Modeling, Testing and Experimental Validation of Laser Machining Micro Quality Response by Artificial Neural Network
One way to reduce uncertainty in problem solving and decision making is by seeking the advice of an expert in related field. On the other hand, when we use computers to reduce uncertainty, the computer itself can become an expert in a specific field through a variety of methods. One such method is m...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IJENS Publishers
2009
|
| Subjects: | |
| Online Access: | http://eprints.utem.edu.my/id/eprint/9167/1/ANN-Laser.pdf http://eprints.utem.edu.my/id/eprint/9167/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | One way to reduce uncertainty in problem solving and decision making is by seeking the advice of an expert in related field. On the other hand, when we use computers to reduce uncertainty, the computer itself can become an expert in a specific field through a variety of methods. One such method is machine learning, which involves computer algorithm to capture hidden knowledge from data. In this research, a problem solving scenario for a metal cutting industry which faces some problems in determining the end product quality of Manganese Molybdenum (Mn-Mo) pressure vessel plate is investigated. Therefore, several real life machining scenarios with some expert knowledge input and machine technology features were incorporated. Three significant design parameters were used, namely; cutting speed, gas pressure and power. Artificial Neural network (ANN) has an ability to derive meaning from complicated data, and can be used to extract patterns and detect trends that are too complex to be noticed by either humans or other computational techniques. Therefore, prediction of laser machining cut quality, namely surface roughness was carried out using machine learning techniques based on Quick Back Propagation Algorithm using ANN. Experimentally observed responses were used to train, map and optimize the network algorithms before the best architecture was selected. Ten different architectures and models were tested and finally the best 3-8-1 model was finalized based on R square values. The model was then fed with new sets of machining parameters to experimentally validate the model’s ability in predicting the cut quality. The findings were found to be very promising and yielded excellent accuracy for both model and experimental validation reaching almost 88% and 92% respectively. |
|---|