NORMA eResearch @NCI Library

A Predictive Maintenance Framework for Remaining Useful Life Classification on IoT Enabled Devices using Neural Networks

Sainz Calderon, Raul Damian (2022) A Predictive Maintenance Framework for Remaining Useful Life Classification on IoT Enabled Devices using Neural Networks. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
PDF (Master of Science)
Download (2MB) | Preview
[thumbnail of Configuration manual]
PDF (Configuration manual)
Download (1MB) | Preview


The unprecedented technological advances in the fields of the Internet of Things (IoT), Artificial Intelligence (AI) are at the heart of the new industrial Revolution known as “Industry 4.0” which has pushed improvements in the manufacturing industry, especially in predictive maintenance research and technologies. With the simplistic premise of fixing a machine before it fails, companies and organizations can achieve great improvements in productivity by reducing downtime and machine failures. Remaining Useful Life (RUL) is a metric used to forecast the performance of the system to predict the time (cycles) left before the machine fails. However, the prognosis RUL can be a challenge, many researchers have used many model-based and data-driven models for trying to solve these issues. The proposed framework compares the performance of most widely used data-driven machine learning models LSTM and CNN versus the state of the art TCN to predict and classify if a machine will fail within a certain time range. The popular dataset CMAPSS from NASA is used for running the experiments of this research. The results show promising results for the proposed model TCN even with noisy and complex conditions datasets. Additionally, techniques for dealing with imbalanced data are compared throughout the experiments, the use of weights when training the models resulted in more stable training/accuracy plots and improvement of classification for imbalanced label classification.

Item Type: Thesis (Masters)
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
Q Science > QH Natural history > QH301 Biology > Methods of research. Technique. Experimental biology > Data processing. Bioinformatics > Artificial intelligence
Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Artificial intelligence
T Technology > TK Electrical engineering. Electronics. Nuclear engineering > Telecommunications > Computer networks > Internet of things
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Tamara Malone
Date Deposited: 10 Mar 2023 16:20
Last Modified: 10 Mar 2023 16:20

Actions (login required)

View Item View Item