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

Linear neuron on-line learning adaptive control method and controller for passive system

A technology of adaptive control and passive system, applied in the direction of adaptive control, general control system, control/regulation system, etc., can solve the problem of many manual interface parameters, unfavorable debugging and maintenance by ordinary operators, and difficult porting of configuration platforms, etc. question

Inactive Publication Date: 2011-04-27
STATE NUCLEAR ELECTRIC POWER PLANNING DESIGN & RES INST CO LTD
View PDF0 Cites 38 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] First of all, the design of the controller of the advanced control algorithm is mostly based on the mathematical model of the complex controlled object. The implementation program and method of the controller are very complicated, but the mathematical model of the controlled object in different industrial systems is different. As a result, different controllers should be designed for different controlled objects, so that the controllers with advanced control algorithms cannot be widely promoted; at the same time, the control effect of the controller depends to a large extent on the pros and cons of the mathematical model, so that the performance of each controller uneven;
[0005] Secondly, because the structure and form of the controller itself of the advanced control algorithm are too complicated, it is not easy to realize it under the configuration platform. The implementation of the advanced control algorithm in the prior art can use advanced programming languages ​​(such as: C ++, Delphi, etc.) to write algorithm programs, use DLL (Dynamic Link Library, dynamic link library) or ActiveX control technology to realize the application of advanced control algorithms, or based on the script language in the configuration platform to realize the application of advanced control algorithms, the former exists The problem is that the implementation of the algorithm program relies on professional technicians, and it is difficult to transplant to different configuration platforms, and it is difficult to achieve true seamless integration. The problem with the latter is that different configuration platforms have different scripting languages, which is also not conducive to Transplant between different configuration platforms;
[0006] Finally, controllers with advanced control algorithms need to debug and adjust many manual interface parameters, and require on-site support from senior experts, which is not conducive to the debugging and maintenance of ordinary operators

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
  • Linear neuron on-line learning adaptive control method and controller for passive system
  • Linear neuron on-line learning adaptive control method and controller for passive system
  • Linear neuron on-line learning adaptive control method and controller for passive system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0039] An embodiment of the present invention provides a passive system linear neuron online learning adaptive control method, see figure 1 , the method flow is as follows:

[0040]101: Establish a linear neuron model, the linear neuron model is:

[0041] Its vector is expressed as Δy(k+1)=Φ(k)·ΔU T (k)+θ;

[0042] 102: According to the linear neuron model, and the online input and output data of the passive system at time k and before k, derive the linear neuron weight online learning algorithm:

[0043] Φ ^ ( k ) = Φ ^ ( k - 1 ) + ΔU ( k - 1 ) μ + ΔU ( k - ...

Embodiment 2

[0050] An embodiment of the present invention provides a passive system linear neuron online learning adaptive control method, see figure 2 , the method flow is as follows:

[0051] 201: Establish a linear neuron model, the linear neuron model is:

[0052] Its vector is expressed as Δy(k+1)=Φ(k)·ΔU T (k)+θ;

[0053] Wherein, the characteristic of the linear neuron model is that the excitation function of the neuron is taken as a simple linear function y=x.

[0054] Specifically, considering the SISO (Single Input Single Output, single input single output) discrete dynamic nonlinear system Y=F(·), both can be expressed as:

[0055] Δy ( k + 1 ) = Σ i = 0 n w i ( k ) Δu ( k - i...

Embodiment 3

[0150] see Figure 7 , the embodiment of the present invention provides a passive system linear neuron online learning adaptive control controller, the controller includes:

[0151] The linear neuron module 701 is used to establish a linear neuron model, and the linear neuron model is:

[0152] Its vector is expressed as Δy(k+1)=Φ(k)·ΔU T (k)+θ;

[0153] The weight online learning algorithm module 702 is used to derive the linear neuron weight online learning algorithm according to the linear neuron model and the online input and output data of the passive system at time k and before k:

[0154] Φ ^ ( k ) = Φ ^ ( k - 1 ) + ΔU ( k - 1 ) μ ...

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 linear neuron on-line learning adaptive control method and a controller for a passive system, belonging to the field of process control. The method comprises the following steps of: establishing a linear neuron model; deriving a linear neuron weight on-line learning algorithm according to the linear neuron model and on-line input and output data of the passive system at the time of k and before k; deriving a linear neuron on-line learning adaptive control algorithm according to the linear neuron model and the linear neuron weight on-line learning algorithm; and controlling a closed loop of the passive system according to the linear neuron on-line learning adaptive control algorithm. The controller comprises a linear neuron module, a weight on-line learning algorithm module, an on-line learning adaptive control algorithm module and a control outputting module. The invention controls the closed loop of the passive system by proposing the linear neuron on-line learning adaptive control algorithm and giving the implementation method of the controller on a configuration platform through the algorithm.

Description

technical field [0001] The invention relates to the field of process control, in particular to a passive system linear neuron online learning adaptive control method and a controller. Background technique [0002] At present, most closed-loop controls in industrial process control still use the PID (Proportion Integration Differentiation, proportional integral differential) control algorithm. The essence of PID control is a linear controller based on a linear combination of proportional, integral, and differential. Conventional PID Control is suitable for industrial process systems with simple controlled objects, few external disturbance factors, and the system works under stable conditions. However, the simple structure and fixed parameters in PID do not match the complex controlled objects. The fixed parameter settings can only make the entire control system work near a small equilibrium point, so that PID control cannot be satisfied with The controlled object has a compl...

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): G05B13/04
Inventor 李传庆
Owner STATE NUCLEAR ELECTRIC POWER PLANNING DESIGN & RES INST CO LTD
Features
  • Generate Ideas
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com