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

Model training method based on small amount of marked data, control system and air conditioner

A technology for labeling data and model training, applied in mechanical equipment and other directions, can solve problems such as difficulty in data collection

Inactive Publication Date: 2021-03-05
GREE ELECTRIC APPLIANCES INC +1
View PDF9 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, at present, the data collection work in traditional industries is difficult to carry out. Even if the manpower and time costs are sufficient, only a small amount of marked data can be collected, especially the comfortable and energy-saving marked data of air-conditioning operation.

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
  • Model training method based on small amount of marked data, control system and air conditioner
  • Model training method based on small amount of marked data, control system and air conditioner

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0019] Embodiment 1: A kind of model training method based on a small amount of labeled data, refer to figure 1 , figure 2 .

[0020] A model training method based on a small amount of labeled data, the specific steps include:

[0021] S1. First, 250 pieces of comfortable and energy-saving air conditioner operating data in the laboratory environment are manually obtained according to expert experience, and a radial basis neural network framework is constructed. The pipe temperature is the input layer feature. Through the input of the above environmental parameters, the effect of fully describing the current environmental state and user needs can be achieved. The network is defined with the output layer features of the air conditioner external fan speed, compressor speed, and electronic expansion valve opening. The above three The first parameter is the core parameter of air conditioning system operation. By adjusting these three parameters, the overall use status and functi...

Embodiment 2

[0024] Embodiment 2: A kind of model control system based on a small amount of labeled data, refer to figure 1 , figure 2 .

[0025] A model control system based on a small amount of labeled data is characterized in that it includes an initial model building module, a data augmentation module, and a model optimization module. The specific steps of each module are described in detail below:

[0026] The initial model construction module: First, 250 pieces of comfortable and energy-saving air-conditioning operation data in the laboratory environment are manually obtained according to expert experience; second, a radial basis neural network framework is constructed, and the indoor temperature, set temperature, outdoor temperature, exhaust temperature, The inner pipe temperature and outer pipe temperature are the characteristics of the input layer (the above parameters are sufficient to describe the current environmental status and user needs), and the network is defined by the ...

Embodiment 3

[0030] Embodiment 3: an air conditioner.

[0031] An air conditioner, including a processor and a memory, the memory is used to store a computer program, characterized in that: when the computer program is invoked by the processor, the model training method based on a small amount of labeled data described in Embodiment 1 is implemented .

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 model training method based on a small amount of marked data, a control system and an air conditioner. The training method specifically comprises the steps of constructing aninitial neural network model, taking environment state parameters of the current air conditioner as model input, inputting data into the model for training and updating after the data are expanded through a data augmentation method, accurately predicting air conditioner operation parameters, and then controlling the air conditioner to enter a comfortable energy-saving model to operate. The controlsystem comprises an initial model construction module, a data augmentation module and a model optimization module. According to the method, model training with high performance is completed by utilizing a small amount of marked data and a large amount of unmarked data, the defect that a model cannot be trained or the performance effect of the trained model is poor due to the fact that data collection is difficult in a traditional task is overcome, and deep combination of the traditional industry and the artificial intelligence technology is further promoted.

Description

technical field [0001] The invention relates to the technical field of air conditioners, in particular to a model training method based on a small amount of labeled data, a control system and an air conditioner. Background technique [0002] In order to better combine artificial intelligence technology with traditional industries, data collection is the key to its application. However, at present, the data collection work in traditional industries is difficult to carry out. Even if the manpower and time costs are sufficient, only a small amount of marked data can be collected, especially the comfortable and energy-saving marked data of air-conditioning operation. Therefore, in order to avoid the difficulty of collecting labeled data, this paper proposes an air conditioning comfort and energy-saving model training method based on a small amount of labeled data. Contents of the invention [0003] Aiming at the deficiencies of the prior art, the present invention proposes a ...

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): F24F11/63
CPCF24F11/63
Inventor 岳冬陈翀宋德超罗晓宇
Owner GREE ELECTRIC APPLIANCES INC
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