CN113656238B - AI industrial application capability test method and system of intelligent terminal - Google Patents
AI industrial application capability test method and system of intelligent terminal Download PDFInfo
- Publication number
- CN113656238B CN113656238B CN202110944557.0A CN202110944557A CN113656238B CN 113656238 B CN113656238 B CN 113656238B CN 202110944557 A CN202110944557 A CN 202110944557A CN 113656238 B CN113656238 B CN 113656238B
- Authority
- CN
- China
- Prior art keywords
- intelligent terminal
- temperature change
- parameters
- equipment
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/22—Detection or location of defective computer hardware by testing during standby operation or during idle time, e.g. start-up testing
- G06F11/26—Functional testing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/22—Detection or location of defective computer hardware by testing during standby operation or during idle time, e.g. start-up testing
- G06F11/2273—Test methods
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
- Investigating Or Analyzing Materials Using Thermal Means (AREA)
Abstract
An AI industrial application capability test method and system of an intelligent terminal, wherein the method comprises the following steps: according to the temperature change values of different AI industrial applications executed by the standard intelligent terminal equipment under different equipment parameters, constructing a relation model of the equipment parameters and the temperature change of the standard intelligent terminal equipment; acquiring equipment parameters of intelligent terminal equipment to be detected, and acquiring a temperature change predicted value of the intelligent terminal equipment to be detected based on a relation model of the equipment parameters and the temperature change of the standard intelligent terminal equipment; acquiring actual temperature change values of the intelligent terminal equipment to be tested for executing different AI industrial applications; and obtaining a test result of the intelligent terminal equipment to be tested according to the temperature change predicted value and the temperature change actual value.
Description
Technical Field
The invention relates to the technical field of AI industrial application capability of intelligent terminals, in particular to a method and a system for testing the AI industrial application capability of an intelligent terminal.
Background
The automated production flow required by industry 4.0 would require a large number of AI detection devices, including automatic identification of the primary type of image, audio, text. The special equipment has long research and development period, large investment and low return rate, and the adoption of the existing mature mobile intelligent terminals such as the intelligent mobile phone and the like for AI detection of the generation flow becomes an optional scheme with high cost performance. But the industrial environment is different from the daily use environment of the smart phone, and monotonous task and long-time execution are basic execution types. Industry 4.0 AI artificial intelligent detection based on intelligent terminal can be applied to various fields such as electronic component detection, printed food, aviation precision manufacturing, precision electronic parts, precision ceramic parts, product assembly link detection, product classification recognition, product positioning detection, printed matter detection, bottle cap detection, glass, tobacco and the like.
The AI industrial application capability test of the intelligent terminal in the industrial environment usually adopts a long-term copying mode, such as an electrified working test lasting thousands of hours, and the test modes directly simulate actual continuous use, but the intelligent terminal equipment is numerous at present and is layered, if the test is still carried out by adopting a long-time electrified working mode, the test efficiency is low, and the test result cannot be obtained in time.
Disclosure of Invention
In view of the above analysis, the embodiment of the invention aims to provide an AI industrial application capability test method and system for an intelligent terminal, which are used for solving the problems of long time consumption and low efficiency of the AI industrial application capability test of the existing intelligent terminal.
On the one hand, the embodiment of the invention provides an AI industrial application capability test method of an intelligent terminal, which comprises the following steps:
according to the temperature change values of different AI industrial applications executed by the standard intelligent terminal equipment under different equipment parameters, constructing a relation model of the equipment parameters and the temperature change of the standard intelligent terminal equipment;
acquiring equipment parameters of intelligent terminal equipment to be detected, and acquiring a temperature change predicted value of the intelligent terminal equipment to be detected based on a relation model of the equipment parameters and the temperature change of the standard intelligent terminal equipment;
acquiring actual temperature change values of the intelligent terminal equipment to be tested for executing different AI industrial applications; and obtaining a test result of the intelligent terminal equipment to be tested according to the temperature change predicted value and the temperature change actual value.
The beneficial effects of the technical scheme are as follows: by constructing a relation model of equipment parameters and temperature changes of standard intelligent terminal equipment, the temperature changes of equipment to be tested for executing different AI industrial applications are predicted, and by comparing the predicted values with actual values, the test result of the intelligent terminal to be tested can be rapidly judged, the test time is short, the test efficiency is high, and the time and labor cost can be greatly saved.
Based on the improvement of the technical characteristics, the equipment parameters comprise: CPU operating parameters, GPU operating parameters, memory throughput, fixed storage throughput, and convergence time for executing different AI industrial applications.
The beneficial effects of the technical scheme are as follows: the relation between the equipment parameters and the temperature change can be accurately measured through the change of different parameters by taking the temperature change as a dependent variable through taking the CPU operation parameters, the GPU operation parameters, the memory throughput rate, the fixed storage throughput rate and the convergence time for executing different AI industrial applications as independent variables, and the obtained relation model is more accurate, so that a foundation is provided for accurately predicting the temperature change of the equipment to be measured.
Further, the relationship model includes a linear regression model.
Further, the method comprises obtaining a relation model of the device parameters and the temperature change of the standard intelligent terminal device according to the temperature change values of the standard intelligent terminal device executing different AI industrial applications under different device parameters, including,
normalizing the equipment parameters;
based on the normalized equipment parameters, establishing a linear regression model of the equipment parameters and temperature change of the standard intelligent terminal equipment by adopting a multiple linear regression method;
determining model parameters of the linear regression model by adopting a least square method;
and carrying out residual analysis on the linear regression model, and if abnormal data exist, eliminating the abnormal data and redefining model parameters of the linear regression model.
The beneficial effects of the technical scheme are as follows: the linear regression model can be accurately and rapidly built through the least square method, whether abnormal data exist in the model or not can be found to influence the accuracy of the model through residual analysis of the model, and the built model can be more accurate through the extraction of the abnormal data.
Further, the determining the model parameters of the linear regression model by using a least square method includes:
the regression model is y=bx+u, where,
y represents an operation temperature change value, X is a device parameter, n is the number of the device parameter, m is the number of the statistical samples, B is a regression coefficient, and U is a regression constant;
the least square method is adopted to sum the squares of the dispersion between the actual value and the estimated value of the model
Minimum target, calculate +.>
Wherein X is i And Y i Representing the observed value of the ith sample, Y i * Representing the regression value of the i-th sample.
Further, the normalizing the device parameter data includes normalizing the device parameter data using the following formula:
wherein X is * Representing normalized equipment parameters, X being original equipment parameters, X mean X is the mean value of the device parameters std Is the standard deviation of the device parameters.
The beneficial effects of the technical scheme are as follows: and through normalization processing of the data, different scales, data and data differences are eliminated, and a data base is provided for a subsequent construction model.
On the other hand, the embodiment of the invention provides an AI industrial application capability test system of an intelligent terminal, which comprises the following modules:
the model construction module is used for constructing a relation model of the equipment parameters and the temperature change of the standard intelligent terminal equipment according to the temperature change values of the standard intelligent terminal equipment executing different AI industrial applications under different equipment parameters;
the temperature prediction module is used for acquiring equipment parameters of the intelligent terminal equipment to be detected, and acquiring an operation temperature change predicted value of the intelligent terminal equipment to be detected based on a relation model of the equipment parameters and the temperature change of the standard intelligent terminal equipment;
the test result acquisition module is used for acquiring the temperature change actual value of the intelligent terminal equipment to be tested for executing the AI industrial application; and obtaining a test result of the intelligent terminal equipment to be tested according to the temperature change predicted value and the temperature change actual value.
Further, the device parameters include: CPU operating parameters, GPU operating parameters, memory throughput, fixed storage throughput, and convergence time for executing different AI industrial applications.
Further, the relationship model includes a linear regression model.
Further, the model building module comprises,
the normalization module is used for carrying out normalization processing on the equipment parameters;
the linear regression model construction module is used for constructing a linear regression model of the equipment parameters and the temperature change of the standard intelligent terminal equipment by adopting a multiple linear regression method based on the normalized equipment parameters;
the model parameter acquisition module is used for determining model parameters of the linear regression model by adopting a least square method;
and the model correction module is used for carrying out residual analysis on the linear regression model, and if abnormal data exist, eliminating the abnormal data and redefining model parameters of the linear regression model.
In the invention, the technical schemes can be mutually combined to realize more preferable combination schemes. Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention may be realized and attained by the structure particularly pointed out in the written description and drawings.
Drawings
The drawings are only for purposes of illustrating particular embodiments and are not to be construed as limiting the invention, like reference numerals being used to refer to like parts throughout the several views.
FIG. 1 is a flowchart of an AI industrial application capability test method of an intelligent terminal according to an embodiment of the invention;
fig. 2 is a block diagram of a system for testing AI industrial application capability of an intelligent terminal according to an embodiment of the present invention.
Detailed Description
Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form a part hereof, and together with the description serve to explain the principles of the invention, and are not intended to limit the scope of the invention.
The automated production flow required by industry 4.0 would require a large number of AI detection devices, including automatic identification of the primary type of image, audio, text. The special equipment has long research and development period, large investment and low return rate, and the adoption of the existing mature mobile intelligent terminals such as the intelligent mobile phone and the like for AI detection of the generation flow becomes an optional scheme with high cost performance. But the industrial environment is different from the daily use environment of the smart phone, and monotonous task and long-time execution are basic execution types. Industry 4.0 AI artificial intelligent detection based on intelligent terminal can be applied to various fields such as electronic component detection, printed food, aviation precision manufacturing, precision electronic parts, precision ceramic parts, product assembly link detection, product classification recognition, product positioning detection, printed matter detection, bottle cap detection, glass, tobacco and the like.
The existing AI application test is mainly focused on a software test layer, and cannot evaluate the AI application capability of the intelligent terminal in an industrial environment, especially the durable operation capability under long-term working conditions. The AI industrial application capability test of the intelligent terminal in the current industrial environment usually adopts a long-term copying mode, such as an electrified working test lasting thousands of hours, and the test modes directly simulate actual continuous use, but the intelligent terminal equipment is numerous at present and is layered, if the test is still carried out by adopting a long-time electrified working mode, the test efficiency is low, and the test result cannot be obtained in time.
The energy consumption in the computation is actually related to the reversibility of the computation. The Landauer principle states that energy must be expended to erase information. (first form) if the computer erases 1 bit of information, the total amount of energy dissipated to the environment is at least K B Tln2, where K B Is the boltzmann constant, and T is the temperature at which the computer is located. In AI calculation, various logic gate operations are needed, bits are erased continuously, and especially the operation requirement of AI tasks on CPU and GPU is very high, the calculation speed is very high, and the heating value is large. First, the different types of AI tasks have different heating effects on the same class of computing terminals. AI model computation densities of text, pictures, speech, etc. are different. And secondly, the distribution of the heating values of the same AI tasks in different computing terminals is different, the GPU mainly processes matrix grid information, the CPU mainly processes comprehensive scheduling tasks, the memory and the fixed storage are used for storing temporary data and long-term data in a computing stage, and the heating value of equipment can be influenced by different processing rates and storage rates. The smaller the temperature rise of the intelligent terminal running the AI application, the longer the intelligent terminal can run stably for a long time.
Therefore, based on consideration of CPU, GPU, storage and different AI tasks and temperature change as a measurement index, whether the intelligent terminal meets the requirement of stable operation capability of an industrial environment can be rapidly judged. On the basis, the invention discloses a method for testing the AI industrial application capability of an intelligent terminal, which comprises the following steps as shown in figure 1:
s1, according to temperature change values of different AI industrial applications executed by standard intelligent terminal equipment under different equipment parameters, constructing a relation model of equipment parameters and temperature change of the standard intelligent terminal equipment.
The standard intelligent terminal equipment adopts methods such as expert scoring method, analytic hierarchy process and the like, and a series of equipment is selected according to hardware parameters, historical failure rate and the like of the equipment, and can meet the running time requirements of AI industrial application under different equipment parameter combinations. The hardware parameters include memory capacity, processor model, battery capacity, battery life, etc. And setting the standard intelligent terminal equipment in the simulated industrial environment, changing the equipment parameters of the standard intelligent terminal equipment, and obtaining the temperature rise time of the standard intelligent terminal equipment for executing different AI tasks under different equipment parameters.
For example, the working environment of a production line, the annual average temperature of 20 ℃, the air humidity of 10% and the dust density of 3mg/m < 3 >, an intelligent terminal used in the production line executes AI application for comprehensively using image recognition and character recognition to detect printing defective products and judge abnormal production line and working condition by voice recognition, and the whole startup and shutdown time of the intelligent terminal is consistent with that of a production line. And selecting a series of intelligent terminal equipment as standard intelligent terminal equipment based on the hardware parameters of the equipment, the historical failure rate and other indexes.
Specifically, the device parameters include: CPU operating parameters, GPU operating parameters, memory throughput, fixed storage throughput, and convergence time for executing different AI industrial applications. The faster the CPU and GPU run, the greater the heating value. The higher the throughput rate of the memory and the fixed memory throughput rate, the larger the heating value. The computation density of different AI models is different, and thus the impact on the temperature rise of the intelligent terminal is also different. By considering the difference of the processor speed, the memory speed and the AI task application, the heating factor of the intelligent terminal when the AI application is executed can be comprehensively considered, so that the relation between the equipment parameter and the temperature rise can be accurately obtained, and a foundation is provided for the subsequent test of the equipment to be tested.
For example, the CPU operating parameter may calculate a test time for pi, with a number of calculated bits of 1 megabit. The GPU operating parameter is a repeated drawing test time, for example, a repeated drawing of a full screen 60 frames of the smartphone. The memory throughput is the memory random access test time, e.g., 10M data is randomly accessed. The fixed memory throughput is a fixed memory random access test time, such as random access 10M data. Different AI industry applications include text-type AI-recognition tasks, sound-type AI-recognition tasks, and picture-type AI-recognition tasks. When the method is implemented, the equipment parameters of the standard intelligent terminal are changed in a test environment with the same specific industrial environment, and temperature change measurement is carried out on different AI industrial applications of the standard intelligent terminal to obtain sample data.
Considering that regression analysis is a commonly used analysis method for predicting the relationship between independent and dependent variables, it is preferable that the relationship model selects a regression model, which specifically includes a linear regression model in this embodiment.
Exemplary, constructing a linear regression model of device parameters and temperature changes of standard intelligent terminal devices includes the following steps:
s11, carrying out normalization processing on the equipment parameters;
specifically, the following formula is adopted to normalize the device parameter data:
wherein X is * Representing normalized equipment parameters, X being original equipment parameters, X mean X is the mean value of the device parameters std Is the standard deviation of the device parameters.
Through normalization processing, the differences of different dimensions and orders of magnitude are eliminated, and modeling analysis is convenient for data.
S12, based on the normalized equipment parameters, a multiple linear regression method is adopted to establish a linear regression model of the equipment parameters and the temperature change of the standard intelligent terminal equipment.
Specifically, the linear regression model is y=bx+u, wherein,
y represents an operation temperature change value, X is a device parameter, n is the number of the device parameter, m is the number of the statistical samples, B is a regression coefficient, and U is a regression constant;
s13, determining model parameters of the linear regression model by adopting a least square method.
Specifically, a least square method is adopted to obtain the sum of squares of the dispersion between the actual value and the estimated value of the model
Minimum target, calculate +.>
Wherein X is i And Y i Representing the observed value of the ith sample, Y i * Representing the regression value of the i-th sample.
S14, carrying out residual analysis on the linear regression model, and if abnormal data exist, eliminating the abnormal data and redefining model parameters of the linear regression model.
After the linear regression model is established, the model is verified through residual analysis, if abnormal point data exist, abnormal data are removed, model parameters of the linear regression model are determined again, and the model meets the equal variance requirement. For example, a residual map may be used to perform residual analysis on the model.
For example, the linear regression model can be subjected to fitness test and significance test, and abnormal data in the sample data can be removed, so that the regression model has high fitness and significance. The specific goodness-of-fit test and significance test are common statistical methods and will not be described in detail herein.
Other types of relational models, such as logistic regression models, may also be built.
S2, acquiring equipment parameters of the intelligent terminal equipment to be detected, and acquiring a temperature change predicted value of the intelligent terminal equipment to be detected based on a relation model of the equipment parameters and the temperature change of the standard intelligent terminal equipment.
Specifically, after obtaining the CPU operation parameters, the GPU operation parameters, the memory throughput rate, the fixed storage throughput rate and the convergence time of executing different AI industrial applications of the intelligent terminal to be tested, the parameters are brought into the relationship model obtained in step S1, so as to obtain the operation temperature change predicted value of the device to be tested for executing the AI industrial applications.
S3, acquiring actual temperature change values of the intelligent terminal equipment to be tested for executing different AI industrial applications; and obtaining a test result of the intelligent terminal equipment to be tested according to the temperature change predicted value and the temperature change actual value.
Specifically, the temperature of the intelligent terminal to be tested is tested before and after the intelligent terminal to be tested executes different AI industrial applications, so as to obtain the actual temperature change value of the intelligent terminal to be tested for executing different AI industrial applications. The intelligent terminal to be measured can be measured for multiple times, and the average value of the multiple measurements is taken as the actual value of the temperature change.
And obtaining a test result of the intelligent terminal equipment to be tested according to the difference value between the temperature change predicted value and the temperature change actual value. For example, if the difference value between the predicted value of the temperature change and the actual value of the temperature change is within a preset threshold value range, the industrial application capability of the intelligent terminal to be tested is indicated to meet the requirement, otherwise, the industrial application capability of the intelligent terminal to be tested is considered to not meet the requirement. The requirements for the temperature change value of the intelligent terminal for executing the AI industrial application in different industrial application environments are different, and the threshold range is different, so that specific setting is required according to the actual industrial environment requirements.
In addition, even if the CPU operating parameters, GPU operating parameters, memory throughput, fixed storage throughput, and convergence time for executing different AI industrial applications are the same, differences between circuit layouts and accessory collocations of different terminals are also necessarily reflected in differences in temperature. Therefore, the AI industrial application capability test method of the intelligent terminal provided by the invention can be used for testing the AI industrial application capability of the intelligent terminal, further reflecting the differences of the wiring mode, the heat dissipation layout, the component level and the like between the terminal to be tested and the standard terminal according to the temperature change predicted value and the temperature change actual value, and measuring the quality level of the wiring mode, the heat dissipation layout and the component level of the intelligent terminal to be tested. The AI industrial application capability test method of the intelligent terminal can be directly applied to a plurality of related fields such as equipment test before AI model deployment, AI terminal equipment purchase evaluation, AI terminal equipment design prototype evaluation and the like.
In order to conveniently check the test result, the evaluation result of the device to be tested can be intuitively displayed in a scoring form, for example, the score of a standard intelligent terminal device is 100 points, and the evaluation result can be obtained according to a formula
And the intelligent terminal to be tested is scored, so that the inspector can conveniently and intuitively obtain the evaluating result. Wherein θ represents a percentile coefficient, deltaT score And the difference value between the temperature change predicted value and the temperature change actual value of the intelligent terminal to be tested is represented.
Compared with the prior art, the AI industrial application capability test method of the intelligent terminal provided by the embodiment of the invention has the following beneficial effects:
1. by constructing a relation model of equipment parameters and temperature changes of standard intelligent terminal equipment, the temperature changes of equipment to be tested for executing different AI industrial applications are predicted, and by comparing the predicted values with actual values, the test result of the intelligent terminal to be tested can be rapidly judged, the test time is short, the test efficiency is high, and the time and labor cost can be greatly saved.
2. The relation between the equipment parameters and the temperature change can be accurately measured through the change of different parameters by taking the temperature change as a dependent variable through taking the CPU operation parameters, the GPU operation parameters, the memory throughput rate, the fixed storage throughput rate and the convergence time for executing different AI industrial applications as independent variables, and the obtained relation model is more accurate, so that a foundation is provided for accurately predicting the temperature change of the equipment to be measured.
3. The linear regression model can be accurately and rapidly built through the least square method, whether abnormal data exist in the model or not can be found to influence the accuracy of the model through residual analysis of the model, and the built model can be more accurate through eliminating the abnormal data.
4. And through normalization processing of the data, different scales, data and data differences are eliminated, and a data base is provided for a subsequent construction model.
The invention discloses an AI industrial application capability test system of an intelligent terminal, which comprises a model construction module, a temperature prediction module of intelligent terminal equipment to be tested and a test result acquisition module as shown in fig. 2.
Specifically, the model construction module is used for constructing a relation model of the device parameters and the temperature change of the standard intelligent terminal device according to the temperature change values of the standard intelligent terminal device executing different AI industrial applications under different device parameters.
Specifically, the device parameters include: CPU operating parameters, GPU operating parameters, memory throughput, fixed storage throughput, and convergence time for executing different AI industrial applications.
The temperature prediction module is used for acquiring equipment parameters of the intelligent terminal equipment to be detected, and acquiring an operation temperature change predicted value of the intelligent terminal equipment to be detected based on a relation model of the equipment parameters and the temperature change of the standard intelligent terminal equipment;
the test result acquisition module is used for acquiring the temperature change actual value of the intelligent terminal equipment to be tested for executing the AI industrial application; and obtaining a test result of the intelligent terminal equipment to be tested according to the temperature change predicted value and the temperature change actual value.
Preferably, the relational model comprises a linear regression model.
Preferably, the model building module comprises,
the normalization module is used for carrying out normalization processing on the equipment parameters;
the linear regression model construction module is used for constructing a linear regression model of the equipment parameters and the temperature change of the standard intelligent terminal equipment by adopting a multiple linear regression method based on the normalized equipment parameters;
the model parameter acquisition module is used for determining model parameters of the linear regression model by adopting a least square method;
and the model correction module is used for carrying out residual analysis on the linear regression model, and if abnormal data exist, eliminating the abnormal data and redefining model parameters of the linear regression model.
Further, the determining the model parameters of the linear regression model by using a least square method includes:
the regression model is y=bx+u, where,
y represents the running temperature change value, X is the equipmentParameters, n is the number of equipment parameters, m is the number of statistical samples, B is a regression coefficient, and U is a regression constant;
the least square method is adopted to sum the squares of the dispersion between the actual value and the estimated value of the model
Minimum target, calculate +.>
Wherein X is i And Y i Representing the observed value of the ith sample, Y i * Representing the regression value of the i-th sample.
Further, the normalizing the device parameter data includes normalizing the device parameter data using the following formula:
wherein X is * Representing normalized equipment parameters, X being original equipment parameters, X mean X is the mean value of the device parameters std Is the standard deviation of the device parameters.
The method embodiment and the system embodiment are based on the same principle, and the related parts can be mutually referred to and can achieve the same technical effect.
Those skilled in the art will appreciate that all or part of the flow of the methods of the embodiments described above may be accomplished by way of a computer program to instruct associated hardware, where the program may be stored on a computer readable storage medium. Wherein the computer readable storage medium is a magnetic disk, an optical disk, a read-only memory or a random access memory, etc.
The present invention is not limited to the above-mentioned embodiments, and any changes or substitutions that can be easily understood by those skilled in the art within the technical scope of the present invention are intended to be included in the scope of the present invention.
Claims (8)
1. The AI industrial application capability test method of the intelligent terminal is characterized by comprising the following steps:
according to the temperature change values of different AI industrial applications executed by the standard intelligent terminal equipment under different equipment parameters, constructing a relation model of the equipment parameters and the temperature change of the standard intelligent terminal equipment;
acquiring equipment parameters of intelligent terminal equipment to be detected, and acquiring a temperature change predicted value of the intelligent terminal equipment to be detected based on a relation model of the equipment parameters and the temperature change of the standard intelligent terminal equipment;
acquiring actual temperature change values of the intelligent terminal equipment to be tested for executing different AI industrial applications; obtaining a test result of the intelligent terminal equipment to be tested according to the temperature change predicted value and the temperature change actual value;
the device parameters include: CPU operation parameters, GPU operation parameters, memory throughput rate, fixed storage throughput rate and convergence time for executing different AI industrial applications;
obtaining a test result of the intelligent terminal equipment to be tested according to the temperature change predicted value and the temperature change actual value, wherein the test result comprises the following steps:
if the difference value between the temperature change predicted value and the temperature change actual value is within a preset threshold range, the AI industrial application capacity of the intelligent terminal equipment to be tested meets the requirements, otherwise, the AI industrial application capacity of the intelligent terminal equipment to be tested does not meet the requirements.
2. The AI industrial applicability testing method of the intelligent terminal of claim 1, wherein the relationship model comprises a linear regression model.
3. The method for testing the AI industrial applicability of an intelligent terminal according to claim 2, wherein the obtaining a relationship model of the device parameter and the temperature change of the standard intelligent terminal device according to the temperature change values of the standard intelligent terminal device executing different AI industrial applicability under different device parameters comprises,
normalizing the equipment parameters;
based on the normalized equipment parameters, establishing a linear regression model of the equipment parameters and temperature change of the standard intelligent terminal equipment by adopting a multiple linear regression method;
determining model parameters of the linear regression model by adopting a least square method;
and carrying out residual analysis on the linear regression model, and if abnormal data exist, eliminating the abnormal data and redefining model parameters of the linear regression model.
4. The AI industrial applicability test method of the intelligent terminal according to claim 3, wherein the determining the model parameters of the linear regression model by using the least square method comprises:
the regression model is y=bx+u, where,
y represents an operation temperature change value, X is a device parameter, n is the number of the device parameter, m is the number of the statistical samples, B is a regression coefficient, and U is a regression constant;
the least square method is adopted to sum the squares of the dispersion between the actual value and the estimated value of the model
Minimum target, calculate +.>
Wherein X is i And Y i Representing the observed value of the ith sample, Y i * Representing the regression value of the i-th sample.
5. The AI industrial applicability test method of the intelligent terminal according to claim 3, wherein the normalizing the device parameters comprises normalizing the device parameters using the following formula:
wherein X is * Representing normalized equipment parameters, X being original equipment parameters, X mean X is the mean value of the device parameters std Is the standard deviation of the device parameters.
6. The AI industrial application capability test system of the intelligent terminal is characterized by comprising the following modules:
the model construction module is used for constructing a relation model of the equipment parameters and the temperature change of the standard intelligent terminal equipment according to the temperature change values of the standard intelligent terminal equipment executing different AI industrial applications under different equipment parameters;
the temperature prediction module is used for acquiring equipment parameters of the intelligent terminal equipment to be detected, and acquiring an operation temperature change predicted value of the intelligent terminal equipment to be detected based on a relation model of the equipment parameters and the temperature change of the standard intelligent terminal equipment;
the test result acquisition module is used for acquiring the temperature change actual value of the intelligent terminal equipment to be tested for executing the AI industrial application; obtaining a test result of the intelligent terminal equipment to be tested according to the temperature change predicted value and the temperature change actual value;
the device parameters include: CPU operation parameters, GPU operation parameters, memory throughput rate, fixed storage throughput rate and convergence time for executing different AI industrial applications;
obtaining a test result of the intelligent terminal equipment to be tested according to the temperature change predicted value and the temperature change actual value, wherein the test result comprises the following steps:
if the difference value between the temperature change predicted value and the temperature change actual value is within a preset threshold range, the AI industrial application capacity of the intelligent terminal equipment to be tested meets the requirements, otherwise, the AI industrial application capacity of the intelligent terminal equipment to be tested does not meet the requirements.
7. The AI industrial applicability testing system of the intelligent terminal of claim 6, wherein the relationship model comprises a linear regression model.
8. The AI industrial applicability test system of the intelligent terminal of claim 7, wherein the model building module comprises:
the normalization module is used for carrying out normalization processing on the equipment parameters;
the linear regression model construction module is used for constructing a linear regression model of the equipment parameters and the temperature change of the standard intelligent terminal equipment by adopting a multiple linear regression method based on the normalized equipment parameters;
the model parameter acquisition module is used for determining model parameters of the linear regression model by adopting a least square method;
and the model correction module is used for carrying out residual analysis on the linear regression model, and if abnormal data exist, eliminating the abnormal data and redefining model parameters of the linear regression model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110944557.0A CN113656238B (en) | 2021-08-17 | 2021-08-17 | AI industrial application capability test method and system of intelligent terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110944557.0A CN113656238B (en) | 2021-08-17 | 2021-08-17 | AI industrial application capability test method and system of intelligent terminal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113656238A CN113656238A (en) | 2021-11-16 |
| CN113656238B true CN113656238B (en) | 2023-07-04 |
Family
ID=78491781
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110944557.0A Active CN113656238B (en) | 2021-08-17 | 2021-08-17 | AI industrial application capability test method and system of intelligent terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113656238B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101453383A (en) * | 2007-12-06 | 2009-06-10 | 中国移动通信集团公司 | Device, network status analysis method and apparatus |
| CN106682781A (en) * | 2016-12-30 | 2017-05-17 | 山东鲁能软件技术有限公司 | Power equipment multi-index prediction method |
| CN107817685A (en) * | 2017-12-05 | 2018-03-20 | 上海中信信息发展股份有限公司 | intelligent temperature monitoring method, device, system and terminal |
| CN109886430A (en) * | 2019-01-24 | 2019-06-14 | 同济大学 | A kind of equipment health state evaluation and prediction technique based on industrial big data |
| CN110515811A (en) * | 2019-08-09 | 2019-11-29 | 中国信息通信研究院 | Terminal artificial intelligence performance benchmark test method and device |
| CN110515790A (en) * | 2018-05-21 | 2019-11-29 | 重庆理工大学 | Terminal temperature prediction and optimization system and method |
| CN111191838A (en) * | 2019-12-27 | 2020-05-22 | 赛腾机电科技（常州）有限公司 | Industrial equipment state control method and device integrating artificial intelligence algorithm |
| CN111639842A (en) * | 2020-05-20 | 2020-09-08 | 湖北博华自动化系统工程有限公司 | Equipment health evaluation method, evaluation system and equipment health prediction method |
| CN112703457A (en) * | 2018-05-07 | 2021-04-23 | 强力物联网投资组合2016有限公司 | Method and system for data collection, learning and machine signal streaming for analysis and maintenance using industrial internet of things |
| CA3158765A1 (en) * | 2019-11-25 | 2021-06-03 | Strong Force Iot Portfolio 2016, Llc | Intelligent vibration digital twin systems and methods for industrial environments |
| CN112926636A (en) * | 2021-02-03 | 2021-06-08 | 中车青岛四方机车车辆股份有限公司 | Method and device for detecting abnormal temperature of traction converter cabinet body |
| CN113051700A (en) * | 2019-12-27 | 2021-06-29 | 北京国双科技有限公司 | Equipment reliability monitoring method and device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10984338B2 (en) * | 2015-05-28 | 2021-04-20 | Raytheon Technologies Corporation | Dynamically updated predictive modeling to predict operational outcomes of interest |
-
2021
- 2021-08-17 CN CN202110944557.0A patent/CN113656238B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101453383A (en) * | 2007-12-06 | 2009-06-10 | 中国移动通信集团公司 | Device, network status analysis method and apparatus |
| CN106682781A (en) * | 2016-12-30 | 2017-05-17 | 山东鲁能软件技术有限公司 | Power equipment multi-index prediction method |
| CN107817685A (en) * | 2017-12-05 | 2018-03-20 | 上海中信信息发展股份有限公司 | intelligent temperature monitoring method, device, system and terminal |
| CN112703457A (en) * | 2018-05-07 | 2021-04-23 | 强力物联网投资组合2016有限公司 | Method and system for data collection, learning and machine signal streaming for analysis and maintenance using industrial internet of things |
| CN110515790A (en) * | 2018-05-21 | 2019-11-29 | 重庆理工大学 | Terminal temperature prediction and optimization system and method |
| CN109886430A (en) * | 2019-01-24 | 2019-06-14 | 同济大学 | A kind of equipment health state evaluation and prediction technique based on industrial big data |
| CN110515811A (en) * | 2019-08-09 | 2019-11-29 | 中国信息通信研究院 | Terminal artificial intelligence performance benchmark test method and device |
| CA3158765A1 (en) * | 2019-11-25 | 2021-06-03 | Strong Force Iot Portfolio 2016, Llc | Intelligent vibration digital twin systems and methods for industrial environments |
| CN111191838A (en) * | 2019-12-27 | 2020-05-22 | 赛腾机电科技（常州）有限公司 | Industrial equipment state control method and device integrating artificial intelligence algorithm |
| CN113051700A (en) * | 2019-12-27 | 2021-06-29 | 北京国双科技有限公司 | Equipment reliability monitoring method and device |
| CN111639842A (en) * | 2020-05-20 | 2020-09-08 | 湖北博华自动化系统工程有限公司 | Equipment health evaluation method, evaluation system and equipment health prediction method |
| CN112926636A (en) * | 2021-02-03 | 2021-06-08 | 中车青岛四方机车车辆股份有限公司 | Method and device for detecting abnormal temperature of traction converter cabinet body |
Non-Patent Citations (1)
| Title |
|---|
| 工业互联网平台评价指标体系构建与应用研究;李君;邱君降;柳杨;窦克勤;成雨;刘帅;文莎;;中国科技论坛(第12期);70-86 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113656238A (en) | 2021-11-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109543906B (en) | Atmospheric visibility prediction method and equipment | |
| CN109685289B (en) | Method, device and system for forward prediction of blast furnace conditions | |
| CN113688042A (en) | Method and device for determining test scene, electronic equipment and readable storage medium | |
| CN114722434B (en) | Block chain-based ledger data control method and device | |
| CN115936262B (en) | Yield prediction method, system and medium based on big data environment interference | |
| CN116453438B (en) | Display screen parameter detection method, device, equipment and storage medium | |
| CN115097796A (en) | Quality control system and method for simulating big data and correcting AQL value | |
| CN114705965A (en) | Third generation semiconductor reliability data analysis system based on big data | |
| CN114676749A (en) | Power distribution network operation data abnormity judgment method based on data mining | |
| CN114091347A (en) | Electronic component service life prediction method | |
| CN113656238B (en) | AI industrial application capability test method and system of intelligent terminal | |
| CN113589172A (en) | Service life estimation method for power grid components | |
| CN110007174B (en) | Power management and control detection method for data center | |
| CN113627755B (en) | Testing method, device, equipment and storage medium of intelligent terminal factory | |
| CN115965625A (en) | Instrument detection device based on visual identification and detection method thereof | |
| CN105227410A (en) | Based on the method and system that the server load of adaptive neural network detects | |
| CN115494431A (en) | Transformer fault warning method, terminal equipment and computer readable storage medium | |
| CN114155412A (en) | Deep learning model iteration method, device, equipment and storage medium | |
| CN113835016A (en) | DDR chip limit performance test method, test device, equipment and storage medium | |
| CN111881259A (en) | Equipment fault probability evaluation method and system based on text mining | |
| CN113032998A (en) | Medical instrument life evaluation method and device | |
| CN117571742B (en) | Method and device for realizing chip quality inspection based on artificial intelligence | |
| CN113191017B (en) | Virtual measurement model construction method, virtual measurement method, system, electronic device and computer-readable storage medium | |
| CN113408221B (en) | Probe service life prediction method, system, device and storage medium | |
| CN117436770B (en) | BIM-based building engineering construction management method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |