PREDICTIVE ANALYTICS FOR EQUIPMENT FAILURE BY USING GATED RECURRENT UNIT – GENETIC ALGORITHM (GRU – GA)
Failure is described as an inability to attain a desired goal and acknowledged to be contradictory with success. It is a scenario that happens frequently across several industries and results in either minor or severe consequences such as maintenance expenses, production disruption and safety concer...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2023
|
| Subjects: | |
| Online Access: | http://utpedia.utp.edu.my/id/eprint/24633/1/ZahirahZainuddin_18003491.pdf http://utpedia.utp.edu.my/id/eprint/24633/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Failure is described as an inability to attain a desired goal and acknowledged to be contradictory with success. It is a scenario that happens frequently across several industries and results in either minor or severe consequences such as maintenance expenses, production disruption and safety concerns. The reasons behind this issue are always related to improper predictive maintenance, prolonged equipment operating hours and many other factors. Hence, the issue can be solved by adopting a prediction activity to monitor and predict the state of equipment in advance. Prediction predicts the upcoming instance by evaluating the assertions obtained from the gears. In this case, Deep Learning (DL) is chosen to construct the prediction activity for estimating the life expectancy of an equipment. Gated Recurrent Unit (GRU) algorithm is used to cater the predicting action of equipment state based on data from an oil and gas industry. |
|---|