Data mining methods for data pattern discovery and prediction
Machine critical component failures are the reason for significant process downtime as well as costly repair work. Therefore, detecting early degradation and faults will improve the product reliability as well as reduce the cost of downtime and repairs by carrying out necessary maintenance timely. T...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project |
| Language: | English |
| Published: |
2013
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/51074 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | Machine critical component failures are the reason for significant process downtime as well as costly repair work. Therefore, detecting early degradation and faults will improve the product reliability as well as reduce the cost of downtime and repairs by carrying out necessary maintenance timely. Theoretically, a good degradation detection and analysis tool can contribute to a near zero loss for potential failures. In this report, the author descript statistical approaches to detect anomalous component and monitor the degradation status of machine critical component in order to predict the remaining useful life of the machine critical component based on the features collected, extracted and selected in previous steps. This approach is implemented in Visual C# and predicted results are compared with actual experimental result for verification of the usefulness of this approach. |
|---|