Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Deviation punishment enhanced stacked automatic encoder processing method and device

A technology of stacking autoencoders and autoencoders, applied in the field of data analysis

Pending Publication Date: 2021-07-30
CENT SOUTH UNIV
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The embodiment of the present application provides an enhanced stacked autoencoder processing method and device with deviation penalty, so as to at least solve the problems existing in the existing stacked autoencoder detection faults

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
  • Deviation punishment enhanced stacked automatic encoder processing method and device
  • Deviation punishment enhanced stacked automatic encoder processing method and device
  • Deviation punishment enhanced stacked automatic encoder processing method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0021] It should be noted that, in the case of no conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and embodiments.

[0022] It should be noted that the steps shown in the flowcharts of the accompanying drawings may be performed in a computer system, such as a set of computer-executable instructions, and that although a logical order is shown in the flowcharts, in some cases, The steps shown or described may be performed in an order different than here.

[0023] The following concepts are involved in the following examples:

[0024] False alarm rate (FAR) and fault detection rate (FDR) are two commonly used criteria for evaluating detection performance. FAR represents the ratio of false alarms to the entire normal data, while FDR represents the ratio of true alarms to the entire fault data.

[0025] Th...

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 deviation punishment enhanced stacked auto-encoder processing method and device, and the method comprises the steps: obtaining the dispersity between a neuron and other neurons, the dispersity being used for indicating the distance between the neuron and other neurons in the same layer, and the neurons being neurons in a stacked auto-encoder; configuring a first penalty coefficient corresponding to each layer for the layer, wherein the first penalty coefficient is used for adjusting the dispersity between the layers; configuring a second penalty coefficient for the reconstruction error, the second penalty coefficient being used for adjusting the reconstruction precision; and configuring the first penalty coefficient and the second penalty coefficient in the stacked auto-encoder. According to the stacking automatic encoder fault detection method and device, the problem existing in fault detection of an existing stacking automatic encoder is solved, and therefore the detection performance is optimized to a higher level.

Description

technical field [0001] The present application relates to the field of data analysis in industrial processes, specifically, an enhanced stacked autoencoder processing method and device for deviation penalty. Background technique [0002] Nonlinearities dominate in industrial processes when large-scale fluctuations or process regulations occur. Fault detection strategies based on conventional multivariate statistical analysis (MVA), such as principal component analysis (PCA) and canonical variable analysis (CVA), provide systematic statistical interpretation for modeling and designing detection metrics. However, they cannot handle nonlinearities. Therefore, nonlinear extensions of MVA based on kernel tricks, such as kernel PCA and kernel CVA, were developed. A natural drawback of these kernel methods is that the detection results are quite sensitive to the kernel function and corresponding hyperparameters. Kernel methods applied to process monitoring as nonparametric techn...

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): G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/047G06F18/2415G06F18/241
Inventor 王凯曹子卉王雅琳袁小锋
Owner CENT SOUTH UNIV
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
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