Application of Factor Analysis Optimization Error Back Propagation Algorithm FA-BP Neural Network Equipment Fault Diagnosis Model

To enhance the precision of equipment fault diagnosis, a novel device fault diagnosis method based on factor analysis (FA) and an improved error back propagation (BP) algorithm has been proposed. An FA-BP neural network diagnosis model has been developed to enable intelligent diagnosis of equipment...

Full description

Saved in:
Bibliographic Details
Published in:2023 IEEE International Conference on Image Processing and Computer Applications (ICIPCA) pp. 32 - 37
Main Author: Liu, Yue
Format: Conference Proceeding
Language:English
Published: IEEE 11.08.2023
Subjects:
Analytical models
Backpropagation
BP neural network
Computational modeling
equipment fault diagnosis
error back propagation
factor analysis
Fault diagnosis
Neural networks
Training
Vibrations
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:To enhance the precision of equipment fault diagnosis, a novel device fault diagnosis method based on factor analysis (FA) and an improved error back propagation (BP) algorithm has been proposed. An FA-BP neural network diagnosis model has been developed to enable intelligent diagnosis of equipment faults. Firstly, three common factors influencing motor faults were extracted using FA, which served as effective input data for the BP neural network. Subsequently, the data that was optimized by factor analysis was utilized as the input layer of the BP neural network, and the data were trained and tested by employing Matlab software. Eventually, the results of fault diagnosis prediction obtained from the FA-BP neural network model were compared with those attained using the traditional BP neural network. The experimental findings affirm that the improved model achieves significantly elevated accuracy of equipment fault prediction and swifter training speed compared to the traditional error back propagation algorithm.
DOI:10.1109/ICIPCA59209.2023.10258011