An Improved IGA-BP Algorithm Applied to Fault Diagnosis

In immune genetic algorithm of float coding, there would be no certain way to calculate the critical value based on the Euclidean distance similarity. For this problem, the paper puts forward a way on "3σ" method; inspired by the elitist strategy, the paper modifies the coefficient of arit...

Full description

Saved in:
Bibliographic Details
Published in2012 Third International Conference on Digital Manufacturing & Automation pp. 194 - 197
Main Authors Lichun Wu, Li Zhang, Yongbo Yang, Lijie Sun, Fengman Li
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.07.2012
Subjects
Arithmetic crossover
Encoding
Euclidean distance
Fault diagnosis
Genetic algorithms
Immune Genetic Algorithm
Neural Network
Neural networks
Self-Adaption
Sociology
Statistics
Online AccessGet full text

Cover

More Information
Summary:In immune genetic algorithm of float coding, there would be no certain way to calculate the critical value based on the Euclidean distance similarity. For this problem, the paper puts forward a way on "3σ" method; inspired by the elitist strategy, the paper modifies the coefficient of arithmetic crossover method, which can adaptively incline toward the higher fitness value of two pairs of antibody. The improved IGA-BP algorithm is applied to the fault diagnosis of transmission gear. The experiment is simulated on MATLAB, and its result shows that the "3σ" method is an effective solution, and this improved algorithm has several advantages such as fast convergence, good adaptability, high accuracy on transmission gear's fault types.
ISBN:9781467322171
1467322172
DOI:10.1109/ICDMA.2012.47