Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

System power consumption optimized lightweight system and method with autonomous learning function

A technology of system power consumption and self-learning, which is applied in the field of deep learning, can solve the problems of not being able to realize the real-time performance of tasks well, and achieve the effect of realizing intelligence, improving efficiency and reducing power consumption

Pending Publication Date: 2021-03-16
廊坊嘉杨鸣科技有限公司
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the time overhead of this kind of method is very considerable, and it cannot realize the real-time performance of the task well.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • System power consumption optimized lightweight system and method with autonomous learning function
  • System power consumption optimized lightweight system and method with autonomous learning function
  • System power consumption optimized lightweight system and method with autonomous learning function

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0030] Such as figure 1 As shown, a lightweight system with self-learning function and optimized system power consumption, the system includes: a data acquisition unit configured to collect real-time operating data of each operating unit in the system; the real-time operating data includes at least: each The temperature data, current data, voltage data and running time of the running unit; the sample state estimation unit is configured to estimate the probability of the collected real-time running data based on the collected real-time running data, and establish a state sample set; The judgment unit is configured to use the established state judgment function based on the established state sample set to judge the operating state of each operating unit in the current system, and obtain whether each operating unit is in a concave curve operating state or a convex curve operating state; The control unit is configured to use the operating state of each operating unit obtained base...

Embodiment 2

[0034] On the basis of the previous embodiment, the sample state estimation unit, based on the collected real-time operation data, performs probability estimation on the collected real-time operation data, and the method for establishing a state sample set performs the following steps: the collected real-time operation data Run the data, when the time is 0, that is, when k is 0, according to the probability distribution, the initial state sample set is established: s, w, Q and P represent the degree data, current data, voltage data and running time respectively; get the weight: Among them, N is the total number of real-time running data; then calculate the instantaneous value at a certain moment: let k=k+1, calculate the instantaneous value at k+1 moment; use the following formula to summarize the peak value of the instantaneous value: Among them, K k is the predicted value of the wave at the kth moment; Indicates the real instantaneous value; z k Indicates the es...

Embodiment 3

[0037] On the basis of the previous embodiment, the state judging unit, based on the established state sample set, uses the established state judgment function to judge the operating state of each operating unit in the current system, and obtains whether each operating unit is out of a concave curve The method of running state or convex curve running state performs the following steps: Use the following formula as the state judgment function: Power min =Min∑ i∈N [s c (i)·w c +Q l (i)·p l +P i (i)·p i ]+∑ i,j∈N,i≠j t(i,j)·p t ; Among them, Power min is the calculated power consumption, N is the total number of real-time running data, c is the first adjusted square number, the value range is: 1~3, l is the second adjusted square number, the value range is 2~4 , p is the adjustment probability, the value range is: 25% to 75%, t is the curve time variable; the data set corresponding to each unit in the state sample set is input into the state judgment function as an input...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention belongs to the technical field of deep learning, and particularly relates to a system power consumption optimized lightweight system and method with an autonomous learning function, andthe system comprises: a data collection unit which is configured to be used for collecting the real-time operation data of each operation unit in the system, wherein the real-time operation data at least comprises temperature data, current data, voltage data and operation time of each operation unit during operation; and the sample state estimation unit that is configured to perform probability estimation on the acquired real-time operation data based on the acquired real-time operation data and establish a state sample set. By analyzing the operation state of each operation unit in the systemin real time, different power consumption control means are used for different operation states, autonomous learning can be performed according to the control effect, intelligence of power consumption control is realized, the power consumption of the system is further reduced, and meanwhile, the power consumption control efficiency of the system is improved.

Description

technical field [0001] The invention belongs to the technical field of deep learning, and in particular relates to a lightweight system and method for optimizing system power consumption with an autonomous learning function. Background technique [0002] Power consumption is one of the most basic electrical characteristic indicators of processor performance. One of the very important reasons is that with the increase of frequency, the increase of power consumption is accompanied by the change of thermal characteristics. There will be serious constraints and impacts. [0003] With the development of science and technology, electronic devices such as mobile phones and tablet computers are becoming more and more intelligent. For example, the mobile phone has multiple functions such as wifi, Internet access, camera, audio and video playback, gravity sensing, games, and Bluetooth. [0004] These complex functions are based on efficient and intelligent processing systems, which ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06F1/329G06N20/00
CPCG06F1/329G06N20/00
Inventor 张兴旺
Owner 廊坊嘉杨鸣科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com