Fault diagnosis in LED illuminating circuits based on cloud model

Qi Liu; Hong-Dong Zhao; Jie Zhao

doi:10.1117/12.868885

15 November 2010 Fault diagnosis in LED illuminating circuits based on cloud model

Qi Liu, Hong-Dong Zhao, Jie Zhao

Proceedings Volume 7852, LED and Display Technologies; 78520T (2010) https://doi.org/10.1117/12.868885
Event: Photonics Asia 2010, 2010, Beijing, China

Abstract

For gaining effective features to realize fault diagnosis in the LED illuminating circuits, a method of fault diagnosis in analog circuits based on cloud model is proposed. In this paper, the analog circuit with a sinusoidal input is simulated and its output is sampled to extract sequences of each layer of wavelet coefficients as the initial fault feature vectors. Then, the backward cloud algorithm of cloud model is applied to obtain corresponding digital features of the wavelet coefficients, which include the Expected value E_x, the Entropy E_n and the Hyper entropy H_e as new fault feature vectors, named fault cloud feature vectors. Finally, Fault cloud feature vectors are used to BP neural networks to classify fault and realize the fault diagnosis in analog circuits. The simulation result on the LED illuminating circuits shows that this method is feasible and has many powerful virtues, such as diagnosing and locating faults quickly and exactly.

Citation Download Citation

Qi Liu, Hong-Dong Zhao, and Jie Zhao "Fault diagnosis in LED illuminating circuits based on cloud model", Proc. SPIE 7852, LED and Display Technologies, 78520T (15 November 2010); https://doi.org/10.1117/12.868885

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available