Model-free self-adaptive water mixing temperature control system and method based on deep reinforcement learning

A model-free self-adaptive and reinforcement learning technology, applied in the direction of self-adaptive control, general control system, control/regulation system, etc., can solve problems such as wasting water resources and difficult temperature regulation, and achieve reliable and accurate mixed water system, avoid The effect of frequent temperature changes

Active Publication Date: 2020-10-16
HARBIN UNIV OF COMMERCE
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The invention solves the problems of difficulty in temperature adjustment and waste of water resources in the current manual temperature adjustment of the existing mixing water device. The invention discloses a "model-free adaptive mixing water temperature control system and method based on deep reinforcement learning"

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-free self-adaptive water mixing temperature control system and method based on deep reinforcement learning
  • Model-free self-adaptive water mixing temperature control system and method based on deep reinforcement learning
  • Model-free self-adaptive water mixing temperature control system and method based on deep reinforcement learning

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0046] Specific implementation mode one: combine Figure 1-Figure 3Describe this embodiment. The model-free adaptive mixed water temperature control system based on deep reinforcement learning in this embodiment includes an action network module and a value network module;

[0047] The action network module includes an estimation network module and an evaluation target network module;

[0048] The action network module is used to define the state space and action space of the mixed water system;

[0049] The value network module is used to judge and evaluate the network environment;

[0050] The action network module and the value network module are used to interact with the environment to obtain the DDPG model.

specific Embodiment approach 2

[0051] Specific implementation mode two: combination Figure 1-Figure 3 Describe this embodiment, the model-free adaptive mixed water temperature control method based on deep reinforcement learning in this embodiment, the specific method steps are as follows:

[0052] Step 1, customize the state space and action space of the mixed water system, and establish action network and value network;

[0053] Step 2, train the action network and value network according to the data generated by interacting with the mixed water environment, and obtain the DDPG model of mixed water temperature regulation;

[0054] Step 3: Deploy the DDPG model on the muddy water equipment, communicate with the cloud server in real time, update the equipment model parameters asynchronously, and realize self-adaptive learning of the new muddy water environment.

specific Embodiment approach 3

[0055] Specific implementation mode three: combination Figure 1-Figure 3 Describe this embodiment, the model-free adaptive mixed water temperature control method based on deep reinforcement learning in this embodiment, in step 1, the action network includes: action network, target action network; the value network includes judgment value network , the state space and the action space of the water mixing system of the target value network, the action space of the water mixing system is adjusting the rotation speed A∈[V of the paddle max , V min ], where V max is the maximum speed of temperature regulation, V min =-V max ;

[0056] The state space S is specifically: Which respectively represent: cold water end temperature before mixing, cold water end pressure before mixing, cold water end water flow before mixing, hot water end temperature before mixing, hot water end pressure before mixing, hot water end water flow before mixing , the current temperature after water mi...

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 model-free adaptive mixed water temperature control system and method based on deep reinforcement learning, and belongs to the field of cold and hot water mixed water temperature control. The problems that manual temperature adjustment of an existing water mixing device is difficult, and water resources are wasted are solved. The system comprises an action network moduleand a value network module, and the method specifically comprises the following steps: step 1, customizing a state space and an action space of a water mixing system, and establishing an action network and a value network; 2, training an action network and a value network according to data generated by interaction with the water mixing environment to obtain a water mixing temperature adjustment DDPG model; and 3, deploying the DDPG model in the water mixing equipment, communicating with a cloud server in real time, and asynchronously updating equipment model parameters to realize adaptive learning of a new water mixing environment. The temperature control system and method can adapt to the use environment, have high adaptability to environmental factors and enable the water mixing system to be reliable and accurate.

Description

technical field [0001] The invention relates to a model-free adaptive mixed water temperature control system and method based on deep reinforcement learning, belonging to the field of cold and hot water mixed water temperature control. Background technique [0002] Most of the traditional water mixing devices use manual temperature adjustment, which has problems such as difficult temperature adjustment and waste of water resources. However, some mixed water intelligent constant temperature systems on the market usually use fixed algorithms, and the problem is that they cannot adapt Different environments have problems such as poor reliability and poor accuracy. [0003] Most of the existing control research work is on univariate PID control, the theory and design of which are well established, understood and practically applied. But the overall multivariable PID system has not been successful, and most industrial processes are multivariable in nature. [0004] The traditio...

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): G05B13/04
CPCG05B13/042
Inventor 黄文俊兰琦琦解泽宇
Owner HARBIN UNIV OF COMMERCE
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