Presenter(s)
Rebecca L. Greider
Files
Download Project (286 KB)
Description
This study utilizes a working artificial neural network (ANN) to monitor an industrial injection molding process. This ANN will be able to adapt and learn using training data obtained from the process. Outputs will be classified as normal or not normal based uponannotations made on the data by a plant engineer. This network will be able to recognize patterns in the data it analyzes and will also be able to model complex relationships in the data. The goal is to use the ANN to predict a future unusable part. ANN performance will beevaluated on how far in advance it can reliably predict an unusable part: several parts in the future versus the next one to be produced.
Publication Date
4-18-2012
Project Designation
Honors Thesis
Primary Advisor
Michael J. Elsass
Primary Advisor's Department
Chemical and Materials Engineering
Keywords
Stander Symposium project
Recommended Citation
"Artificial Neural Networks and Their Use in Process Monitoring and Diagnosis of an Industrial Injection Molding Process" (2012). Stander Symposium Projects. 73.
https://ecommons.udayton.edu/stander_posters/73
![Artificial Neural Networks and Their Use in Process Monitoring and Diagnosis of an Industrial Injection Molding Process](https://ecommons.udayton.edu/stander_posters/1072/thumbnail.jpg)