports and cpk reports, etc., and can perform batch testing and statistics on power amplifiers. It should be noted that the above-described embodiments illustrate rather than limit the invention, and that alternative embodiments may be devised by those skilled in the art without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means may be embodied by one and the same item of hardware. The use of the words first, second, and third, etc. do not denote any order. These words can be interpreted as names.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, but not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that it can still be The technical solutions described in the foregoing embodiments are modified, or some technical features thereof are equivalently replaced; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (8)

1. A method for evaluating the storage life of an electronic complete machine, comprising: Obtain the performance degradation data and natural storage period of the electronic complete machine; Build a degradation data trend model according to the test samples in the performance degradation data, and obtain the predicted life of the electronic complete machine according to the degradation data trend model and the verification samples in the performance degradation data; Obtain the acceleration factor according to the natural storage period and the pre-established acceleration model; Obtain the characteristic life of the electronic complete device according to the predicted life of the electronic complete device and the acceleration factor; Wherein, before obtaining the acceleration factor according to the natural storage period and the pre-established acceleration model, the method further includes: According to the temperature stress data included in the performance degradation data, combined with the degradation data trend model, obtain the predicted lifespan of a plurality of test electronic complete machines under different temperature stresses; The acceleration model is constructed according to the predicted life of multiple test electronic complete machines under different temperature stress and the natural storage life of each test electronic complete machine; Correspondingly, the obtaining of the acceleration factor according to the natural storage period and the pre-established acceleration model includes: Evaluate the parameters in the acceleration model according to the predicted life of a plurality of test electronic complete machines under different temperature stresses; Acceleration factors of the electronic complete machine under different temperature stresses are obtained by calculating according to the acceleration model. 2. The method according to claim 1, wherein the performance degradation data comprises: a plurality of groups comprising product performance data and detection time sample data corresponding to the product performance data; Correspondingly, the building a degradation data trend model according to the test samples in the performance degradation data includes: dividing the performance degradation data into test samples and verification samples according to the detection time; Use support vector machine to establish the initial degradation data trend model; Taking the product performance data in the test sample as the input vector and the performance degradation data as the output vector, the initial degradation data trend model is trained by using the least squares support vector machine algorithm to construct the degradation data trend model. 3 . The method according to claim 2 , wherein, according to the degradation data trend model and the verification samples in the performance degradation data, obtaining the predicted life of the electronic complete machine comprises: 3 . According to the order of detection time, the prediction step is performed; The predicting step includes: Taking the detection time in the first group of sample data in the verification sample as input, and combining with the degradation data trend model, obtain the predicted value of the corresponding product performance parameter; Determine whether the predicted value of the product performance parameter reaches the upper limit or lower limit of the product failure threshold; If so, the detection time corresponding to the predicted value of the product performance parameter is taken as the predicted life of the electronic complete machine. 4. The method according to claim 3, wherein, if the predicted value of the product performance parameter does not reach the upper limit or the lower limit of the product failure threshold, then according to the first group of sample data in the verification sample Update the degraded data trend model described above; delete the first group of sample data in the verification sample; The predicting step is repeatedly performed until the predicted lifetime of the electronic complete machine is obtained. 5. A device for evaluating the storage life of an electronic complete machine, comprising: The acquisition module is used to acquire the performance degradation data and natural storage period of the electronic whole machine; an evaluation module, configured to construct a degradation data trend model according to the test samples in the performance degradation data, and obtain the predicted life of the electronic complete machine according to the degradation data trend model and the verification samples in the performance degradation data; a processing module, configured to obtain an acceleration factor according to the natural storage period and a pre-established acceleration model; an optimization module, used for obtaining the characteristic life of the electronic complete machine according to the predicted life of the electronic complete machine and the acceleration factor; Wherein, the device further includes: a modeling module; The modeling module is configured to obtain the predicted life spans of multiple test electronic complete machines under different temperature stresses according to the temperature stress data included in the performance degradation data and in combination with the degradation data trend model; Build an accelerated model for the predicted life of the test electronic complete machine and the natural storage life of each test electronic complete machine; Correspondingly, the processing module is used to evaluate the parameters in the acceleration model according to the predicted lifespan of a plurality of test electronic complete machines under different temperature stresses; calculate and obtain the electronic complete machine under different temperature stresses according to the acceleration model. acceleration factor. 6 . The device according to claim 5 , wherein the performance degradation data comprises: multiple sets of product performance data and detection time sample data corresponding to the product performance data; 6 . Correspondingly, the evaluation module is used to divide the performance degradation data into a test sample and a verification sample according to the detection time; use a support vector machine to establish an initial degradation data trend model; use the product performance data in the test sample is the input vector, the performance degradation data value is the output vector, and the initial degradation data trend model is trained by using the least squares support vector machine algorithm to construct the degradation data trend model. 7. The device according to claim 6, wherein the evaluation module is further configured to perform the prediction step according to the order of detection time; The predicting step includes: taking the detection time in the first group of sample data in the verification sample as an input, and combining with the degradation data trend model to obtain the predicted value of the corresponding product performance parameter; judging the value of the product performance parameter. Whether the predicted value reaches the upper limit or lower limit of the product failure threshold; if so, the detection time corresponding to the predicted value of the product performance parameter is taken as the predicted life of the electronic complete machine. 8. The device according to claim 7, wherein the evaluation module is further configured to determine that the predicted value of the product performance parameter does not reach the upper limit or the lower limit of the product failure threshold, according to the verification The first group of sample data in the sample updates the degradation data trend model; the first group of sample data in the verification sample is deleted; and the predicting step is repeatedly performed until the predicted life of the electronic complete machine is obtained.

Images