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

Gaussian mixture hidden Markov model and regression analysis remaining life prediction method

A hidden Markov and Gaussian mixture technology, which is applied in special data processing applications, instruments, electrical digital data processing, etc., can solve problems such as poor real-time performance, high cost, and long prediction period

Active Publication Date: 2015-04-08
HUNAN UNIV OF SCI & TECH
View PDF1 Cites 46 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0010] There are also problems such as high cost, long forecast period, and poor real-time performance in the crossover method produced by combining two or three of the above methods.

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
  • Gaussian mixture hidden Markov model and regression analysis remaining life prediction method
  • Gaussian mixture hidden Markov model and regression analysis remaining life prediction method
  • Gaussian mixture hidden Markov model and regression analysis remaining life prediction method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0059] Such as figure 1 As shown, the Gaussian mixture hidden Markov model and regression analysis method of the present invention for remaining life prediction. This assessment methodology includes:

[0060] (1) Use the wavelet packet algorithm to decompose and extract the feature vector. The equipment condition monitoring signal is processed by wavelet packet decomposition, and the characteristic information of different failure modes of parts is extracted as the observation value of Gaussian Mixture Hidden Markov Model (GMM-HMM).

[0061] (2) Establish a Gaussian mixture hidden Markov model library of different failure modes. The historical data of different failure modes is used for sample training, which is used to generate hidden Markov model libraries under different states of key components, which belongs to the offline learning stage.

[0062] (3) Establish failure time models of components with different failure modes. The probability statistics of the time histo...

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 Gaussian mixture hidden Markov model and regression analysis remaining life prediction method. The Gaussian mixture hidden Markov model and regression analysis remaining life prediction method comprises the specific steps of 1 performing characteristic vector extraction through wavelet packet algorithm decomposition, 2 establishing Gaussian mixture hidden Markov model libraries of different fault modes; 3 establishing part failure time modes of different fault modes; 4 performing part fault mode recognition and failure state assessment; 5 utilizing regression analysis to predict remaining life. The Gaussian mixture hidden Markov model and regression analysis remaining life prediction method integrates data drive and probability statistics fully utilizes respective advantages that the hidden Markov models predict the remaining life and have randomness and real-timeliness; the par failure process is divided into multiple stages, the probability statistics and the regression analysis are adopted to correct current usage time, and the remaining life prediction accuracy is improved. The Gaussian mixture hidden Markov model and regression analysis remaining life prediction method has the advantages of being high in prediction accuracy, operation speed and real-timeliness, low in cost and the like.

Description

technical field [0001] The invention belongs to the field of fault diagnosis and life prediction, in particular to a remaining life prediction method based on Gaussian mixture hidden Markov model and regression analysis. Background technique [0002] In the demand for the competitiveness of electromechanical products, in addition to continuously improving the practicability, reliability, and safety of products, it is also required to continuously reduce product maintenance and maintenance costs. At present, most electromechanical products adopt after-event maintenance and regular maintenance. Although these two maintenance methods are operable, because they do not take into account the actual operation and aging status of the equipment, the effect of these maintenance methods is actually not ideal, resulting in insufficient maintenance or maintenance. excessive phenomenon. [0003] Condition-Based Maintenance (CBM) is a viable solution that helps reduce the cost of traditio...

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): G06F19/00
Inventor 黄良沛吴超威黄傲黄华刘勇华
Owner HUNAN UNIV OF SCI & 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