Application of PA-LSTM-FCN model in fault diagnosis of circuit board

Li Wang; Yichi Zhang; Xiaohuai Xie; Xuepeng Liu

doi:10.1117/12.2652543

10 November 2022 Application of PA-LSTM-FCN model in fault diagnosis of circuit board

Li Wang, Yichi Zhang, Xiaohuai Xie, Xuepeng Liu

Proceedings Volume 12331, International Conference on Mechanisms and Robotics (ICMAR 2022); 1233137 (2022) https://doi.org/10.1117/12.2652543
Event: International Conference on Mechanisms and Robotics (ICMAR 2022), 2022, Zhuhai, China

Abstract

When processing printed circuit board (PCB) defects using an infrared fault diagnostic system, we found that the small dimensionality of the recorded PCB temperature data and the difficulty of extracting some features of the error mode may lead to low fault diagnosis accuracy. To solve this problem, based on the temperature data of PCB components, a PALSTM- FCN(Parall-Attention-LSTM-FCN) board error diagnostic model based on multidimensional temperature data was proposed. First, we use the temperature data acquisition system to collect chip temperature information for different failure modes. Then, a multidimensional chip temperature dataset is created to optimize the diagnostic model input data features. Finally, the PA-LSTM-FCN model is trained to extract the temperature data features to complete the circuit board fault diagnosis. The experiment uses an adjustable regulated power supply circuit board for reliability analysis. The results show that the diagnostic accuracy of the model proposed in this paper reached 97.34%, achieving high accuracy diagnosis of circuit board failure modes.

Citation Download Citation

Li Wang, Yichi Zhang, Xiaohuai Xie, and Xuepeng Liu "Application of PA-LSTM-FCN model in fault diagnosis of circuit board", Proc. SPIE 12331, International Conference on Mechanisms and Robotics (ICMAR 2022), 1233137 (10 November 2022); https://doi.org/10.1117/12.2652543

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