Data center energy efficiency optimization method based on transfer learning

A data center and transfer learning technology, which is applied in the fields of electrical digital data processing, digital data processing components, energy-saving computing, etc., can solve problems such as high data volume and data quality requirements, difficult modeling, and poor generalization performance , to achieve the effect of increasing accuracy and robustness, improving convergence speed, and improving generalization ability

Active Publication Date: 2020-01-17
创新奇智(上海)科技有限公司
View PDF7 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this modeling method has the following defects: modeling is difficult, it requires high data volume and data quality, it is easy to overfit, and its generalization performance is poor.

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
  • Data center energy efficiency optimization method based on transfer learning
  • Data center energy efficiency optimization method based on transfer learning
  • Data center energy efficiency optimization method based on transfer learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0052] refer to Figure 1-2 , the present invention provides a technical solution: the model training method in a data center energy efficiency optimization method based on transfer learning specifically includes the following steps:

[0053] S1: Input historical raw data;

[0054] S2: Perform data preprocessing and feature engineering on the original data;

[0055] S3: Extract training samples of each unit;

[0056] S4: Construct Base model training samples: merge the training samples of each unit and randomly shuffle them;

[0057] S5: Train the Base model to obtain a prediction model with high precision and large variance;

[0058] S6: Construct the List-wise model training samples, select the training samples of the specified unit, randomly shuffle and combine them into list samples;

[0059] S7: Migrate the pre-trained weights of the Base model, fine-tune the List-wise model, transfer the parameters of the pre-trained hidden layer in the Base model to the shared hidde...

Embodiment 2

[0062] refer to image 3 , the present invention provides a technical solution: a data center energy efficiency optimization method based on transfer learning also includes performing the model reasoning and decision-making method, which specifically includes the following steps:

[0063] S9: input online raw data;

[0064] S10: Perform data preprocessing and feature engineering on the original data;

[0065] S11: Extracting forecast samples of each unit;

[0066] S12: Execute optimal control parameter search and solution based on the energy consumption prediction module Predictor Model;

[0067] S13: Obtain several sets of optimal control parameters;

[0068] S14: Use the ranking module Rank Model to sort the control parameters;

[0069] S15: output optimal control parameters.

Embodiment 3

[0071] refer to Figure 4 , the present invention provides a technical solution: a data center energy efficiency optimization method based on transfer learning also includes the main steps of performing the data preprocessing and feature engineering including:

[0072] Remove outliers, using a sliding window to remove outliers greater than three times the variance of the mean;

[0073] Small window sliding average to solve the problem of data fluctuation;

[0074] construct features based on the physical properties of the device;

[0075] Perform feature combination and feature intersection;

[0076] Feature screening based on feature importance.

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 discloses a data center energy efficiency optimization method based on transfer learning in the field of data mining and machine learning. The method comprises a model training method and a model reasoning decision-making method. A basic model is pre-trained based on Base Model by using data samples of all units; according to the method, hidden layer parameters learned by a basic model are migrated to List-wiseModel of each unit, and fine adjustment is carried out by using a relatively low learning rate, so that the problem of missing of part of unit samples is solved, and the generalization ability of the model is improved; by designing a multi-task learning model and adding rank constraint to multi-target loss, the problems of overfitting and model high variance caused by noise are solved, and a sample pair is selected in a random and periodic sliding window mode to improve the convergence rate of the model; the optimal unit control parameters are obtained by using theoptimal linear search of the energy consumption prediction task, the control parameters are sorted by using the rank prediction task sorting model, and the optimal control parameters are comprehensively selected, so that the accuracy and robustness of the optimal control parameters are improved.

Description

technical field [0001] The invention relates to the technical fields of data mining and machine learning, in particular to a data center energy efficiency optimization method based on migration learning. Background technique [0002] The emergence of energy and environmental issues has put energy conservation and emission reduction on the important agenda. With the development of technologies such as cloud services, big data, and AI computing, enterprises and governments have invested in building a large number of data centers. At present, the energy consumption of data centers in China Generally high, the average PUE value is between 2.2-3.0. China's data center electricity consumption accounts for 3% of the electricity consumption of the whole society, and it is expected to reach 3.3% in 2020. [0003] At present, there are many researches on data center energy saving, and energy consumption simulation software is often used to simulate and compare the energy efficiency p...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06F1/3287G06F9/48
CPCG06F1/3287G06F9/4806Y02D10/00
Inventor 张发恩马凡贺
Owner 创新奇智(上海)科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap