|
|
 |
| |
|
|
|
Predictive Maintenance for Vehicle Performance using Bidirectional LSTM |
|
|
|
PP: 1469-1479 |
|
|
|
doi:10.18576/amis/180623
|
|
|
|
Author(s) |
|
|
|
Asokan Vasudevan,
K. Gandhimathi,
Suleiman Ibrahim Mohammad,
M. Harsavarthini,
N. Raja,
Eddie Eu Hui Soon,
Ahmad A. Abu-Shareha,
Muhammad Turki Alshurideh,
|
|
|
|
Abstract |
|
|
| The advancement of sensor and network technologies has led to an abundance of condition- monitoring and performance data, particularly in the automotive sector. This data and big data analytics offer opportunities to enhance predictive maintenance strategies. Various data preprocessing techniques, such as handling missing values and data normalization, are involved before the data is fed into the algorithm. The Feature selection process and Data splitting process also play a major role in determining which attributes in the data are more important and splitting the data for the testing and training process. The evolution of Deep Learning (DL) techniques becomes achievable to address potential equipment failures like a brake pad, fuel consumption, tire rotation, crankshaft detection, etc., and estimate the remaining useful life of the vehicle by using the algorithm bidirectional Long Short-Term Memory (LSTM) deep networks as a primary algorithm for predictive maintenance in vehicles.
|
|
|
|
|
|
|
|
