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

Terminal precision air conditioner optimization control method and system based on reinforcement learning

A technology of intensive learning and precision air conditioning, applied in design optimization/simulation, electrical equipment construction parts, instruments, etc., can solve problems such as inability to adjust air conditioners, slow response to cold and hot spots, energy waste, etc., to achieve control automation and automation Controlling and avoiding the effect of manual intervention

Inactive Publication Date: 2021-05-14
南京群顶科技股份有限公司
View PDF7 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] (1) There is hysteresis in PID control, and the response to cold and hot spots is slow, so it is impossible to adjust the air conditioner in time according to the temperature change in the computer room;
[0005] (2) PID control is only based on a single temperature measurement point, without considering the overall situation of the computer room, and it is difficult to achieve the expected control effect;
[0006] (3) PID parameters need to be adjusted frequently by humans, and in order to reserve a certain fluctuation space, there is usually an excessive use of air volume and chilled water, resulting in a waste of energy

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
  • Terminal precision air conditioner optimization control method and system based on reinforcement learning
  • Terminal precision air conditioner optimization control method and system based on reinforcement learning
  • Terminal precision air conditioner optimization control method and system based on reinforcement learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0041] see figure 1 ,figure 1 A schematic diagram of the steps of a method for optimal control of terminal precision air conditioners based on reinforcement learning provided by the embodiment of the present invention is as follows:

[0042] Step S100, obtaining sample data of data center computer room equipment within a preset time, and extracting a set of sub-sample sequences according to the obtained sample data;

[0043] Specifically, the preset time can be set according to actual needs. The equipment in the data center computer room can be heating and cooling equipment in the data center computer room. The sample data can be equipment control parameters and temperature data. The sub-sample sequence can be any sample data extracted from the sample data. Data within a fixed time window forward from moment to moment.

[0044] In some implementations, the data center equipment room heating and cooling equipment control parameters and temperature data are obtained from the se...

Embodiment 2

[0063] see figure 2 , figure 2 A block diagram of a terminal precision air-conditioning optimization control system based on reinforcement learning provided by the embodiment of the present invention is as follows:

[0064] The data collection and sub-sample sequence extraction module 100 is used to obtain sample data of data center equipment room equipment within a preset time, and extract a sub-sample sequence set according to the obtained sample data;

[0065] The heat balance equation generation module 200 is used to construct a relationship model between heat load and refrigeration equipment through sample data, and generate a heat balance equation;

[0066] The heat balance equation solving module 300 is used to solve the heat balance equation by using the EM algorithm according to the sub-sample sequence set to obtain the coefficient of action of the system heat balance;

[0067] The optimization objective function definition and air conditioner control parameter so...

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 provides a terminal precision air conditioner optimization control method and system based on reinforcement learning, and relates to the field of machine room air conditioner control. The terminal precision air conditioner optimization control method based on reinforcement learning comprises the steps that data center machine room equipment sample data within preset time is obtained, and a sub-sample sequence set is extracted according to the obtained sample data; a relation model between the heat load and the refrigeration equipment is constructed through the sample data, and a heat balance equation is generated; according to the subsample sequence set, solving the heat balance equation by using an EM algorithm to obtain an action coefficient of system heat balance; and defining an optimization objective function, and solving control parameters of the air conditioner by using a reinforcement learning method. In addition, the invention further provides a terminal precision air conditioner optimization control system based on reinforcement learning. The terminal precision air conditioner optimization control system comprises a data collection and subsample sequence extraction module, a heat balance equation generation module, a heat balance equation solving module and an optimization objective function definition and air conditioner control parameter solving module.

Description

technical field [0001] The invention relates to the field of computer room air-conditioning control, in particular to a method and system for optimal control of terminal precision air-conditioning based on reinforcement learning. Background technique [0002] The data center computer room needs precision air conditioners to control the temperature within an appropriate range to ensure the normal operation of servers and storage racks. Precision air conditioner refers to the precision air conditioner dedicated to the computer room that can fully meet the environmental conditions of the computer room. The redundancy can ensure that the air conditioner can operate normally throughout the year. [0003] In the prior art, the output is mainly adjusted through the PID control of the air conditioner itself to better control the ambient temperature. Generally, the PID parameters are combined with the temperature sense of the return air duct to control the parameters of the fan and...

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): G06F30/20H05K7/20G06F111/08G06F113/02G06F119/08
CPCG06F30/20G06F2111/08G06F2113/02G06F2119/08H05K7/20745H05K7/20836
Inventor 杨鹏杨波
Owner 南京群顶科技股份有限公司
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