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

A Learning Control Method for Complex Nonlinear Systems Based on Non-Strictly Repeated Problems

A nonlinear system, learning control technology, applied in the field of learning control of complex nonlinear systems based on non-strictly repetitive problems

Active Publication Date: 2021-07-13
JIANGSU INST OF ECONOMIC & TRADE TECH
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Purpose of the invention: For the problems existing in the prior art, the purpose of the invention is to provide a learning control method for complex nonlinear systems based on non-strictly repeated problems, so as to solve the problem of how to solve the problem of unknown variables, control gains and disturbances in the system. Make full use of the known boundary conditions and non-strictly repetitive laws of unknown variables in the actual system, combine the classic adaptive iterative learning control method with known boundary conditions, design a learning control method, and simultaneously solve a variety of non-strictly repetitive problems in the system , to achieve robust asymptotic tracking of the target trajectory

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
  • A Learning Control Method for Complex Nonlinear Systems Based on Non-Strictly Repeated Problems
  • A Learning Control Method for Complex Nonlinear Systems Based on Non-Strictly Repeated Problems
  • A Learning Control Method for Complex Nonlinear Systems Based on Non-Strictly Repeated Problems

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0090] Embodiment one: the system equation of the controlled nonlinear system is as follows:

[0091]

[0092] in, is an unknown parameter that is not strictly repeated, and its change law satisfies the high-order internal model; is a system function vector that satisfies linear growth. unknown parameters Bounded variation in the interval [-1.4,+1.4]; The bounded change interval of is [-3,+3]; The bounded change interval of is [-0.1,+0.1]; the bounded change interval of external disturbance is [-0.1,+0.1]. The system runs iteratively in the discrete time interval {0,1,…,100}.

[0093] The iterative change law of satisfies different high-order internal models respectively, as shown in the following formula:

[0094]

[0095] It can be seen from the above formula that the unknown parameter and respectively satisfy the second-order internal model, and Then satisfy the first-order internal model, that is, the unknown parameter Varies only with time, inde...

Embodiment 2

[0116] Embodiment 2: In order to better investigate the scope of application of the proposed complex nonlinear system learning control method based on non-strictly repetitive problems, considering that strict repetition in the iterative domain is a special case of non-strict repetition, the designed learning control The method is applied to the following control problem of a permanent magnet linear motor:

[0117]

[0118] Among them, v k (t) represents the angular velocity of the mover of the permanent magnet linear motor, and the nonlinear functions of the system are and The iteration domain of unknown parameters of the system is strictly repeated, as follows: θ (1) =0.8237, θ (2) = θ (3) = θ (4) =-0.014, control gain b=0.0014, disturbance d(t)=-0.07sin(0.001πt). The system tracking reference trajectory is where coefficient Randomly takes values ​​in the interval (0,1] as the iteration changes.

[0119] At this time, the boundary of the unknown parameters o...

Embodiment 3

[0121] Embodiment 3: For a single-input single-output system with multiple non-strictly repeated problems, the system equation is considered as follows:

[0122]

[0123] Among them, the unknown parameter θ k (t) changes with the time domain-iterative domain, and the range of change is known, and the change law is the same as that in Embodiment 1 Unknown control gain B(t)=(1+sin(0.5t)), disturbance d(t)=0.1cos(0.05t). System tracking reference trajectory is the same as that in Embodiment 1 The controlled system runs iteratively within a finite time interval t∈{0,1,…,100}.

[0124] According to the formula (17), take the initial value of the learning gain matrix A complex nonlinear system learning control method based on non-strictly repeated problems is run iteratively 100 times in the discrete time interval {0,1,...,100}, and the learning convergence of the maximum absolute value error of state tracking as table 1 and Figure 4 shown.

[0125] Table 1. The maximum...

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 complex nonlinear system learning control method based on non-strictly repeated problems. For a class of complex nonlinear systems that can be repeatedly operated, considering the existence of unknown variables, control gains and disturbances in the system, fully utilize Known boundary conditions and non-strictly repetitive laws of unknown variables in the actual system. Combining the classic adaptive iterative learning control method with known boundary conditions, a learning control method is designed to solve various non-strictly repetitive problems in the system. The invention can make a kind of complex nonlinear system asymptotically converge to a non-strictly repeated target trajectory within a limited time, and has certain robustness.

Description

technical field [0001] The invention relates to a method in the field of learning control, in particular to a learning control method for complex nonlinear systems based on non-strictly repeated problems. Background technique [0002] The non-strictly repetitive problem refers to the problem that some states of the system are not strictly consistent in each iteration in a repeatable system. Since the main purpose of iterative learning control theory is to make full use of the repeatable operation characteristics of the controlled system, by comparing the measured error with the ideal value, continuous self-learning and error correction are carried out in the iteration, so the problem of non-strict repetition is seriously shackles. The development of iterative learning control theory. At present, the discussions on non-strictly repetitive problems in the system mainly include non-strictly repetitive initial states, non-strictly repetitive reference trajectories, and non-stri...

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 Patents(China)
IPC IPC(8): G05B13/04
Inventor 周伟刘保彬于淼
Owner JIANGSU INST OF ECONOMIC & TRADE TECH
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