Data-driven multi-agent system PID control protocol self-learning method

A technology of multi-agent system and self-learning method, which is applied to electric controllers, controllers with specific characteristics, etc., to achieve the effect of ensuring performance optimization and ensuring optimality

Pending Publication Date: 2022-02-18
LIAONING UNIVERSITY OF PETROLEUM AND CHEMICAL TECHNOLOGY
View PDF10 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Aiming at the problem of model-free PID multi-agent system consistency, the present invention provides a data-driven multi-agent system PID control protocol self-learning method, which can make All agents reach a consistent state, that is, to achieve consistency, and at the same time ensure the optimality of the specified performance of multi-agents

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-driven multi-agent system PID control protocol self-learning method
  • Data-driven multi-agent system PID control protocol self-learning method
  • Data-driven multi-agent system PID control protocol self-learning method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0198] Multi-agent system matrix:

[0199]

[0200] figure 1 As shown, given the communication topology diagram, the multi-agent edge weight is:

[0201] e 12 =0.8,e 14 =0.7,e 21 =0.8,e 31 =0.8,e 32 =0.5,e 43 = 0.8.

[0202] Figure 2-9 The evolution of the critic neural network weights for four agents is shown. Once the critic neural network weights are found, the PID and PI control parameters can be calculated. It can be found that under the learned PID control protocol, all agents have reached an agreement, such as Figure 10 and Figure 12 is shown in , where the neighbor error states and the state at which all agents reach the optimal consensus are plotted.

[0203] To demonstrate that the control performance achieved under the PID consistent control protocol may be better than state feedback control or proportional-integral (PI) control, and to demonstrate the generality of the developed model-free PID consistent control method, we contrast PI with PID.

...

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-driven multi-agent system PID control protocol self-learning method. According to themethod,an optimal consistency problem for a PID-controlled multi-agent system is provided; the solving of control protocol parameters based on PID control is converted into the solving of a non-zero sum game problem, a non-strategy Q learning algorithm is provided, and online self-learning of the PID control protocol parameters and optimal consistency of the multi-agent system are achieved; a reinforcement learning (RL) technology and a neural network function estimation method are combined, the agent system is not required to be dynamically known, and measurable data are completely utilized to self-learn PID control protocol parameters. According to the method, all the agents can reach the consistent state with the dynamic model of the multi-agent system unknown, namely, the consistency is achieved, and meanwhile, the optimality of the specified performance of the multiple agents can be guaranteed.

Description

technical field [0001] The invention relates to a controller, in particular to a data-driven multi-agent system PID control protocol self-learning method. Background technique [0002] Multi-agent consistency has always been one of the key issues in complex dynamical systems, which means that all agents distributed in the network can asymptotically reach the same state. Although there have been research results on the consensus or optimal consensus of multi-agent systems, these methods mainly use the "current" information of the system to design the consensus or synchronization protocol. This presents a promising new research topic, that is, not only "present" information, but also "past" information and even "future" trends. [0003] The PID controller can make good use of the past value, present value and future trend of the tracking error in order to obtain a good transient state and stable performance. The results of existing PID-based multi-agent system consistency pr...

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): G05B11/42
CPCG05B11/42
Inventor 李金娜王佳琦
Owner LIAONING UNIVERSITY OF PETROLEUM AND CHEMICAL TECHNOLOGY
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