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

A data-driven dynamic modeling method for deep neural network systems with large time delays

A deep neural network and dynamic modeling technology, applied in the field of dynamic modeling of deep neural network large time-delay systems, can solve the problems of slow update of weights and low efficiency.

Active Publication Date: 2022-08-02
CHINA SOUTH IND GRP AUTOMATION RES INST +1
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to overcome the deficiencies of the prior art, to provide a dynamic modeling method based on data-driven deep neural network large time-delay system, to solve the problem that the neural network easily encounters slow weight update, The problem of inefficiency

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 data-driven dynamic modeling method for deep neural network systems with large time delays
  • A data-driven dynamic modeling method for deep neural network systems with large time delays
  • A data-driven dynamic modeling method for deep neural network systems with large time delays

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0019] The specific embodiments of the present invention are described below with reference to the accompanying drawings, so that those skilled in the art can better understand the present invention. It should be noted that, in the following description, when the detailed description of known functions and designs may dilute the main content of the present invention, these descriptions will be omitted here.

[0020] When the large delay system is a multi-input system, the mechanism analysis is used to analyze the multiple input data, and the relevant input data that affects the output of the system are preselected. However, considering the influence of various factors such as system operating conditions and external interference, the adaptive LASSO algorithm is used to re-screen the pre-selected input data.

[0021] The adaptive LASSO algorithm is as follows:

[0022]

[0023] in γ>0. It can be seen from the above formula that λ as the weight value of the penalty item o...

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 deep neural network large time-delay system dynamic modeling method, which mainly relates to a data preprocessing method and a novel deep neural network modeling technology. The invention discloses a deep neural network-based modeling method for a large time delay system, aiming at the problems of high modeling difficulty, strong coupling, and multiple disturbances that occur in large time delay systems common in forming control and industrial manufacturing. In the process of data preprocessing, the large time-delay characteristics of the system response are considered, and the shortcomings of poor weight update accuracy and slow speed of the deep neural network algorithm are optimized to enhance its generalization ability. The invention can effectively solve the problem that the mechanism model of the large time delay system is difficult to establish, and the obtained deep network model also has good generalization performance.

Description

technical field [0001] The invention belongs to the technical field of nonlinear system modeling, and more particularly relates to a data-driven deep neural network dynamic modeling method for large time-delay systems. Background technique [0002] Nonlinear large-delay systems are ubiquitous in most of many industrial processes today. The modeling, identification and control of large-time-delay systems have received extensive attention due to the long delay time of the system, many inputs and outputs, easy to be affected by interference, and the internal structure is complex and difficult to analyze. [0003] The choice of system modeling method directly affects the accuracy and reliability of system modeling. Nowadays, many nonlinear modeling methods such as least squares estimation and local dynamic linearization have been widely used in various industrial occasions. With the development of machine learning, machine learning methods such as support vector machines, core...

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
CPCG05B13/042
Inventor 黄求安高丰陈勇伍凌川韩智鹏鲁前成刘越智
Owner CHINA SOUTH IND GRP AUTOMATION RES INST
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