pisco_log
banner

Deep Learning Analog Circuit Fault Diagnosis Based on Self-attention Mechanism

Fangmin Shi

Abstract


This paper discusses the application of deep learning based on self-attention mechanism in analog circuit fault diagnosis, aiming to
solve the limitation of traditional diagnosis methods in complex fault identification. The research team designed an optimized deep learning
model that enhances multi-classification capabilities by introducing SoftMax layers, utilizes Dropout mechanism to mitigate overfitting, innovatively converts circuit signals into spectrograms as input, and combines location coding with class coding to improve the sequence processing capability of the self-attention mechanism. In the experiment, Sallen-Key low-pass filter was selected as the test object, and various data
sets including 24 fault types were generated by simulation software, which fully covered single fault and double fault cases. Transfer learning
strategy was adopted in model training, and the customized AlexNet network showed high accuracy in different fault classification. The results show that this method can effectively improve the accuracy and efficiency of fault diagnosis, and is of great significance to promote the
development of analog circuit fault detection technology.

Keywords


Self-attention mechanism; Deep learning; Analog circuit fault diagnosis; Sallen-Key Low-pass filter

Full Text:

PDF

Included Database


References


[1] HUYY, PENG MF, TIAN CL, et al. Analog circuit fault diagnosis using multi-wavelet transform and SVM[C]. Third International Conference on Digital Manufacturing & Automation, IEEE, 2012:214-217.

[2] LIU H, CHEN G, SONG G, et al. Analog circuit fault diagnosis using bagging ensemble method with cross-validation[ C]. International

Conference on Mechatronics and Automation, IEEE, 2009:4430-4434.

[3] SHOKROLAHI SM, KARIMIZIARANI M. A deep network solution for intelligent fault detection in analog circuit [J]. Analog Integrated Circuits and Signal Processing, 2021, 107(3):597-604.

[4] Shengdong W, Zhenbao L, Zhen J, et al. Intermittent fault diagnosis of analog circuit based on enhanced one-dimensional vision transformer and transfer learning strategy[J]. Engineering Applications of Artificial Intelligence, 2024(2), 127.

[5] Suman B, Kumar G M, Nilanjan C.Investigation of Extreme Learning Machine-Based Fault Diagnosis to Identify Faulty Components in

Analog Circuits[J]. Circuits, Systems, and Signal Processing, 2023, 43(2):711-728.

[6] Saravana R R, Cecil L M P.Diagnosis of Analog and Digital Circuit Faults Using Exponential Deep Learning Neural Network[J]. Journal of Electronic Testing, 2023, 39(4):421-433.

[7] YU WX, SUI Y, WANG J. The faults diagnostic analysis for analog circuit based on FA-TM-ELM[J]. Journal of Electronic Testing,

2016, 32(4):459-465.

[8] ZHANG A, CHEN C. Fault diagnosis based semi-supervised global LSSVM for analog circuit [C]. 2014 International Conference on

Mechatronics and Control (ICMC), IEEE, 2014:744-748.

[9] Zhen L, Xuemei L, Songlin X, et al.A Novel Fault Diagnosis Method for Analog Circuits Based on Multi-Input Deep Residual Networks with an Improved Empirical Wavelet Transform[J]. Applied Sciences, 2022, 12(3):1675-1675.

[10] Fangli L, Yanbo W, Song X, et al.Fault Diagnosis of Permanent Magnet Synchronous Motor Inter Turn Short Circuit Based on Deep Reinforcement Learning[J]. Journal of Physics: Conference Series, 2021(1):2137.

[11] Yang Y, Wang L, Nie X, et al.Incipient fault diagnosis of analog circuits based on wavelet transform and improved deep convolutional

neural network:LETTER[J]. IEICE Electronics Express, 2021, 18(13):20210174-20210174.

[12] Alireza M, Mohamad S K.Simultaneous fault localization and detection of analog circuits using deep learning approach[J]. Computers

and Electrical Engineering, 2021, 92.

[13] Yuanjiang L, Yanbo W, Yi Z, et al.Diagnosis of Inter-turn Short Circuit of Permanent Magnet Synchronous Motor Based on Deep learning and Small Fault Samples[J]. Neurocomputing, 2021, 442348-358.




DOI: http://dx.doi.org/10.70711/aitr.v2i4.4875

Refbacks

  • There are currently no refbacks.