CN112732060A - Power configuration method and power configuration device based on artificial intelligence - Google Patents
Power configuration method and power configuration device based on artificial intelligence Download PDFInfo
- Publication number
- CN112732060A CN112732060A CN202011155981.9A CN202011155981A CN112732060A CN 112732060 A CN112732060 A CN 112732060A CN 202011155981 A CN202011155981 A CN 202011155981A CN 112732060 A CN112732060 A CN 112732060A
- Authority
- CN
- China
- Prior art keywords
- power
- power configuration
- verification result
- limit range
- configuration
- 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.)
- Pending
Links
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 44
- 238000000034 method Methods 0.000 title claims abstract description 30
- 238000012795 verification Methods 0.000 claims abstract description 98
- 238000012360 testing method Methods 0.000 claims abstract description 45
- 230000004044 response Effects 0.000 claims abstract description 28
- 238000013461 design Methods 0.000 claims description 13
- 230000017525 heat dissipation Effects 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 5
- 238000010586 diagram Methods 0.000 description 9
- 238000012545 processing Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000009966 trimming Methods 0.000 description 2
- 230000015556 catabolic process Effects 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013021 overheating Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
- G06F1/26—Power supply means, e.g. regulation thereof
- G06F1/32—Means for saving power
- G06F1/3203—Power management, i.e. event-based initiation of a power-saving mode
- G06F1/3206—Monitoring of events, devices or parameters that trigger a change in power modality
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
- G06F1/26—Power supply means, e.g. regulation thereof
- G06F1/32—Means for saving power
- G06F1/3203—Power management, i.e. event-based initiation of a power-saving mode
- G06F1/3234—Power saving characterised by the action undertaken
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3058—Monitoring arrangements for monitoring environmental properties or parameters of the computing system or of the computing system component, e.g. monitoring of power, currents, temperature, humidity, position, vibrations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3058—Monitoring arrangements for monitoring environmental properties or parameters of the computing system or of the computing system component, e.g. monitoring of power, currents, temperature, humidity, position, vibrations
- G06F11/3062—Monitoring arrangements for monitoring environmental properties or parameters of the computing system or of the computing system component, e.g. monitoring of power, currents, temperature, humidity, position, vibrations where the monitored property is the power consumption
-
- 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
Abstract
The invention provides a power configuration method and a power configuration device based on artificial intelligence, which are suitable for optimizing the power configuration of a device to be tested. The method comprises the following steps: obtaining the specification of a device to be tested; generating a power configuration according to the specification and the artificial intelligence model; configuring power of the device under test according to the power configuration; measuring the temperature of the device under test to generate a heat conduction verification result corresponding to the power configuration; and updating the power configuration in response to the thermal conductivity verification result not matching the specification.
Description
Technical Field
The present invention relates to power configuration technologies, and in particular, to a power configuration method and a power configuration apparatus based on artificial intelligence.
Background
With the progress of science and technology, the demand of people for the performance of electronic products is gradually increasing. When purchasing electronic products, the performance of the electronic products is often an important factor affecting sales. Increasing the power of an electronic product can significantly increase the performance of the electronic product, but can also overheat the electronic product and cause performance degradation. Therefore, it is one of the objectives of the skilled person to adjust the power configuration and temperature of the electronic product to optimize the performance of the electronic product.
Disclosure of Invention
The invention provides a power configuration method and a power configuration device based on artificial intelligence, which can automatically optimize the power configuration of a device to be tested based on artificial intelligence.
The invention relates to a power configuration device based on artificial intelligence, which is suitable for optimizing the power configuration of a device to be tested and comprises a processor, a storage medium, a power configuration circuit and a heat sensor. The power distribution circuit is coupled to the device under test. The storage medium stores specifications of the device under test and a plurality of modules. The processor is coupled to the storage medium, the power configuration circuit, and the thermal sensor, and accesses and executes a plurality of modules, wherein the plurality of modules includes a trim module and a thermal conductivity verification module. The fine tuning module generates a power configuration according to the specification and the artificial intelligence model. The thermal conduction verification module configures power of the device to be tested according to the power configuration through the power configuration circuit, measures the temperature of the device to be tested through the thermal sensor to generate a thermal conduction verification result corresponding to the power configuration, and updates the power configuration in response to the thermal conduction verification result not matching with the specification.
In an embodiment of the invention, the heat conduction verification module determines the heat dissipation design power of the dut according to the power configuration in response to the heat conduction verification result matching the specification.
In an embodiment of the invention, the modules further include a performance verification module. The performance verification module measures the performance of the device under test in response to the thermal conduction verification result matching the specification to generate a performance verification result corresponding to the power configuration, and updates the power configuration in response to the performance verification result not matching the specification.
In an embodiment of the invention, the performance verification module determines the heat dissipation design power of the dut according to the power configuration in response to the performance verification result matching the specification.
In an embodiment of the invention, the power configuration apparatus further includes an output device. The output device is coupled to the processor, wherein the processor outputs the alarm information through the output device in response to a mismatch between one of the thermal conduction verification result and the performance verification result and the specification.
In an embodiment of the invention, the specification includes a first power limit range and a second power limit range, and the plurality of modules further includes a preprocessing module. The preprocessing module generates a plurality of initial data according to the specification, wherein the plurality of initial data comprise first initial data corresponding to a first maximum value of a first power limit range and a second maximum value of a second power limit range, second initial data corresponding to a first minimum value of the first power limit range and a second minimum value of the second power limit range, and third initial data corresponding to random values of the first power limit range and the second power limit range.
In an embodiment of the invention, the fine tuning module includes a scoring module. The scoring module generates scores according to the plurality of initial data, wherein the artificial intelligence model generates a power configuration according to the scores.
In an embodiment of the invention, the scoring module updates the score according to the power configuration, and the artificial intelligence model updates the power configuration according to the score.
The invention provides a power configuration method based on artificial intelligence, which is suitable for optimizing the power configuration of a device to be tested and comprises the following steps: obtaining the specification of a device to be tested; generating a power configuration according to the specification and the artificial intelligence model; configuring power of the device under test according to the power configuration; measuring the temperature of the device under test to generate a heat conduction verification result corresponding to the power configuration; and updating the power configuration in response to the thermal conductivity verification result not matching the specification.
In an embodiment of the invention, the power configuration method further includes: and determining the heat dissipation design power of the device to be tested according to the power configuration in response to the heat conduction verification result being matched with the specification.
In an embodiment of the invention, the power configuration method further includes: measuring the performance of the device under test in response to the thermal conductivity verification result matching the specification to generate a performance verification result corresponding to the power configuration; and updating the power configuration in response to the performance verification result not matching the specification.
In an embodiment of the invention, the power configuration method further includes: and determining the heat dissipation design power of the device to be tested according to the power configuration in response to the performance verification result being matched with the specification.
In an embodiment of the invention, the power configuration method further includes: and outputting alarm information in response to a mismatch between one of the thermal conduction verification result and the performance verification result and the specification.
In an embodiment of the invention, the specification includes a first power limit range and a second power limit range, wherein the step of generating the power configuration according to the specification and the artificial intelligence model includes: generating a plurality of initial data according to the specification, wherein the plurality of initial data includes first initial data corresponding to a first maximum value of a first power limit range and a second maximum value of a second power limit range, second initial data corresponding to a first minimum value of the first power limit range and a second minimum value of the second power limit range, and third initial data corresponding to random values of the first power limit range and the second power limit range.
In an embodiment of the invention, the step of generating the power configuration according to the specification and the artificial intelligence model further includes: generating scores according to the plurality of initial data, wherein the artificial intelligence model generates the power configuration according to the scores.
In an embodiment of the invention, the step of generating the power configuration according to the specification and the artificial intelligence model further includes: the score is updated according to the power configuration, and the power configuration is updated by the artificial intelligence model according to the score.
Based on the above, the power configuration apparatus of the present invention can automatically output the optimal power configuration corresponding to the device under test without spending the user time to adjust the optimal power configuration.
Drawings
FIG. 1 shows a schematic diagram of an artificial intelligence based power configuration apparatus, according to an embodiment of the invention.
FIG. 2 illustrates a flow diagram of an artificial intelligence based power configuration method, according to an embodiment of the invention.
Fig. 3A is a diagram illustrating a comparison of scores corresponding to a default power configuration and an optimal power configuration, according to an embodiment of the invention.
FIG. 3B is a diagram illustrating a comparison of CPU temperature (CPU) for a default power configuration and an optimal power configuration, according to one embodiment of the present invention.
FIG. 3C is a diagram illustrating a comparison of apparent temperatures (skins) corresponding to a default power configuration and an optimal power configuration, according to an embodiment of the invention.
Description of the reference numerals
100: a power configuration device;
110: a processor;
120: a storage medium;
121: a pretreatment module;
122: a fine tuning module;
1221: a scoring module;
1222: an artificial intelligence model;
123: a thermal conduction verification module;
124: an efficiency verification module;
130: an output device;
140: a power configuration circuit;
150: a thermal sensor;
31. 32, 33, 34, 35, 36: a curve;
s201, S202, S203, S204, S205, S206, S207, S208, S209, S210, S211, S212, S213, S214, S215, S216: and (5) carrying out the following steps.
Detailed Description
Reference will now be made in detail to exemplary embodiments of the invention, examples of which are illustrated in the accompanying drawings. Wherever possible, the same reference numbers will be used throughout the drawings and the description to refer to the same or like parts.
FIG. 1 shows a schematic diagram of an artificial intelligence based power configuration apparatus 100, according to an embodiment of the invention. The power configuration apparatus 100 is adapted to optimize the power configuration of the device under test. Power configuration device 100 may include a processor 110, a storage medium 120, an output device 130, a power configuration circuit 140, and a thermal sensor 150.
The processor 110 is, for example, a Central Processing Unit (CPU), or other programmable general purpose or special purpose Micro Control Unit (MCU), a microprocessor (microprocessor), a Digital Signal Processor (DSP), a programmable controller, an Application Specific Integrated Circuit (ASIC), a Graphics Processing Unit (GPU), a video signal processor (ISP), an Image Processing Unit (IPU), an Arithmetic Logic Unit (ALU), a Complex Programmable Logic Device (CPLD), a field programmable logic device (FPGA), or other similar components. Processor 110 may be coupled to storage medium 120, output device 130, power configuration circuit 140, and thermal sensor 150, and may access and execute the various modules and various applications stored in storage medium 120.
The storage medium 120 is, for example, any type of fixed or removable Random Access Memory (RAM), read-only memory (ROM), flash memory (flash memory), hard disk (HDD), Solid State Drive (SSD), or the like or a combination thereof, and is used to store a plurality of modules or various applications executable by the processor 110. In the present embodiment, the storage medium 120 may store a plurality of modules including a preprocessing module 121, a fine tuning module 122, a thermal conduction verification module 123, and a performance verification module 124, the functions of which will be described later. In addition, the storage medium 120 may also store specifications of the device under test, wherein the specifications may be associated with information such as power limit, maximum appearance temperature, or maximum cpu temperature, and the invention is not limited thereto.
The output device 130 is a device such as a display, a speaker, or a signal transmitter that can be used to transmit information. Processor 110 may output information via output device 130.
The power distribution circuit 140 is coupled to the dut. The processor 110 may be coupled to the dut through the power distribution circuit 140 to provide power to the dut. In addition, the processor 110 may also obtain parameter values corresponding to power and related to performance from the device under test through the power configuration circuit 140 in response to providing the power to the device under test.
The thermal sensor 150 may be used to measure a temperature generated by the dut when operating according to the power configuration, wherein the temperature includes, for example, a cpu temperature or an external temperature, and the invention is not limited thereto.
Fig. 2 is a flowchart illustrating an artificial intelligence based power allocation method according to an embodiment of the present invention, wherein the power allocation method is suitable for optimizing the power configuration of a device under test, and the power allocation method can be implemented by the power allocation apparatus 100 shown in fig. 1.
In step S201, the trimming module 122 may obtain the specification of the dut from the storage medium 120. The specification of the device under test may be input to the power configuration apparatus 100 by a user through an external input device, for example. For example, a user may enter the specification of the device under test into the power configuration device 100 through a keyboard. Power configuration device 100 may store the specifications of the device under test to storage medium 120.
In step S202, the trimming module 122 may generate a power configuration of the dut according to the specification. In particular, the specification may include a plurality of power limit ranges that are respectively applicable to different scenarios. For example, the specification may include a first range of power limits and a second range of power limits different from the first range of power limits. The power configuration may include a power value corresponding to a first power limit range and a power value corresponding to a second power limit range, which is not limited in the present invention.
The preprocessing module 121 can generate a plurality of initial data according to the specification, wherein each of the plurality of initial data can be matched with the specification. The plurality of pieces of initial data may include first initial data. The first initial data may include a power value corresponding to a first power limit range, and the power value is a maximum value in the first power limit range. The first initial data may also include a power value corresponding to a second power limit range, and the power value is a maximum value in the second power limit range.
The plurality of initial data may also include second initial data. The second initial data may include a power value corresponding to the first power limit range, and the power value is a minimum value in the first power limit range. The second initial data may also include a power value corresponding to a second power limit range, and the power value is the smallest value in the second power limit range.
The plurality of initial data may further include one or more third initial data. For example, the plurality of pieces of initial data may include eight pieces of third initial data. The third initial data may include a power value corresponding to the first power limit range, and the power value is a random value in the first power limit range. The third initial data may also include a power value corresponding to a second power limit range, and the power value is a random value in the second power limit range.
The fine-tuning module 122 may include a scoring module 1221 and an artificial intelligence model 1222. The scoring module 1221 may generate a score based on the plurality of initial data. Specifically, the scoring module 1221 may instruct the power configuration circuit 140 to configure the power of the device under test according to the plurality of initial data, and obtain a plurality of parameter values corresponding to the power and related to the performance from the device under test. The scoring module 1221 may then generate scores corresponding to the plurality of initial data according to the plurality of parameter values. The scoring module 1221 may calculate the score according to, for example, tensrflow software or 3d mark11 software, but the invention is not limited thereto.
The artificial intelligence model 1222 may receive a plurality of initial data and scores corresponding to the plurality of initial data from the scoring module 1221, wherein the scores are related to the performance of the device under test, for example (e.g., a higher score indicates a higher performance of the device under test). The artificial intelligence model 1222 may then generate an initial power configuration based on the scores. The algorithm employed by the artificial intelligence model 1222 may be adjusted according to the needs of the user, and the invention is not limited thereto.
In step S203, the scoring module 1221 may receive the generated power configuration from the artificial intelligence model 1222 and update the score according to the power configuration. The scoring module 1221 may calculate the score of the current power configuration to update the score according to, for example, the tensrflow software or the 3d park 11 software, but the invention is not limited thereto.
In step S204, the processor 110 may determine whether the updated score exceeds a score threshold. If the updated score exceeds the score threshold, the process proceeds to step S206. If the updated score does not exceed the score threshold, the process proceeds to step S205.
In step S205, the artificial intelligence model 1222 may update the power configuration according to the score.
In step S206, the thermal conduction verification module 123 obtains the latest power configuration from the artificial intelligence model 1222. Then, the thermal conduction verification module 123 may configure the power of the device under test according to the power configuration through the power configuration circuit 140, and measure the temperature of the device under test through the thermal sensor 150 to generate a thermal conduction verification result corresponding to the power configuration. The thermal conduction verification module 123 may generate the thermal conduction verification result according to, for example, 3d mark11 software, although the invention is not limited in this respect. The heat conduction verification result may be related to the cpu temperature, the appearance temperature, the different part temperature or the environment temperature of the dut, but the invention is not limited thereto.
In step S207, the processor 110 may determine whether the heat conduction verification result matches the specification. If the heat conduction verification result matches the specification, the process proceeds to step S211. If the heat conduction verification result does not match the specification, the process proceeds to step S208.
In one embodiment, the thermal conduction verification module 123 may determine a thermal design power (thermal design power) of the dut according to a power configuration corresponding to the thermal conduction verification result in response to the thermal conduction verification result matching the specification. The thermal conduction verification module 123 may output the thermal dissipation design power to a user reference via the output device 130.
In step S208, the processor 110 may determine whether the number of occurrences of the thermal conductivity verification result not matching the specification exceeds a first threshold. If the number of occurrences of the thermal conduction verification result not matching the specification exceeds the first threshold, the process proceeds to step S209. If the number of occurrences of the thermal conduction verification result not matching the specification does not exceed the first threshold, step S210 is performed.
In step S209, the processor 110 may output an alarm message through the output device 130 to alert a user that the thermal conduction verification result of the dut has not been verified for a plurality of times.
In step S210, the thermal conduction verification module 123 may determine that the number of occurrences of the thermal conduction verification result not matching the specification is increased once. In addition, the thermal conduction verification module 123 may update the power configuration. For example, the thermal conduction verification module 123 may reduce the power value corresponding to the first power limit range in the power configuration, and may reduce the power value corresponding to the second power limit range in the power configuration.
In step S211, the performance verification module 124 may configure the power of the device under test according to the power configuration by the power configuration circuit 140, and measure the performance of the device under test by the power configuration circuit 140 to generate a performance verification result corresponding to the power configuration. The performance verification module 124 may generate the performance verification result according to, for example, Cinebench software, but the invention is not limited thereto.
In step S212, the processor 110 may determine whether the performance verification result matches the specification. If the performance verification result matches the specification, the process proceeds to step S216. If the performance verification result does not match the specification, the process proceeds to step S213.
In one embodiment, the performance verification module 124 may determine the thermal design power of the dut according to the power configuration corresponding to the performance verification result in response to the performance verification result matching the specification. The performance verification module 124 may output the heat dissipation design power to the user reference through the output device 130.
In step S213, the processor 110 may determine whether the number of occurrences of the performance verification result not matching the specification exceeds a second threshold. If the number of occurrences of the performance verification result not matching the specification exceeds the second threshold, step S214 is entered. If the number of occurrences of the performance verification result not matching the specification does not exceed the second threshold, step S215 is performed.
In step S214, the processor 110 may output an alarm message through the output device 130 to alert a user that the performance verification result of the dut has not been verified for a plurality of times.
In step S215, the performance verification module 124 may determine that the number of occurrences of the performance verification result not matching the specification is increased once. In addition, the performance verification module 124 may update the power configuration. For example, the performance verification module 124 may increase the power value corresponding to the first power limit range in the power configuration and may increase the power value corresponding to the second power limit range in the power configuration.
In step S216, the processor 110 may output the power configuration through the output device 130 for the user to refer to. The power configuration output by the output device 130 is the optimal power configuration of the dut.
Fig. 3A is a diagram illustrating a comparison of scores corresponding to a default power configuration and an optimal power configuration, according to an embodiment of the invention. Fig. 3A may include a curve 31 corresponding to a default power configuration, which is a power configuration defined by a user to meet the specification of a device under test, and a curve 32 corresponding to an optimal power configuration, which is a power configuration generated by the power distribution apparatus 100 according to the specification of the device under test. As can be seen from fig. 3A, the score of the dut operating according to the optimal power configuration is higher than the score of the dut operating according to the default power configuration when performing the test, wherein the score can be calculated according to the tensrflow software or the 3d scan 11 software, for example, but the invention is not limited thereto.
FIG. 3B is a diagram illustrating a comparison of CPU temperatures corresponding to a default power configuration and an optimal power configuration, according to one embodiment of the present invention. Fig. 3B may include a curve 33 corresponding to a default power configuration and a curve 34 corresponding to an optimal power configuration, wherein the default power configuration is a power configuration that is defined by a user and meets the specification of a device under test, and the optimal power configuration is a power configuration generated by the power distribution apparatus 100 according to the specification of the device under test. As can be seen from fig. 3B, during the test, the cpu temperature of the dut operating according to the optimal power configuration is lower than the cpu temperature of the dut operating according to the default power configuration.
Fig. 3C is a schematic diagram illustrating a comparison of apparent temperatures corresponding to a default power configuration and an optimal power configuration, according to an embodiment of the invention. Fig. 3C may include a curve 35 corresponding to a default power configuration, which is a power configuration defined by a user to meet the specification of a device under test, and a curve 36 corresponding to an optimal power configuration, which is a power configuration generated by the power distribution apparatus 100 according to the specification of the device under test. As can be seen from fig. 3C, during the test, the external temperature of the dut operating according to the optimal power configuration is lower than the external temperature of the dut operating according to the default power configuration.
In summary, the artificial intelligence model of the power configuration apparatus of the present invention can automatically generate the initial power configuration with reference to the specification determined by the engineer for the dut. The scoring module can score the power configuration generated by the artificial intelligence model to determine whether to update the power configuration. The artificial intelligence model may continually update the power configuration until the updated power configuration reaches a particular score. Then, the heat conduction verification module and the performance verification module can conduct heat conduction verification and performance verification on the power configuration to judge whether the power configuration meets the specification, and if not, the heat conduction verification module and the performance verification module can be selected to adjust the power configuration. Finally, the power configuration device can output the optimal power configuration of the device to be tested. The optimal power configuration enables the device under test to achieve optimal operational performance without overheating.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (16)
1. A power configuration device based on artificial intelligence is suitable for optimizing the power configuration of a device to be tested, and is characterized by comprising:
a thermal sensor;
a power configuration circuit coupled to the device under test;
a storage medium storing specifications of the device under test and a plurality of modules; and
a processor coupled to the storage medium, the power configuration circuit, and the thermal sensor, and accessing and executing the plurality of modules, wherein the plurality of modules comprises:
the fine tuning module generates the power configuration according to the specification and the artificial intelligence model; and
a thermal conduction verification module to configure, by the power configuration circuit, power of the device under test according to the power configuration, measure, by the thermal sensor, a temperature of the device under test to generate a thermal conduction verification result corresponding to the power configuration, and update the power configuration in response to the thermal conduction verification result not matching the specification.
2. The power distribution apparatus of claim 1, wherein the thermal conduction verification module determines a thermal design power of the device under test according to the power configuration in response to the thermal conduction verification result matching the specification.
3. The power configuration apparatus of claim 1, wherein the plurality of modules further comprises:
a performance verification module to measure a performance of the device under test in response to the thermal conduction verification result matching the specification to generate a performance verification result corresponding to the power configuration, and to update the power configuration in response to the performance verification result not matching the specification.
4. The power distribution apparatus of claim 3, wherein the performance verification module determines a thermal design power of the device under test according to the power configuration in response to the performance verification result matching the specification.
5. The power configuration apparatus of claim 3, further comprising:
an output device coupled to the processor, wherein the processor outputs an alarm message via the output device in response to one of the thermal conduction verification result and the performance verification result not matching the specification.
6. The power configuration apparatus of claim 1, wherein the specification comprises a first power limit range and a second power limit range, and the plurality of modules further comprises:
the preprocessing module generates a plurality of initial data according to the specification, wherein the plurality of initial data comprise first initial data corresponding to a first maximum value of the first power limit range and a second maximum value of the second power limit range, second initial data corresponding to a first minimum value of the first power limit range and a second minimum value of the second power limit range, and third initial data corresponding to random values of the first power limit range and the second power limit range.
7. The power configuration apparatus of claim 6, wherein the fine tuning module comprises:
and the scoring module generates scores according to the plurality of initial data, wherein the artificial intelligence model generates the power configuration according to the scores.
8. The power configuration device of claim 7, wherein the scoring module updates the score according to the power configuration, and the artificial intelligence model updates the power configuration according to the score.
9. A power configuration method based on artificial intelligence is suitable for optimizing the power configuration of a device to be tested, and is characterized by comprising the following steps:
obtaining the specification of the device to be tested;
generating the power configuration according to the specification and an artificial intelligence model;
configuring power of the device under test according to the power configuration;
measuring the temperature of the device under test to generate a thermal conduction verification result corresponding to the power configuration; and
updating the power configuration in response to the thermal conductivity verification result not matching the specification.
10. The power configuration method of claim 9, further comprising:
and determining the heat dissipation design power of the device to be tested according to the power configuration in response to the heat conduction verification result being matched with the specification.
11. The power configuration method of claim 9, further comprising:
measuring the performance of the device under test in response to the thermal conductivity verification result matching the specification to generate a performance verification result corresponding to the power configuration; and
updating the power configuration in response to the performance verification result not matching the specification.
12. The power configuration method of claim 11, further comprising:
and determining the heat dissipation design power of the device to be tested according to the power configuration in response to the performance verification result being matched with the specification.
13. The power configuration method of claim 11, further comprising:
outputting an alarm message in response to one of the heat conduction verification result and the performance verification result not matching the specification.
14. The power allocation method of claim 9, wherein the specification comprises a first power limit range and a second power limit range, and wherein generating the power configuration according to the specification and the artificial intelligence model comprises:
generating a plurality of initial data according to the specification, wherein the plurality of initial data includes first initial data corresponding to a first maximum value of the first power limit range and a second maximum value of the second power limit range, second initial data corresponding to a first minimum value of the first power limit range and a second minimum value of the second power limit range, and third initial data corresponding to random values of the first power limit range and the second power limit range.
15. The power allocation method of claim 14, wherein the step of generating the power configuration according to the specification and the artificial intelligence model further comprises:
generating scores according to the plurality of initial data, wherein the artificial intelligence model generates the power configuration according to the scores.
16. The power allocation method of claim 15, wherein the step of generating the power configuration according to the specification and the artificial intelligence model further comprises:
updating the score according to the power configuration, and updating the power configuration by the artificial intelligence model according to the score.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US201962927097P | 2019-10-28 | 2019-10-28 | |
| US62/927,097 | 2019-10-28 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112732060A true CN112732060A (en) | 2021-04-30 |
Family
ID=75597283
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011155981.9A Pending CN112732060A (en) | 2019-10-28 | 2020-10-26 | Power configuration method and power configuration device based on artificial intelligence |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN112732060A (en) |
| TW (1) | TWI755090B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106292773A (en) * | 2016-08-19 | 2017-01-04 | 青岛海信移动通信技术股份有限公司 | Temperature adjustment device and method for regulating temperature |
| CN109642829A (en) * | 2016-09-08 | 2019-04-16 | 高通股份有限公司 | Time-temperature sensor position offset error correction |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011164927A (en) * | 2010-02-09 | 2011-08-25 | Nec Access Technica Ltd | Home gateway device and power consumption suppressing method for the same |
| US9489027B1 (en) * | 2015-08-31 | 2016-11-08 | Wave Resource Strategies, Inc. | System and method for the accurate recordation of power consumption in a computing device utilizing power profiles |
| US9830413B2 (en) * | 2015-09-15 | 2017-11-28 | Taiwan Semiconductor Manufacturing Co., Ltd. | System and method for estimating performance, power, area and cost (PPAC) |
| KR102282301B1 (en) * | 2017-09-22 | 2021-07-27 | 광동 오포 모바일 텔레커뮤니케이션즈 코포레이션 리미티드 | POWER SUPPLY CIRCUIT, POWER SUPPLY DEVICE AND CONTROL METHOD |
-
2020
- 2020-10-08 TW TW109134916A patent/TWI755090B/en active
- 2020-10-26 CN CN202011155981.9A patent/CN112732060A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106292773A (en) * | 2016-08-19 | 2017-01-04 | 青岛海信移动通信技术股份有限公司 | Temperature adjustment device and method for regulating temperature |
| CN109642829A (en) * | 2016-09-08 | 2019-04-16 | 高通股份有限公司 | Time-temperature sensor position offset error correction |
Also Published As
| Publication number | Publication date |
|---|---|
| TW202117495A (en) | 2021-05-01 |
| TWI755090B (en) | 2022-02-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2020045236A1 (en) | Augmentation device, augmentation method, and augmentation program | |
| KR20170125720A (en) | Recognition apparatus based on deep neural network, training apparatus and methods thereof | |
| WO2016050129A1 (en) | Application processing method and device and terminal | |
| CN109163794B (en) | Method for detecting bandwidth of linear vibration motor | |
| US11157380B2 (en) | Device temperature impact management using machine learning techniques | |
| KR20170015097A (en) | Method and apparatus for power management | |
| US11681261B2 (en) | Control system, control method, learning device, control device, learning method for controlling an operation of a subject device on basis of a detemined command value | |
| TW201835716A (en) | Semiconductor device, operating condition controlling method, and non-transitory computer readable medium | |
| WO2020198023A1 (en) | System and method of machine learning-based design and manufacture of ear-dwelling devices | |
| CN113965313B (en) | Model training method, device, equipment and storage medium based on homomorphic encryption | |
| CN112732060A (en) | Power configuration method and power configuration device based on artificial intelligence | |
| CN111310905B (en) | Neural network model training method and device and heating and ventilation system energy efficiency optimization method | |
| WO2021186685A1 (en) | Simulation execution system, simulation execution method, and simulation execution program | |
| US10203733B2 (en) | Systems and methods for recommending optimum datacenter parameters to meet rack level airflow constraints | |
| CN112801302A (en) | Machine learning model publishing method and system based on interface | |
| CN110974038B (en) | Food material cooking degree determining method and device, cooking control equipment and readable storage medium | |
| US20200334338A1 (en) | Wafer characteristic prediction method and electronic device | |
| CN115374659B (en) | Electric turbulence control method and device and electronic equipment | |
| WO2017097118A1 (en) | Text classification processing method and apparatus | |
| JP2019016193A (en) | Analysis method, analysis device, and analysis program | |
| CN116401903A (en) | Human electromagnetic radiation quantitative analysis method, device, server and storage medium | |
| CN116579252A (en) | Particle swarm optimization-based vortex mechanical exhaust hole optimization method and system | |
| CN113516066B (en) | Power quality disturbance signal classification method and device, storage medium and electronic equipment | |
| CN113362827B (en) | Speech recognition method, device, computer equipment and storage medium | |
| JP4539356B2 (en) | Field effect transistor model generation device, field effect transistor model generation method, etc. |
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 |