CN116596336B

CN116596336B - State evaluation method and device of electronic equipment, electronic equipment and storage medium

Info

Publication number: CN116596336B
Application number: CN202310562035.3A
Authority: CN
Inventors: 江海洋; 刘浩; 袁振华; 王骏荣; 钟权; 张学钢
Original assignee: Hefei Lianbao Information Technology Co Ltd
Current assignee: Hefei Lianbao Information Technology Co Ltd
Priority date: 2023-05-16
Filing date: 2023-05-16
Publication date: 2023-10-31
Anticipated expiration: 2043-05-16
Also published as: CN116596336A

Abstract

The application provides a state evaluation method and device of electronic equipment, the electronic equipment and a storage medium, wherein the method comprises the following steps: acquiring the probability of abnormality occurrence of each target component based on the current use data of each target component in at least one target component which has historically generated abnormality in the electronic device; determining a reference weight of each target component based on the historical anomaly data of each target component; obtaining current evaluation indexes of all target components according to the reference weight of each target component and the probability of abnormality occurrence at present, wherein the current evaluation indexes are used for representing the current health condition of the target components; the current state of the electronic device is evaluated based on the current evaluation index of each target component. Technical support is provided for realizing effective evaluation of the state of the electronic equipment.

Description

State evaluation method and device of electronic equipment, electronic equipment and storage medium

Technical Field

The present application relates to the field of data processing, and in particular, to a method and apparatus for evaluating a state of an electronic device, and a storage medium.

Background

In the related art, due to the lack of effective evaluation on the state of electronic equipment such as a notebook computer in the production stage, quality inspection records remained at a factory end before leaving a factory of many electronic equipment are qualified, but after entering the market, the condition of abnormal quality still occurs, and great after-sales repair cost is brought. How to effectively evaluate the state of an electronic device is a technical problem to be solved.

Disclosure of Invention

The application provides a state evaluation method and device of electronic equipment, the electronic equipment and a storage medium, and aims to at least solve the technical problems in the prior art.

According to a first aspect of the present application, there is provided a state evaluation method of an electronic device, the method comprising:

acquiring the probability of abnormality occurrence of each target component based on the current use data of each target component in at least one target component which has historically generated abnormality in the electronic device;

determining a reference weight of each target component based on the historical anomaly data of each target component;

obtaining current evaluation indexes of all target components according to the reference weight of each target component and the probability of abnormality occurrence at present, wherein the current evaluation indexes are used for representing the current health condition of the target components;

The current state of the electronic device is evaluated based on the current evaluation index of each target component.

In the above aspect, the determining the reference weight of each target component based on the historical abnormal data of each target component includes:

the reference weight of each target component is determined based on the historical anomaly data for each target component and the current usage data for each target component.

In the above solution, the obtaining the current evaluation index of each target component according to the reference weight of each target component and the probability of abnormality at the current time includes:

and obtaining the current evaluation index of each target component according to the reference weight of each target component, the current abnormal occurrence probability and the business weight value of each target component.

In the above aspect, the obtaining the probability of occurrence of the abnormality of each target component based on the current usage data of each target component in at least one target component in the electronic device, which has historically generated the abnormality, includes:

inputting the current use data of each target component into a detection model of each target component to obtain a detection result of the current use data of each target component, wherein the detection result is used for representing the probability of abnormality of each target component at present; the detection model of each target component is obtained by training a model to be trained of each target component by sample data with normal sample labels and sample data with abnormal sample labels of each target component.

In the above scheme, the method further comprises:

preprocessing current use data of each target component according to the data characteristics of each target component to obtain target data of each target component;

inputting the current usage data of each target component into a detection model of each target component to obtain a detection result of the current usage data of each target component, wherein the detection result comprises the following steps:

and inputting the target data of each target component into a detection model of each target component to obtain a detection result of the target data of each target component.

In the above solution, the evaluating the current state of the electronic device based on the current evaluation index of each target component includes:

obtaining a state measurement value of the electronic equipment based on the current evaluation index of each target component;

and evaluating the current state of the electronic equipment according to the state metric value of the electronic equipment.

In the above solution, the detection model of each target component is obtained by training a model to be trained of each target component by using sample data with a normal sample tag and sample data with an abnormal sample tag of each target component, and includes:

inputting sample data with normal sample labels and sample data with abnormal sample labels of all target components to the to-be-trained model of all target components, and training the to-be-trained model of all target components;

And each detection model obtained by training the model to be trained of each target component is used for detecting whether the current use data of each target component is abnormal or not.

In the above scheme, the inputting the sample data with the normal sample label and the sample data with the abnormal sample label of each target component to the model to be trained of each target component, training the model to be trained of each target component includes:

preprocessing sample data with normal sample labels and sample data with abnormal sample labels of all target components according to the data characteristics of all target components to obtain target sample data of all target components;

and inputting the target sample data of each target component into the model to be trained of each target component, and training the model to be trained of each target component.

According to a second aspect of the present application, there is provided a state evaluation apparatus of an electronic device, the apparatus comprising:

a first acquisition unit configured to acquire a probability that an abnormality occurs at a present time for each target component based on present use data of each target component of at least one target component in the electronic device that has historically produced the abnormality;

a determining unit configured to determine a reference weight of each target component based on the history abnormality data of each target component;

The second acquisition unit is used for obtaining the current evaluation index of each target component according to the reference weight of each target component and the probability of abnormality at present, wherein the current evaluation index is used for representing the current health condition of the target component;

and the evaluation unit is used for evaluating the current state of the electronic equipment based on the current evaluation index of each target component.

According to a third aspect of the present application, there is provided an electronic device comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein, the liquid crystal display device comprises a liquid crystal display device,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the methods of the present application.

According to a fourth aspect of the present application there is provided a non-transitory computer readable storage medium storing computer instructions for causing a computer to perform the method of the present application.

In the application, the current state of the electronic equipment can be evaluated based on the current use data of each target component in at least one target component which has historically generated abnormality in the electronic equipment and the historical abnormality data of each target component. A unified and global state evaluation system is constructed, so that the state of the electronic equipment is effectively evaluated, the quality of the electronic equipment is ensured, and the after-sale maintenance cost is greatly reduced.

It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the application or to delineate the scope of the application. Other features of the present application will become apparent from the description that follows.

Drawings

The above, as well as additional purposes, features, and advantages of exemplary embodiments of the present application will become readily apparent from the following detailed description when read in conjunction with the accompanying drawings. Several embodiments of the present application are illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings and in which:

in the drawings, the same or corresponding reference numerals indicate the same or corresponding parts.

Fig. 1 shows a schematic implementation flow diagram of a state evaluation method of an electronic device according to an embodiment of the present application;

fig. 2 shows a second implementation flow chart of a state evaluation method of an electronic device according to an embodiment of the present application;

fig. 3 is a schematic diagram showing a composition structure of a state evaluation device of an electronic apparatus according to an embodiment of the present application;

fig. 4 is a schematic diagram showing a composition structure of an electronic device according to an embodiment of the present application.

Detailed Description

In order to make the objects, features and advantages of the present application more comprehensible, the technical solutions according to the embodiments of the present application will be clearly described in the following with reference to the accompanying drawings, and it is obvious that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

In the related art, in a production stage of an electronic device such as a notebook computer, a plurality of tests are generally performed on the electronic device to ensure that the electronic device meets a factory standard. Through simulation of a plurality of user use scenes, the test results of the electronic equipment which can be delivered or not can be delivered are given. If the test result indicates that the electronic equipment can leave the factory, the electronic equipment is normally off-line and leaves the factory. If the test result indicates that the electronic equipment cannot be delivered, multiple business strategies such as repeated testing and maintenance and replacement are executed on the electronic equipment, and all delivered electronic equipment is ensured to pass the quality test. But this approach can only avoid some qualitative quality problems such as failure to light. The generated and retained data has extremely low guiding significance on analyzing the problems of the after-sale abnormal electronic equipment, and the overall state of the electronic equipment cannot be monitored or predicted. And all the test items are independent and have no global association, and independent test results are respectively independent as trigger conditions of quality control operation, so that a unified and global state evaluation system cannot be constructed.

In the related art, statistical analysis is performed on the electronic equipment with abnormality at the after-sale stage of the electronic equipment, and the statistical result is fed back to the production end to check the abnormal condition of the equipment at the production stage. This scheme has an after-market data feedback period of up to several months from the beginning to the basic end, and the process requires human effort to continually collect and feed back large amounts of data, requiring significant human effort and time costs. In addition, factory side electronics and production line configurations may have changed significantly while waiting for feedback. Aiming at the statistical result of the historical data, the state of the electronic equipment cannot be effectively evaluated dynamically. If the state of the electronic equipment can be effectively evaluated in the production stage, a great deal of manpower and time cost can be saved, the quality of the electronic equipment is ensured, and the after-sale maintenance cost is further reduced.

In the embodiment of the application, the probability of abnormality occurrence of each target component is obtained based on the current use data of each target component in at least one target component which has historically generated abnormality in the electronic equipment; determining a reference weight of each target component based on the historical anomaly data of each target component; obtaining current evaluation indexes of all target components according to the reference weight of each target component and the probability of abnormality occurrence at present, wherein the evaluation indexes are used for representing the health condition of the target components; the current state of the electronic device is evaluated based on the current evaluation index of each target component. A unified and global state evaluation system is constructed, so that the state of the electronic equipment is effectively evaluated, and the quality of the electronic equipment is ensured.

The method for evaluating the state of the electronic device according to the embodiment of the application is described in detail below.

An embodiment of the present application provides a method for evaluating a state of an electronic device, as shown in fig. 1, where the method includes:

s101: based on current usage data of each of at least one target component in the electronic device that has historically generated an abnormality, a probability of each target component being abnormal at the current time is obtained.

In the present application, two relative time concepts are involved, one being historical and one being current. On the time axis, the historical time is before and the current time is after. It will be appreciated that a target component that has historically developed an anomaly may be considered a component that has developed an anomaly at a time prior to the current time. The probability of the abnormality of the target component at the present time may be regarded as detecting the probability of whether the abnormality of the target component occurs at the time when it is desired to detect whether the abnormality of the target component occurs.

In this step, the electronic device contains many components inside. Taking an electronic device as a notebook computer as an example, the notebook computer internally comprises a magnetic disk, a CPU (Central Processing Unit, a central processing unit), a GPU (graphics processing unit, a graphics processor), a memory, a main board, a battery and other components. The target component is a component that has historically been abnormal in notebook computers, as opposed to the current ones. The number of the target members may be one or two or more. Based on the after-market factory side feedback, it is possible to know which part or parts of the notebook computer have been abnormal in history, and thus, the target part can be determined from all parts of the notebook computer.

It is understood that the target components of the notebook computer may be all components in which an abnormality has occurred, or may be part of components in which an abnormality has occurred. For example, if the battery is abnormal in the notebook computer, but the number of times of abnormality is only one or two times, the number of times is small, and the proportion of times of abnormality occurrence of other components is relatively high, the battery may not be used as a target component for state evaluation of the notebook computer. That is, the target component may be a component in which the electronic device has historically been abnormal and the number of times of occurrence of the abnormality is high, such as higher than the first set threshold.

The usage data of the target component refers to data generated at the time of the target component. For example, taking a target component as a CPU, the usage data of the CPU includes data such as a minimum frequency of the CPU. Taking the target component as a GPU as an example, the use data of the GPU comprises data such as GPU fundamental frequency design standards and the like. If the time when the state of the electronic device is desired to be evaluated is regarded as the current time, the minimum frequency of the CPU used at the time when the state is desired to be evaluated can be regarded as the current usage data of the target component, such as the CPU. The GPU baseband design criteria data may be considered as current usage data for the GPU, the target component.

In general, when the state of the electronic device is evaluated in the production stage of the electronic device, the current usage data of the target component is the current usage data of the target component of the electronic device to be evaluated in the production stage (such as the data of the minimum CPU frequency, the GPU fundamental frequency design standard, etc. of the notebook computer in the production stage). Based on the current usage data of each target component that has historically generated an abnormality, the probability of each target component that is currently generating an abnormality is obtained.

The technical scheme of obtaining the probability of abnormality of each target component at present according to the current use data of the target component which has generated abnormality in history can be particularly referred to other description.

S102: the reference weight of each target component is determined based on the historical anomaly data for each target component.

In this step, the reference weight of each target component is the same as the abnormal weight at which abnormality has occurred in each target component in history. Similarly to the case where the sum of the abnormality weights of the target parts, which have historically been abnormal, is equal to 1, the sum of the reference weights of the target parts is also equal to 1. And obtaining the times of occurrence of historic abnormality of each target component according to the historic abnormality data counted in the after-sales stage. The abnormality weight of each target component that has historically been abnormal is determined by the number of times each target component has historically been abnormal.

Illustratively, the historical exception data for the CPU includes a CPU minimum frequency exception, assuming that the after-market stage counts a total number of times that all target components have historically been abnormal to 100 times, and wherein the number of times that the CPU has been abnormal is 40 times, the abnormality weight for the CPU has historically been abnormal is 40/100=0.4, the number of times that the GPU has been abnormal is 30 times, and the abnormality weight for the GPU has historically been 30/100=0.3.

S103: and obtaining the current evaluation index of each target component according to the reference weight of each target component and the probability of abnormality occurrence at present, wherein the current evaluation index is used for representing the current health condition of the target component.

In this step, the current evaluation index of each target component may be a value for characterizing the health of each target component. Specifically, taking a certain target component as an example, the current evaluation index of the target component may be a product of the reference weight of the target component and the probability that the target component is abnormal at present.

And obtaining the health condition of each current target component in the electronic equipment based on the current evaluation index of each target component. For example, when the current evaluation index of the target component is a higher value, the target component is in an unhealthy state to a higher degree under the current condition. When the current evaluation index of the target component is a small value, the target component is in a low degree of unhealthy under the current condition. I.e. to a higher degree of health. Wherein the current evaluation index is relatively high or low. For example, assuming that three electronic devices (such as notebook computers) are currently being subjected to state evaluation, the value of the current evaluation index of the first electronic device CPU is 15, the value of the current evaluation index of the second electronic device CPU is 13, and the value of the current evaluation index of the third electronic device CPU is 19. The values of the current evaluation indexes of the first and third electronic devices CPU are higher than those of the second electronic device CPU, and accordingly, the values of the current evaluation indexes of the second electronic device CPU are lower than those of the first and third electronic devices CPU. Still further, the value of the current evaluation index of the third electronic device CPU is a higher value relative to the first electronic device, and correspondingly, the value of the current evaluation index of the first electronic device CPU is a lower value relative to the third electronic device. That is, by comparing the current evaluation indexes of the same target component and different electronic devices, it can be determined that the target component is unhealthy under the current condition, and if the value of the current evaluation index of the third electronic device CPU is the highest, the third electronic device CPU is unhealthy under the current condition. And if the value of the current evaluation index of the second electronic equipment CPU is the lowest, the second electronic equipment CPU is in the lowest unhealthy degree under the current condition. The value of the current evaluation index of the first electronic equipment CPU is medium, and the first electronic equipment CPU is in unhealthy degree under the current condition.

S104: the current state of the electronic device is evaluated based on the current evaluation index of each target component.

In this step, the current evaluation index of each target component represents the health condition of each target component in the electronic device, and the current overall state of the electronic device can be evaluated according to the current health condition of each target component.

In the steps S101-S104, the probability of abnormality occurrence of each target component is obtained based on the current use data of each target component in at least one target component which has historically generated abnormality in the electronic equipment; determining a reference weight of each target component based on the historical anomaly data of each target component; obtaining current evaluation indexes of all target components according to the reference weight of each target component and the probability of abnormality occurrence at present, wherein the current evaluation indexes are used for representing the current health condition of the target components; the current state of the electronic device is evaluated based on the current evaluation index of each target component. A unified and global state evaluation system is constructed, so that the state of the electronic equipment is effectively evaluated, and the delivery quality of the electronic equipment is ensured.

In an alternative solution, the determining the reference weight of each target component based on the historical anomaly data of each target component includes:

In the application, the reference weight of each target component can also be determined by the historical abnormal data of each target component and the current use data of each target component. The evaluation of the status of the electronic device is typically performed during a production phase of the electronic device, i.e. the current usage data of each target component of the electronic device may be usage data of each target component of the electronic device during the production phase. The abnormality data of each target part in the after-sales stage can be used as the history abnormality data of each target part of the electronic device with respect to each target part in the production stage. Illustratively, if the total number of historic data anomalies for all target components counted at the after-market stage is 100, the number of historic anomalies for the data of the CPU is 40, and the number of historic anomalies for the data of the GPU is 30. If the CPU current usage data operates normally (i.e. is currently used normally), the GPU current usage data is abnormal (i.e. is used abnormally currently), and the total number of data abnormality of all target components is 101 up to the present. Of the total number of anomalies of 101, the number of anomalies of the CPU accounts for 40 times, and the reference weight of the CPU is 40/101=0.396. The number of exceptions of the GPU is 30+1=31 times, and the reference weight of the GPU is 31/101=0.307. With respect to the aforementioned reference weights of each target component determined only by the historical abnormality data of each target component, if the target component whose current usage data is abnormal in operation occurs, the reference weight values of both the CPU and the GPU are caused to change.

According to the method, the historical abnormal data and the current use data of each target component are combined, the reference weight of each target component is determined, the influence of the historical abnormal data and the current use data on the reference weight of the target component is considered, the accuracy of the reference weight of each target component can be ensured, and the method is simple and easy to implement.

The calculation of the reference weight is performed by combining the current actual use condition of the target component, so that the accuracy of the calculation of the reference weight can be ensured.

In an alternative solution, as shown in fig. 2, the obtaining, according to the reference weight of each target component and the probability of abnormality occurring at present, a current evaluation index of each target component includes:

s203: and obtaining the current evaluation index of each target component according to the reference weight of each target component, the current abnormal occurrence probability and the business weight value of each target component.

In the application, the current evaluation index of each target component is jointly determined by the reference weight of each target component, the probability of abnormality occurrence currently and the business weight value of each target component. Specifically, in the case where the reference weight, the probability of occurrence of an abnormality at present, and the business weight value are all numerical values, the current evaluation index of each target component may be a product of the reference weight of each target component, the probability of occurrence of an abnormality at present, and the business weight value of each target component.

Wherein the business weight value represents the importance of each target component in the business layer in the electronic equipment. The service weight value of each target component may be a preset value. For example, as a core component for computing and controlling an electronic device such as a notebook computer, a CPU naturally occupies a higher service weight value. For example, the traffic weight of the CPU is set to 0.5. Other components, such as a battery, a motherboard, etc., are low relative to the CPU, naturally, the service weight of these components is set to a low value, such as the service weight value of the battery is set to 0.1, and the service weight value of the motherboard is set to 0.2. The product of the reference weight of each target component, the current abnormal probability and the business weight value of each target component is adopted as the current evaluation index of each target component, so that the current evaluation index of the target component with high business weight value in the electronic equipment can be improved, and the state evaluation of the whole electronic equipment tends to be influenced on the target component with larger influence on the electronic equipment. Therefore, the electronic equipment which is possibly abnormal after leaving the factory can be better intercepted and tracked in the production stage, and the risk of abnormality of the electronic equipment after leaving the factory is reduced.

In the application, the current evaluation index of each target component is obtained based on the reference weight of each target component, the current occurrence probability of abnormality and the business weight value of each target component. Considering the influence of three factors, namely the reference weight, the current abnormal probability and the business weight value, on the evaluation index, more refined health degree measurement and monitoring can be carried out on each target component, and further effective evaluation on the state of the electronic equipment can be realized.

In an alternative solution, the obtaining, based on current usage data of each target component in at least one target component that has historically generated an abnormality in the electronic device, a probability that each target component is abnormal at the current time includes:

In the application, each target component has a corresponding detection model. The detection model corresponding to each target component is used for detecting the current use data of each target component, so that the probability of abnormality of each target component is predicted by the detection model. The detection models corresponding to the target components are obtained by training the to-be-trained models of the target components by sample data of the target components with normal sample labels and sample data of the target components with abnormal sample labels.

The history data of each target component is taken as sample data of each target component. According to the abnormal feedback of the after-sales stage, the historical data with the abnormality is marked as an abnormal sample label, and the historical data without the abnormality is marked as a normal sample label. The historical data with the normal sample labels and the historical data with the abnormal sample labels of all the target components are input into the to-be-trained models corresponding to all the target components, so that the to-be-trained models corresponding to all the target components are trained.

In the training scheme, the loss function value generated in the training process can be calculated, and when the loss function value is smaller than or equal to a set threshold value, the training is finished. The corresponding model to be trained of each target component at the end of training can be used as the detection model of each target component.

In the applied scheme, current usage data of each target component is input into a detection model of each target component, and the probability of occurrence of abnormality of each target component is output from the detection model of each target component. Illustratively, when the probability of the occurrence of an abnormality of the CPU currently output by the detection model of the CPU is 0.8, the probability of the occurrence of an abnormality of the CPU currently is high. Alternatively, the CPU may be currently in a high probability of abnormality. When the probability of the GPU which is output by the GPU detection model and is abnormal is 0.2, the probability of the GPU which is abnormal is lower. Alternatively, the GPU is not currently in an abnormal state with a high probability. Thereby obtaining the detection result of whether the abnormality of each target component is possible to happen at present.

According to the application, the probability of abnormality of each target component is predicted by using the detection model of each target component, so that the accuracy of a detection result is ensured, and the real-time monitoring and real-time adjustment of each target component are realized. Each target component detects the probability of abnormality at present through a corresponding detection model, so that targeted detection is realized, and the detection accuracy of each target component can be ensured.

In an alternative, the method further comprises:

Correspondingly, the step of inputting the current usage data of each target component into the detection model of each target component to obtain a detection result of the current usage data of each target component includes:

In the application, the current use data of each target component can be preprocessed, and the preprocessed current use data of each target component is input into the detection model of each target component to obtain the probability of abnormality of each target component. Since the data dimensions of the target components are different, the data characteristics corresponding to the data dimensions of the target components are also different. Naturally, the way in which the current usage data of each target component is preprocessed will also vary.

Illustratively, the data dimensions of the CPU include CPU minimum frequency, CPU average power consumption, and the like. For the component of the CPU, the data dimensions of the CPU have differences in the value range, and the data characteristics corresponding to the data dimensions of the CPU are as follows: the data dimensions have large differences in the value ranges. And performing characteristic scaling, such as regularization, on the data dimensions of the CPU to eliminate the influence caused by the large difference of the value ranges among the data dimensions. And (3) inputting the dimension data of the CPU subjected to regularization processing as target data of the CPU into a detection model corresponding to the CPU, and obtaining a detection result of the target data of the CPU.

The data dimensions of the motherboard include an S3 related data dimension (computer standby wake-up data), an S4 related data dimension (computer sleep wake-up data), an Audio test data dimension, a USB (Universal Serial Bus ) test data dimension, and the like. Because the value range difference of each data dimension of the main board is smaller, the data dimension of the main board corresponds to the data characteristics that: the data dimensions tend to be uniform over the range of values, so that no regularization preprocessing steps such as component CPU are required. Compared with a CPU, the service weight of the main board is lower, so that the data dimension used for detection of the main board can be all data dimensions of the main board or part of data dimensions obtained by screening according to actual service requirements. Therefore, for the target component of the motherboard, a preprocessing step of dimension screening is required for the motherboard, and the useful data dimension of the motherboard is screened out and used as target data of the motherboard to be input into a detection model corresponding to the motherboard, so as to obtain a detection result of the target data of the motherboard.

In popular terms, for a target component with high service weight, regularization processing may be used to preprocess the target component to obtain target data that may be input to a corresponding detection model. For target components with low or medium business weights, screening can be used to preprocess the target components to obtain target data which can be input into a corresponding detection model and is characterized by specific dimensions.

According to the application, the current use data of each target component is preprocessed according to the data characteristics of different target components, and the preprocessed current use data is used as the input of the detection model of each target component to further obtain the detection result, so that the accuracy of the detection result is ensured, and the validity and accuracy of the overall state evaluation of the electronic equipment are further ensured.

In an optional solution, the evaluating the current state of the electronic device based on the current evaluation index of each target component includes:

In the application, the current state of the electronic equipment is evaluated according to the state measurement value of the electronic equipment. Wherein the state metric of the electronic device is derived based on the current evaluation index of each target component. Specifically, the operational relationship between the state metric value of the electronic device and the current evaluation index of each target component is shown in formula (1):

formula (1)

Wherein, the liquid crystal display device comprises a liquid crystal display device,representing the state metric of the electronic device. / >Representing the probability that the ith target part is currently abnormal. />Representing the reference weight of the i-th target component. />Representing the traffic weight value of the i-th target component. />Representing the current evaluation index of the i-th target component. />Representing the sum of the current evaluation indexes of each target component.Representing the sum of the reference weights of the respective target components. i is an integer greater than or equal to 1, representing the total number of target parts in the electronic device.

And (3) obtaining a state metric value of the electronic equipment through the formula (1). And comparing the state metric value of the electronic equipment with a preset state threshold value to obtain a current state evaluation result of the electronic equipment. For example, when the state metric value of the electronic device is greater than the preset state threshold value, the current state of the electronic device is indicated to be healthy. And when the state measurement value of the electronic equipment is smaller than the preset state threshold value, the current state of the electronic equipment is unhealthy. When the state metric value of the electronic device is equal to the preset state threshold value, the current state of the electronic device is represented as sub-health, and the like.

For example, assuming that the preset state threshold of the electronic device is 20, when the state metric of the electronic device is 15, and at this time, the state metric of the electronic device is smaller than the preset state threshold, the current state of the electronic device is unhealthy. When the state metric value of the electronic equipment is 25, and at the moment, the state metric value of the electronic equipment is larger than a preset state threshold value, the current state of the electronic equipment is healthy. When the state measurement value of the electronic equipment is just 20, and at the moment, the state measurement value of the electronic equipment is equal to the preset state threshold value, the current state of the electronic equipment is sub-healthy. In the foregoing, the description is given taking the current state of the electronic device including healthy, unhealthy and sub-healthy as an example. The specific state evaluation criteria may also include other ways, which the present application does not specifically limit.

According to the application, the current state of the electronic equipment is evaluated through the state metric value of the electronic equipment, so that whether the current state of the electronic equipment is good or not can be intuitively reflected, the high-risk electronic equipment can be conveniently intercepted or retested in time in the production stage, the low-risk electronic equipment is tracked and spot inspected, and the risk of abnormality of the electronic equipment after leaving a factory is reduced.

In an alternative solution, the detection model of each target component is obtained by training a model to be trained of each target component by using sample data with normal sample labels and sample data with abnormal sample labels of each target component, and the method includes:

In the application, each target component has a corresponding model to be trained, and after training, a detection model corresponding to each target component is obtained. A Light GBM (Light Gradient Boosting Machine, gradient elevator) machine learning algorithm model based on a lifting method can be selected as a model to be trained for each target component. In the training process of the model to be trained, super parameters such as feature sub-sampling, leaf node number, maximum depth of a tree structure and the like are utilized to perform super parameter tuning on the model to be trained so as to minimize a loss function of the model to be trained. And under the condition of minimizing the loss function, the training of the model to be trained is completed or ended.

The trained model to be trained comprehensively evaluates the performance quality of the trained model to be trained through three evaluation indexes, namely Precision, recall and a blending average value F1 score of the Precision and the Recall. The calculation modes of the three evaluation indexes are shown in the formula (2), the formula (3) and the formula (4) respectively:

Precision=formula (2)

Recall=Formula (3)

Formula (4)

Wherein, the liquid crystal display device comprises a liquid crystal display device,and the real normal sample data are represented, and the trained model to be trained is judged to be the number of the normal sample data. />And the actual abnormal sample data is represented, and the trained model to be trained is judged to be the number of the abnormal sample data. />And the number of the abnormal sample data is judged by the trained model to be trained. The higher the values of three evaluation indexes, namely Precision, recall and a blending average value F1 score of the Precision and Recall, the better the performance of the detection model obtained by training the model to be trained, and the more accurate the detection result.

According to the application, the model to be trained of each target component is trained through the sample data of each target component with the normal label and each target component with the abnormal label, and the performance quality of the model is obtained through evaluation training by adopting the evaluation index, so that the accuracy of the training result of the model to be trained is improved, and a foundation is provided for effectively evaluating the state of the electronic equipment.

In an alternative solution, the inputting the sample data with the normal sample label and the sample data with the abnormal sample label of each target component to the model to be trained of each target component, and training the model to be trained of each target component includes:

In the present application, sample data of each target component may be preprocessed. And inputting the sample data of each target component after pretreatment into the model to be trained of each target component so as to train the model to be trained of each target component. Since the sample data dimensions of each target component are different, the data characteristics corresponding to the sample data dimensions of each target component will also be different. Naturally, the manner in which the sample data of each target part is preprocessed will also vary.

Illustratively, the sample data dimension of the CPU includes CPU minimum frequency, CPU average power consumption, and the like. For the component of the CPU, the sample data dimensions of the CPU have differences in the value range, and the data characteristics corresponding to the sample data dimensions of the CPU are as follows: the dimensions of the sample data have large differences in the value range. And performing characteristic scaling, such as regularization, on the sample data dimensions of the CPU to eliminate the influence caused by the large difference of the value ranges among the sample data dimensions. And (3) inputting the dimension sample data of the CPU subjected to regularization treatment as target sample data of the CPU to a model to be trained corresponding to the CPU, and training the model to be trained corresponding to the CPU.

The sample data dimension of the motherboard includes an S3 related data dimension (computer standby wake-up data), an S4 related data dimension (computer sleep wake-up data), an Audio (Audio) test data dimension, a USB test data dimension, and the like. Because the value range difference of each sample data dimension of the main board is smaller, the data characteristics corresponding to the sample data dimension of the main board are as follows: the data dimensions tend to be uniform over the range of values, so that no regularization preprocessing steps such as component CPU are required. Compared with a CPU, the service weight of the main board is lower, so that the sample data dimension of the main board for training can be all sample data dimensions of the main board, or can be part of sample data dimensions obtained by screening according to actual service requirements. Therefore, for the target component of the motherboard, the dimension screening of the motherboard is required, the useful sample data dimension of the motherboard is screened out and used as target sample data of the motherboard to be input into the model to be trained corresponding to the motherboard, and the model to be trained corresponding to the motherboard is trained.

In popular terms, for a target component with high service weight, regularization processing can be adopted to preprocess the target component so as to obtain target sample data which can be input into a corresponding model to be trained. For target components with low or medium service weight, screening can be adopted to preprocess the target components so as to obtain target sample data which can be input into the corresponding model to be trained and is characterized by specific dimension.

According to the application, the sample data of each target component is preprocessed according to the sample data characteristics of different target components, and the preprocessed target sample data is used as the input of the model to be trained of each target component so as to train the model to be trained, thereby ensuring the accuracy of the training result and further ensuring the validity and accuracy of the overall state evaluation of the electronic equipment.

An embodiment of the present application provides a state evaluation device of an electronic apparatus, as shown in fig. 3, where the device includes:

a first obtaining unit 301 configured to obtain, based on current usage data of each of at least one target component in the electronic device that has historically generated an abnormality, a probability that each target component is at a current occurrence of an abnormality;

a determining unit 302 for determining a reference weight of each target component based on the history abnormality data of each target component;

a second obtaining unit 303, configured to obtain, according to the reference weight of each target component and the probability of occurrence of an abnormality at present, a current evaluation index of each target component, where the current evaluation index is used to characterize a current health condition of the target component;

an evaluation unit 304, configured to evaluate the current state of the electronic device based on the current evaluation index of each target component.

In an alternative, the determining unit 302 is configured to determine the reference weight of each target component based on the historical anomaly data of each target component and the current usage data of each target component.

In an alternative solution, the second obtaining unit 303 is configured to obtain the current evaluation index of each target component according to the reference weight of each target component, the probability of occurrence of the abnormality currently, and the service weight value of each target component.

In an optional solution, the first obtaining unit 301 is configured to input current usage data of each target component into a detection model of each target component, to obtain a detection result of the current usage data of each target component, where the detection result is used to characterize a probability that each target component is abnormal at present; the detection model of each target component is obtained by training a model to be trained of each target component by sample data with normal sample labels and sample data with abnormal sample labels of each target component.

In an alternative solution, the first obtaining unit 301 is configured to pre-process current usage data of each target component according to a data feature of each target component, so as to obtain target data of each target component; and inputting the target data of each target component into a detection model of each target component to obtain a detection result of the target data of each target component.

In an alternative solution, the evaluation unit 304 is configured to obtain a state metric value of the electronic device based on the current evaluation index of each target component; and evaluating the current state of the electronic equipment according to the state metric value of the electronic equipment.

In an alternative solution, the first obtaining unit 301 is configured to input, to a model to be trained of each target component, sample data with a normal sample tag and sample data with an abnormal sample tag of each target component, and train the model to be trained of each target component; and each detection model obtained by training the model to be trained of each target component is used for detecting whether the current use data of each target component is abnormal or not.

In an alternative solution, the first obtaining unit 301 is configured to pre-process, according to the data characteristics of each target component, sample data with a normal sample tag and sample data with an abnormal sample tag of each target component, so as to obtain target sample data of each target component; and inputting the target sample data of each target component into the model to be trained of each target component, and training the model to be trained of each target component.

It should be noted that, in the state evaluation device for an electronic device according to the embodiment of the present application, since the principle of solving the problem of the state evaluation device for an electronic device is similar to that of the state evaluation method for an electronic device described above, the implementation process, implementation principle and beneficial effects of the state evaluation device for an electronic device can be referred to the description of the implementation process, implementation principle and beneficial effects of the method described above, and the repetition is omitted.

According to an embodiment of the present application, the present application also provides an electronic device and a readable storage medium.

Fig. 4 shows a schematic block diagram of an example electronic device 400 that may be used to implement an embodiment of the application. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the applications described and/or claimed herein.

As shown in fig. 4, the electronic device 400 includes a computing unit 401 that can perform various suitable actions and processes according to a computer program stored in a Read Only Memory (ROM) 402 or a computer program loaded from a storage unit 408 into a Random Access Memory (RAM) 403. In the RAM 403, various programs and data required for the operation of the electronic device 400 may also be stored. The computing unit 401, ROM 402, and RAM 403 are connected to each other by a bus 404. An input/output (I/O) interface 405 is also connected to bus 404.

Various components in electronic device 400 are connected to I/O interface 405, including: an input unit 406 such as a keyboard, a mouse, etc.; an output unit 407 such as various types of displays, speakers, and the like; a storage unit 408, such as a magnetic disk, optical disk, etc.; and a communication unit 409 such as a network card, modem, wireless communication transceiver, etc. The communication unit 409 allows the electronic device 400 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.

The computing unit 401 may be a variety of general purpose and/or special purpose processing components having processing and computing capabilities. Some examples of computing unit 401 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 401 performs the respective methods and processes described above, for example, a state evaluation method of the electronic device. For example, in some embodiments, the state evaluation method of the electronic device may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as the storage unit 408. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 400 via the ROM 402 and/or the communication unit 409. When the computer program is loaded into RAM 403 and executed by computing unit 401, one or more steps of the state evaluation method of the electronic device described above may be performed. Alternatively, in other embodiments, the computing unit 401 may be configured to perform the state evaluation method of the electronic device by any other suitable means (e.g. by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuitry, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems-on-a-chip (SOCs), complex Programmable Logic Devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for carrying out methods of the present application may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of the present application, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here 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 a user; and a keyboard and pointing device (e.g., a mouse or trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), and the internet.

The computer system may include a client and a server. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server of a distributed system, or a server incorporating a blockchain.

It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present application may be performed in parallel, sequentially, or in a different order, so long as the desired results of the technical solution disclosed in the present application can be achieved, and are not limited herein.

Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In the description of the present application, the meaning of "a plurality" is two or more, unless explicitly defined otherwise.

The foregoing is merely illustrative of the present application, and the present application is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A method for evaluating a state of an electronic device, the method comprising:

preprocessing current use data of each target component according to the data characteristics of each target component to obtain target data of each target component; inputting the target data of each target component into a detection model of each target component to obtain a detection result of the target data of each target component; the detection result is used for representing the probability of abnormality of each target component at present;

determining a reference weight of each target component based on the historical anomaly data of each target component; the reference weight of each target component is an abnormal weight of each target component which is abnormal in history;

evaluating the current state of the electronic equipment based on the current evaluation index of each target component;

wherein the determining the reference weight of each target component based on the historical abnormal data of each target component comprises: obtaining the number of times that each target component has historically been abnormal based on the historical abnormality data of each target component; an anomaly weight for each target component that has historically been anomalous is determined based on the number of times each target component has historically been anomalous and the total number of times each target component has historically been anomalous.

2. The method of claim 1, wherein determining the reference weight for each target part based on historical anomaly data for each target part comprises:

3. The method according to claim 1, wherein the obtaining the current evaluation index of each target component according to the reference weight of each target component and the probability of abnormality at the current time includes:

4. The method according to claim 1, wherein the detection model of each target component is obtained by training a model to be trained of each target component from sample data with a normal sample tag and sample data with an abnormal sample tag of each target component.

5. The method of claim 1, wherein evaluating the current state of the electronic device based on the current evaluation index of each target component comprises:

6. The method according to claim 4, wherein the detection model of each target component is obtained by training a model to be trained of each target component from sample data with a normal sample tag and sample data with an abnormal sample tag of each target component, comprising:

7. The method of claim 6, wherein inputting the normal sample tagged sample data and the abnormal sample tagged sample data for each target part to the model to be trained for each target part, training the model to be trained for each target part, comprises:

8. A state evaluation apparatus of an electronic device, the apparatus comprising:

a determining unit configured to determine a reference weight of each target component based on the history abnormality data of each target component; the reference weight of each target component is an abnormal weight of each target component which is abnormal in history;

The evaluation unit is used for evaluating the current state of the electronic equipment based on the current evaluation index of each target component;

the first acquisition unit is used for preprocessing the current use data of each target component according to the data characteristics of each target component to obtain target data of each target component; inputting the target data of each target component into a detection model of each target component to obtain a detection result of the target data of each target component; the detection result is used for representing the probability of abnormality of each target component at present;

the determining unit is used for obtaining the times of occurrence of abnormality of each target component in history based on the history abnormality data of each target component; an anomaly weight for each target component that has historically been anomalous is determined based on the number of times each target component has historically been anomalous and the total number of times each target component has historically been anomalous.

9. An electronic device, comprising:

at least one processor; and

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-7.

10. A non-transitory computer readable storage medium storing computer instructions for causing a computer to perform the method of any one of claims 1-7.