A deep network solution for intelligent fault detection in analog circuit
- PDF / 927,802 Bytes
- 8 Pages / 595.276 x 790.866 pts Page_size
- 1 Downloads / 181 Views
A deep network solution for intelligent fault detection in analog circuit Seyed Moslem Shokrolahi1
•
Mohammadsepehr Karimiziarani2
Received: 17 January 2020 / Revised: 8 August 2020 / Accepted: 6 October 2020 Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Automatic fault diagnosis in analog electronic circuits is one of the interesting and important cases for researchers of this field that has gained substantial improvements in recent decades. Fault detection issue could be transferred into a classification problem. In this paper, a new fault detection method is proposed based on deep Convolution Neural Network (CNN). We used the real part of Power Spectrum Density (PSD) of faulty signals as the input images of CNN. The main reason for this is extracting microstructure features among signals by using PSD which result in a better discrimination amongst wide range of faults. Our method is evaluated by two benchmark circuits. The superior performance of our method is proved by simulation results and compared with other state of arts. Keywords Fault detection CNN Feature extraction Power spectrum density
1 Introduction Nowadays, electronic circuits have a very important role in our everyday lives and in meeting our various everyday needs. Every circuit is designed and manufactured for a special purpose. For each circuit, a special behavior and output is expected according to its elements, internal parameters and the applied input stimulation. Any change in these elements or the input, causes the circuit response and output to change. These unwanted changes are called ‘‘faults’’ and the circuit is considered ‘‘faulty’’. Sometimes the defective parts could be repaired or replaced and the circuit can be repaired through locating the defective elements. Differentiating faulty circuits from non-faulty ones and locating the defective elements is called ‘‘fault detection’’. Manufacturing of circuits is highly expensive and it is crucial to detect their faults. Automatic Fault Detection, Isolation and Reconfiguration (FDIR) with determining exact position in the electronic devices is a demanding and inseparable part of the
& Seyed Moslem Shokrolahi [email protected] 1
Electrical and Computer Engineering Isfahan, Isfahan University of Technology, Isfahan Blvd Isfahan, Isfahan, Iran
2
Department of Computer Science, The University of Alabama, Tuscaloosa, USA
production process. The reason is the difficulty of finding the faults position in the circuits with the conventional tools because circuits are increasingly complicating and developing so that the traditional methods for troubleshooting is so time consuming. Moreover, measuring current in a route requires cutting the circuit which is not possible most of the time. In recent researches, some effective methods have been presented for troubleshooting [1–3]. A large number of classic methods are proposed by using time domain analysis, frequency domain analysis, and DC analysis. Fault detection in electronic circuits by the nov
Data Loading...
