By Topic

Process optimization using a fuzzy logic response surface method

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

4 Author(s)
Xie, H. ; Intel Corp., Chandler, AZ, USA ; Lee, Y.C. ; Mahajan, R.L. ; Su, R.

A new response surface method using fuzzy logic models (FL-RSM) has been proposed. The algorithm starts with a fuzzy logic model (FLM) constructed on the experimental data obtained with design of experiments (DOE). The gradient search method is used with a specified step size, and a confirming experiment is conducted at each step. The search continues until no further improvement in the objective function is observed in that gradient direction. The FLM is trained with the new experimental data combined with the old DOE data, and a new gradient is evaluated. The process is repeated until the working point is close to the optimum, as indicated by a marginal improvement in the objective function. Then the algorithm switches to the optimum search mode. It calculates the optimum based on the model, and a confirming experiment is conducted at the suggested optimum settings. The procedure is repeated until the exit criterion is satisfied. The optimization procedure has been applied to a vertical chemical vapor deposition (CVD) process with various noise levels. The results demonstrate the effectiveness of the proposed FL-RSM. It is similar to the existing regression-model-based RSM approaches. The main difference is that it uses one self-adjusted FLM to replace the combination of linear and nonlinear regression models. As a result, FL-RSM can be more user friendly and efficient in many applications

Published in:

Components, Packaging, and Manufacturing Technology, Part A, IEEE Transactions on  (Volume:17 ,  Issue: 2 )