کنفرانس بین المللی ماشین و محرکه های الکتریکی

| ورود به سامانه | ثبت نام

صفحه اصلی / icemd2023

Open-Circuit Fault Diagnosis of Fault-Tolerant DAB Converter for Improving Smart Transformers Reliability using 1-D CNN

نویسندگان :

Peyman Sheikh Ghomi¹ Milad Babalou² Hossein Torkaman³

1- Shahid Beheshti University Tehran, Iran 2- Shahid Beheshti University Tehran, Iran 3- Shahid Beheshti University Tehran, Iran

کلمات کلیدی :

Smart transformer (ST)،dual-active bridge (DAB)،open-circuit fault (OCF)،fault diagnosis (FD)،convolutional neural network (CNN)

چکیده :

This paper addresses the challenges brought about by rapid population growth and increasing demand for electrical energy. It focuses on diagnosing open-circuit faults (OCFs) in the dual active bridge (DAB) converter, a crucial component of smart transformers (STs) in smart grids (SGs). The DAB converter's reliability is paramount for the safe operation of power electronic systems, as power semiconductor devices are prone to failures, particularly those causing OCFs. Traditional model-based diagnosis methods have limitations, which led to the exploration of data-driven alternatives such as the convolutional neural network (CNN). The proposed 1-D CNN model displays exceptional performance, surpassing other notable machine and deep learning techniques in accuracy, recall, precision, and F1-score. The confusion matrix analysis confirms the 1-D CNN model's superior predictive abilities in accurately distinguishing fault types, especially in challenging situations. In summary, the research improves STs' reliability and performance in SGs by effectively diagnosing OCF in the DAB converter.