An Intelligent Model for Residual life Prediction of Thyristor
Cherry Bhargava1, Jagdeep Singh2, Pardeep Kumar Sharma3

1Dr. Cherry Bhargava, SEEE, Lovely Professional University, Phagwara (Punjab), India.
2Dr. Jagdeep Singh, IIIEE, Lund University, Lund, Sweden.
3Pardeep Kumar Sharma, LSPS, Lovely Professional University, Phagwara (Punjab), India.

Manuscript received on 18 June 2019 | Revised Manuscript received on 25 June 2019 | Manuscript published on 30 June 2019 | PP: 1862-1866 | Volume-8 Issue-5, June 2019 | Retrieval Number: E7891068519/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (

Abstract: Modern age is the age of integration, where millions of electronic components are integrated and installed on a single chip, to minimize the size of device and automatically increases the speed. But, as a greater number of components are placed on a single device, reliability becomes a concern issue, as failure of one component can degrade the complete device. From dimmer to high voltage power transmission, thyristors are widely used. The failure of thyristor can be proven dangerous for mankind, so the reliability prediction of thyristor is highly desirable. This paper is based on the accelerated life testing based experimental technique for reliability assessment. An intelligent model is designed using artificial intelligence techniques i.e. ANN, Fuzzy and ANFIS and comparative analysis is conducted to estimate the most accurate technique. Fuzzy based Graphical User Interface (GUI) is framed which informs the user about the live status of thyristor under various environmental conditions. The intelligent techniques are validated using experimental technique. An error analysis is conducted to predict the most accurate and reliable system for residual life prediction of thyristor. Out of all prediction techniques, ANFIS has the highest accuracy i.e. 95.3%, whereas ANN and Fuzzy inference system has accuracy range 86.1% and 89.2% respectively.
Keywords: Artificial Intelligence, Accelerated life testing, Graphical user interface, Reliability Prediction, Thyristor

Scope of the Article: Structural Reliability Analysis