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

Nerve network optimization controller and control method

A neural network and optimal control technology, applied in the direction of adaptive control, general control system, control/regulation system, etc., can solve problems such as failure to consider control quantity constraints, inability to achieve optimal control, and inability to measure

Inactive Publication Date: 2005-03-23
BEIJING JIAOTONG UNIV
View PDF0 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This idea has a certain reference value for solving the problem of optimal control of complex process objects, because complex process objects generally lack accurate mathematical models, making optimal control impossible
However, using this method to solve complex process control needs to solve the following problems: 1) The training structure and algorithm of the neural network controller are determined based on the state space model, and the state variables are required to be measurable. However, in general complex process control, the state variables Or it does not exist or cannot be measured, and the input-output model is mainly used; 2) It does not consider the constraints of the control variables, but the control variables in complex process control have constraints; 3) It only solves the problem of single-objective optimal control, while in complex process control It is also necessary to consider the multi-objective optimal control problem

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
  • Nerve network optimization controller and control method
  • Nerve network optimization controller and control method
  • Nerve network optimization controller and control method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0029] Embodiment 1: a neural network optimization control method, comprising:

[0030] The controller neural network is used to calculate the optimized control signal sequence according to the complex process object output signal sequence;

[0031] A training structure for converting the optimal control law synthesis problem into a combined network training problem consisting of a controller neural network and a complex process object simulator neural network;

[0032] A training algorithm, deriving an equivalent error calculation formula for the connection between different controller neural networks and complex process object simulator neural networks in the combined network, which is used to train the combined network to obtain an optimized controller neural network;

[0033] The training structure divides the entire control process time into at least 2 time units, and connects the neural networks of at least 2 time units according to the corresponding relationship between...

Embodiment 2

[0084] Example 2: Induction heating involves physical processes such as magnetism, electricity, and heat conduction. Due to the complexity of the mechanism, it is difficult to propose an accurate mathematical model, and the traditional optimal control cannot be realized. The neural network optimization control method can be used to achieve optimal control of the heating process. The control system is as follows: image 3 shown.

[0085] The induction heating neural network optimization control system implemented according to the present invention refers to Figure 1-Figure 8 design. Determine the input-output model structure according to the controlled object as n=1, m=3, the object simulator network is a 4-input and 1-output neural network; the controller neural network structure is π(1×4×1), such as figure 1 Shown; the number of time units in the control process is 5. Combining networks and their training structures such as figure 2 shown, where C 0 、C 1 、C 2 、C 3 、C...

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 neural network optimizing controlling equipment and method. The invention includes controller neural network which is used to calculate optimizing control signal serial according to complicated process object output signal serial; training structure equipment which is used to transform optimizing control rule integrated problem to combining network training problem, composed by controller neural network and complicated process object emulator neural network; training arithmetic equipment, which deducts jointing equal error calculating formula among different controller neural networks and complicated process object emulator neural networks in the combining network, and is used to train the combining network to get optimized controller neural network. The control method of the invention sets the consistency function of training network output and perfect output as guideline function of optimizing control, the invention has the excellence of achieving multiobjective control to complicated process objects which have no precise mathematical model, under restricted controlling condition, with intelligent method.

Description

Technical field [0001] The invention relates to a neural network optimization controller and a control method. Background technique [0002] Neural network control belongs to intelligent control. Its physical implementation is simple, and compared with other types of intelligent control, it is more practical. It has been successfully applied in inverted pendulum control, manipulator control and underwater remote control robot longitudinal attitude adjustment. However, it has not been reported in the complex process control lacking mathematical models, such as industrial process control. [0003] One of the representative results of neural network control is the method proposed by Nguyen1990 (Derrick H, Nguyen, et al. Neural Networks for Self-learning Control Systems. IEEE Cont. Sys. Mag. 1990, (4): 18-23). In this paper, a multi-layer feed-forward network is used to identify the controlled object, another neural network is used as the controller, and the neural network con...

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): G05B13/00G05B13/02
Inventor 朱刚姜丽君赵冬梅谢永斌
Owner BEIJING JIAOTONG UNIV
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