It has been shown that a manufacturing process can be modeled (learned) using Multi-Layer Perceptron (MLP) neural network and then optimized directly using the learned network. This paper extends the previous work by examining several different MLP training algorithms for manufacturing process modeling and three methods for process optimization. The transformation method is used to convert a constrained objective function into an unconstrained one, which is then used as the error function in the process optimization stage. The simulation results indicate that: (i) the conjugate gradient algorithms with backtracking line search outperform the standard BP algorithm in convergence speed; (ii) the neural network approaches could yield more accurate process models than the regression method; (iii) the BP with simulated annealing method is the most reliable optimization method to generate the best optimal solution, and (iv) process optimization directly performed on the neural network is possible but cannot be especially automated totally, especially when the process concerned is a mixed integer problem.

Issue Section:
Research Papers
Topics:
Manufacturing, Modeling, Multilayer perceptron, Optimization, Artificial neural networks, Algorithms, Error functions, Gradient methods, Simulated annealing, Simulation results
1.
Ackley, D. H., Hinton, G. E., and Seinowski, T. J., “A Learning Algorithm for Boltzmann Machines,” Cognitive Science, Vol. 9, 1985.
2.
Barnard
E.
, “
Optimization for Training Neural Nets
,”
IEEE Trans. on Neural Networks
, Vol.
3
, No.
2
,
1992
, pp.
232
–
240
.
3.
Battiti
R.
, “
First- and Second-Order Methods for Learning: Between Steepest Descent and Newton’s Method
,”
Neural Computation
, Vol.
4
,
1992
, pp.
141
–
166
.
4.
Bello
M. G.
, “
Enhanced Training Algorithms, and Integrated Training/Architecture Selection for Multilayer Perceptron Networks
,”
IEEE Trans. on Neural Networks
, Vol.
3
, No.
6
,
1992
, pp.
864
–
874
.
5.
Chang
T. S.
, and
Abdel-Ghaffar
K. A. S.
, “
A Universal Neural Net with Guaranteed Convergence to Zero System Error
,”
IEEE Trans. on Signal Processing
, Vol.
40
, No.
12
,
1992
, pp.
3022
–
3031
.
6.
Chen, L. J., “Multi-Layered Perceptron Network Training Algorithms and Their Applications in Manufacturing Process Modelling and Optimization,” M.S. Thesis, Electrical & Computer Engineering Department, Louisiana State University, Dec, 1993.
7.
Chryssolouris
G.
, and
Guillot
M.
, “
A Comparison of Statistical and AI Approaches to the Selection of Process Parameters in Intelligent Machining
,”
ASME JOURNAL OF ENGINEERING FOR INDUSTRY
, Vol.
112
,
1990
, pp.
122
–
131
.
8.
Dennis, J. E., Jr., and Schnabel, R. B.,Numerical Methods for Unconstrained Optimization and Nonlinear Equations, Englewood Cliffs, NJ, Prentic Hall, 1983.
9.
Hirose
Y.
,
Yamashita
K.
, and
Hijiya
S.
, “
Back-Propagation Algorithm Which Varies the Number of Hidden Unit
,”
Neural Networks
, Vol.
4
,
1991
, pp.
61
–
66
.
10.
Huang
T.
,
Zhang
C.
,
Lee
S.
, and
Wang
H.-P. (Ben)
, “
Implementation and Comparison of Three Neural Network Learning Algorithms
,”
Kybernetes
, Vol.
22
, No.
1
,
1993
, pp.
22
–
38
.
11.
Hush, D. R., and Home, B. G., “Progress in Supervised Neural Networks,” IEEE Signal Processing Magazine, January 1993, pp. 8–39.
12.
Ilhan
R. E.
,
Sathyanarayanan
G.
,
Storer
R. H.
, and
Liao
T. W.
, “
Offline Multiresponse Optimization of Electrochemical Surface Grinding by a Multiobjective Programming Method
,”
Int. J. Mech. Tools Manufact.
, Vol.
32
, No.
3
,
1992
, pp.
435
–
451
.
13.
Liao, T. W., “Creep Feed Grinding of Alumina with Diamond Wheels,” Ph.D. Dissertation, Leigh University, October, 1990.
14.
Liao
T. W.
, and
Chen
L. J.
, “
A Neural Network Approach for Grinding Processes: Modeling and Optimizing
,”
Int. J. Mech. Tools Manufact.
, Vol.
34
, No.
7
,
1994
, pp.
919
–
937
.
15.
Liao
T. W.
, “
MLP Neural Network Models of CMM Measuring Processes
,”
J. Of Intelligent Manufacturing
,
7
,
1996
, pp.
413
–
425
.
16.
Lippmann, R. P., “An Introduction to Computing with Neural Nets,” IEEE ASSP Magazine, April 1987, pp. 4–22.
17.
Metropolis, N., “Equation of State Calculation by Fast Computing Machines,” The J. Of Chemical Physics, Vol. 21, No. 6, June 1953.
18.
Monostori
L.
, and
Barschdorff
D.
, “
Artificial Neural Networks in Intelligent Manufacturing
,”
Robotics & Computer-Integrated Manufacturing
, Vol.
9
, No.
6
,
1992
, pp.
412
–
436
.
19.
Rangwala
S.
, and
Dornfeld
D. A.
, “
Learning and Optimization of Machining Operations Using Computing Abilities of Neural Networks
,”
IEEE Trans. on Systems, Man, and Cybernetics
, Vol.
19
, No.
2
,
1989
, pp.
299
–
314
.
20.
Reklaitis, G. V., Ravindran, A., and Ragsdell, K. M., Engineering Optimization: Methods and Applications, New York, Wiley, 1983.
21.
Rumelhart, D. E., Hinton, G. E., and Williams, R. J., “Learning Internal Representations by Error Propagation,” Parallel Distributed Processing: Explorations in the Microstructure of Cognition, Vol. 1: Foundations, D. E. Rumelhart & J. L. McClelland, eds., MIT Press, 1986.
22.
Sathyanarayanan
G.
,
Lin
I. J.
, and
Chen
M.-K.
, “
Neural Network Modelling and Multi-Objective Optimization of Creep Feed Grinding of Superalloys
,”
Int. J. Prod. Res.
, Vol.
30
, No.
10
,
1992
, pp.
2421
–
2438
.
23.
Troll
A.
, and
Feiten
W.
, “
Improving Neural Net Training by Mathematical Optimization
,”
Cybernetics and Systems: An International Journal
, Vol.
23
,
1992
, pp.
447
–
457
.
This content is only available via PDF.
Copyright © 1998
 by The American Society of Mechanical Engineers
You do not currently have access to this content.