Remaining Useful Lifetime Prediction of Discrete Power Devices by Means of Artificial Neural Networks

Abstract

This work proposes a deep learning-based model for predicting the lifetime of power devices subjected to power cycling. To this purpose, a neural network based on bidirectional long short-term memory is adopted. The neural network is trained with experimental on-voltage degradation profiles. The application of the proposed method is based on the monitoring of a precursor, that is the on-voltage degradation. According to considered precursor, the model allows predicting the remaining useful lifetime (RUL) of power components. In order to prove the accuracy of the model, TO-247 power devices are stressed under power cycling and their wear-out is experimentally investigated. RUL predicted by the neural network is then compared with the experimental lifetime of power devices. Thanks to the proposed deep learning model, the accuracy in the lifetime estimation improves as long as more information about the state of health of the device under test is acquired.

Service States and Reliability Analysis of Low-Voltage Circuit Breaker Overload Protection Considering the Safety Margin

“…The matrix form is: P(t + ∆t) = P(t)P(∆t) (14) where the probability of the state i at time t pi(t) = P(X(t) = i), i = 1 ∼ 5, the state probabilities P(t) = (p 1 (t), p 2 (t), p 3 (t), p 4 (t), p 5 (t)), P(t + ∆t) = (p 1 (t + ∆t), p 2 (t + ∆t), p 3 (t + ∆t), p 4 (t + ∆t), p 5 (t + ∆t)).…”

“…In recent years, artificial intelligence algorithms have been rapidly developed that do not need to obtain the product degradation distribution and life distribution and thus have been more applied in the field of switchgear and electrical appliances. The reliability modeling method based on AI algorithms is a method to use the degradation data containing failure information monitored in real time during experiments as a test or training data set and input it into the corresponding AI model for training, analysis, and determination [14][15][16]. Data-driven modeling methods based on artificial intelligence mainly include fuzzy comprehensive judgment method, gray theory, support vector machines, artificial neural networks, convolutional neural networks, and chaos theory methods [17][18][19][20][21][22][23].…”