Development Of Fuzzy Logic Model For Turning Process Of Steel Alloy And Titanium Alloys
The study is about the application of fuzzy logic in representing the machinability data for the turning process. Machining is a very complex process with respect to the influences of the machining parameters such as cutting speed, feed rate, and depth of cut. In order to perform a good machining...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English English |
| Published: |
2004
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/5129/1/FK_2004_97.pdf http://psasir.upm.edu.my/id/eprint/5129/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The study is about the application of fuzzy logic in representing the machinability
data for the turning process. Machining is a very complex process with respect to the
influences of the machining parameters such as cutting speed, feed rate, and depth of
cut. In order to perform a good machining practice the proper selection of the
machinability data, which includes the machining parameters and cutting tools is
very important. Normally, the selection of the machinability data is done by the
skilled machinist. The manufacturer may face trouble without the presence of the
skilled machinists. Thus, there is a necessity to represent the knowledge of the skilled
machinists into model, so that any normal machinists will be able to perform a good
machining practice by retrieving the information which prescribed in the model.
Consequently, fuzzy logic was chosen as a tool to describe the strategy and action of
the skilled machinist.
In this study, two types of fuzzy models for different workpiece material have been
developed, and they are alloy steel and titanium alloys fuzzy models. Both fuzzy
models serve the purpose of predicting the appropriate cutting speed and feed rate with respect to the corresponding input variables. Generally, the development of
fuzzy model involves the design of three main elements, which are inputs
membership functions, fuzzy rules (inference mechanism), and output membership
functions. So far, there is no any clear procedure that can be used to develop these
three elements. Thus, the strategy for generalizing the development of alloy steel
fuzzy model has been suggested. This strategy is useful and less effort is required for
developing a related new fuzzy models.
The design of fuzzy rules is always the difficult part in developing the fuzzy model
due to the tedious way of defining fuzzy rules with the conventional method.
Therefore, a new method of developing fuzzy rules, namely fuzzy rule mapping has
been introduced and implemented. Through fuzzy rule mapping method, the effort
and the time required in developing the fuzzy rules has been reduced. This method
has been applied in the developing the fuzzy model for alloy steel.
All the predicted outputs (cutting speed and feed rate) from the alloy steel (with
general strategy and fuzzy rule mapping) and titanium alloys fuzzy models were
being compared with the data obtained from “Machining Data Handbook”, by
Metcut Research Associate, and a good match have been obtained throughout the
comparison. The average percentage errors for alloy steel fuzzy models with the
implementation of general strategy and fuzzy rule mapping are about the ranges of
3.1% to 5.6% and 3.0% to 10.7%, respectively. On the other hand, the average
percentage error for titanium alloys fuzzy model is about 1.8% to 5.1%. These results
have showed that the machinability data information for the turning of alloy steel and titanium alloys can be represented by fuzzy model. Besides that, it has also proved
the feasibility of using the suggested strategy and fuzzy rule mapping method. |
|---|