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

Complex equipment fault prediction method and system based on LPP-HMM method

A LPP-HMM, fault prediction technology, applied in prediction, neural learning methods, data processing applications, etc., can solve the problem that the signal processing method is difficult to effectively obtain fault characteristics, cannot identify intermittent faults and random faults, and is difficult to fully reflect the system state, etc. problems, to achieve the effect of improving the fault identification rate, facilitating fault identification, and achieving excellent results

Pending Publication Date: 2020-11-10
安徽三禾一信息科技有限公司
View PDF3 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] Fault diagnosis technology is the core content of PHM technology for complex equipment systems. Due to the nonlinear and time-varying characteristics of complex equipment systems, its fault signals have high-dimensional, nonlinear, non-stationary characteristics, and contain external noise or interference, resulting in Conventional signal processing methods are difficult to effectively obtain fault characteristics
In the diagnosis process, the traditional fault diagnosis method only divides the system state into normal state and fault state, which is difficult to fully reflect the system state
Unable to identify intermittent and random failures, unable to monitor intermediate states of the system, and more difficult to detect gradual failures

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
  • Complex equipment fault prediction method and system based on LPP-HMM method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0031] A complex equipment fault prediction method and system based on the LPP-HMM method, including the following steps;

[0032] S1. Data processing and feature extraction: collect sample data, and use the LPP algorithm to preprocess the sample data;

[0033] S2. Determine the hidden state number L of the HMM model according to the actual situation of the equipment, set the initial parameters of the model, establish the initial HMM model λ=(π,A,B), and perform maximum likelihood estimation for each state type , select any k eigenvectors to form a set of state sequences to train its HMM model until the convergence condition is met, the state type is a hidden state, and each state type needs to train an HMM model;

[0034] S3. Input the original feature information in the preprocessed data in S1 into the fuzzy neural network, and perform network calculation to select n HMM models with the strongest correlation;

[0035] S4. According to the state prediction information of the...

Embodiment 2

[0052] The present invention introduces nonlinear manifold learning into complex equipment fault diagnosis, and proposes a fault diagnosis method using LPP and HMM, which directly learns original high-dimensional fault data and converts complex high-dimensional space into low-dimensional features space to achieve dimensionality reduction and extract the inherent low-dimensional manifold features of the data. Through simulation experiments, it is shown that this method can better preserve the overall structure information in the fault data, which is beneficial to fault identification. The LPP algorithm can be used to reduce the nonlinear fault characteristics of analog circuits, and the effect is better than that of linear dimensionality reduction methods such as PCA. The hybrid HMM classifier reflects the real state of each stage of the system. A good recognition effect can effectively identify the early faults and intermediate states of the system. Combining the LPP method wi...

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 complex equipment fault prediction method and system based on an LPP-HMM method, belongs to the field of complex equipment fault prediction, introduces nonlinear manifold learning into complex equipment fault diagnosis, and provides a fault diagnosis method using LPP and HMM.According to the method, original high-dimensional fault data are directly learned; the complex high-dimensional space is converted into a low-dimensional feature space to achieve dimensionality reduction, and extracting intrinsic low-dimensional manifold features of the data. Simulation experiments show that the method can well reserve the overall structure information in the fault data, and is beneficial to fault identification. The LPP algorithm can be used for analog circuit nonlinear fault feature reduction, and the effect is better than that of PCA and other linear dimension reduction methods.

Description

technical field [0001] The invention relates to the field of complex equipment failure prediction, in particular to a complex equipment failure prediction method and system based on the LPP-HMM method. Background technique [0002] Fault diagnosis technology is the core content of PHM technology for complex equipment systems. Due to the nonlinear and time-varying characteristics of complex equipment systems, its fault signals have high-dimensional, nonlinear, non-stationary characteristics, and contain external noise or interference, resulting in Conventional signal processing methods are difficult to effectively obtain fault features. In the diagnosis process, the traditional fault diagnosis method only divides the system state into normal state and fault state, which is difficult to fully reflect the system state. Intermittent and random faults cannot be identified, intermediate states of the system cannot be monitored, and gradual faults are more difficult to monitor. ...

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): G06Q10/04G06N3/04G06N3/08G06K9/62
CPCG06Q10/04G06N3/08G06N3/045G06F18/214
Inventor 都竞李军梁天范文豪徐启胜江水张殷日
Owner 安徽三禾一信息科技有限公司
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