Due to the high labour costs in industrialised countries and the increasing demands on the quality of plastic moulded parts produced by the injection moulding process, companies are challenged to increase their production efficiency. Data-based approaches provide an opportunity for systematic and objective process set-up and control. Especially the process set-up can be a tedious, iterative and complex task which can consume a significant amount of time in the preparation of mass production. Artificial neural networks (ANN) can be used by evolutionary algorithms to search for suitable machine setting parameters for the desired part and process quality features. This often results in locally optimal solutions, which do not guarantee the best possible machine settings. Invertible neural networks (INN) could be a solution as with the help of these models, globally optimal solutions may be identified for definitive optimisation as the entire solution space for given quality parameters can be mapped in short time. In our approach, data for the implementation of the INN is collected in practical tests. The resulting weight and several geometric dimensions of the parts are determined as quality parameters. Based on these data, ANN are trained and utilized for the generation of synthetic data for the training of the INN. The creation of synthetic data is necessary because INN have millions of trainable parameters and therefore degrees of freedom which need to be determined using a bigger amount of training data in comparison to traditional feedforward ANN. In this paper, two different databases with a size of 10,000 samples each are being investigated. The two databases differ in the distributions of holding pressure and cooling times in order to investigate the influenceability of the prediction quality of the INN by training data. The sample settings for the injection velocity, packing pressure, mould temperature and melt temperature are uniformly distributed. Two different types of conventional and conditional invertible neural networks are implemented based on literature recommendations. Measured on validation data, the model quality (R² value) for the prediction of the part weight is 0.995 and 0.975 for the part dimensions. The machine setting parameters proposed by the INN can be analysed in regard to their impact. The holding pressure time is identified as the machine setting with the greatest influence on both part weight and part dimension. The predictions of the INN are validated in practical experiments with an R² value of 0.997 for the part weight. The weight of the parts produced with the proposed machine settings deviates on average by 0.24 % from the target.

1.
C.
Brecher
,
S.
Jeschke
,
G.
Schuh
, A. J.,
J.
Arnoscht
,
F.
Bauhoff
,
C.
Fuchs
,
O.
Jooß
,
S.
Kouielski
,
S.
Orilski
,
A.
Richert
,
A.
Roderburg
,
M.
Schiffer
,
J.
Schubert
,
S.
Stiller
,
S.
Tönissen
and
F.
Welter
,
Integrative Produktionstechnik für Hochlohnländer
. (
Springer Verlag
,
Berlin, Heidelberg
,
2011
).
2.
A.
Meiabadi
,
A.
Vafaeesefat
and
F.
Sharifi
,
Journal of Optimization in Industrial Engineering
6
,
49
54
(
2013
).
3.
D. V.
Rosato
and
M. G.
Rosato
,
Injection Molding Handbook
. (
Springer US
,
New York
,
2012
).
4.
V. L.
Popov
,
M.
Heß
and
E.
Willert
,
Handbook of Contact Mechanics
. (
Springer Verlag
,
Berlin, Heidelberg
,
2019
).
5.
F.
Shi
,
Z. L.
Lou
,
J. G.
Lu
and
Y. Q.
Zhang
,
International Journal of Advanced Manufacturing Technologies
21
,
656
661
(
2003
).
6.
S.
Changyu
,
W.
Lixia
and
L.
Qian
,
Journal of Materials Processing Technology
187
,
412
418
(
2007
).
7.
W. C.
Chen
,
M. W.
Wang
,
C. T.
Chen
and
G. L.
Fu
,
International Journal of Advanced Manufacturing Technologies
44
,
501
511
(
2009
).
8.
C. C.
Chen
,
P. L.
Su
,
C. B.
Chiou
and
K. T.
Chiang
,
Materials and Manufacturing Processes
26
,
534
540
(
2011
).
9.
K.
Tsai
and
H.
Luo
,
Journal of Intelligent Manufacturing
28
,
473
487
(
2017
).
10.
R. J.
Bensingh
,
R.
Machavaram
,
S. R.
Boopathy
and
C.
Jebaraj
,
Measurement
134
,
359
374
(
2019
).
11.
Y.
Cao
,
X.
Fan
,
Y.
Guo
,
S.
Li
and
H.
Huang
,
Journal of Polymer engineering
40
,
360
371
(
2020
).
12.
R.
Spina
,
Journal of Achievements in Materials and Manufacturing Engineering
15
(
1-2
),
146
152
(
2006
).
13.
M. A.
Azman
,
A. M.
Zain
and
N. A. M.
Halimin
,
Journal of Soft Computing and Decision Support Systems
2
(
5
),
10
15
(
2015
).
14.
Y.
Xu
,
Q. W.
Zhang
,
W.
Zhang
and
P.
Zhang
,
The International Journal of Advanced Manufacturing Technology
76
,
2199
2208
(
2015
).
15.
M. H. N.
Hidayah
,
Z.
Shayfull
,
N. Z.
Noriman
,
S. M.
Sazli
,
R.
Norshahira
and
A. T. N. A.
Miza
,
AIP Conference Proceedings
2030
(
1
),
1
10
(
2018
).
16.
L.
Ardizzone
,
J.
Kruse
,
S.
Wirkert
,
D.
Rahner
,
E. W.
Pellegrini
,
R. S.
Klessen
,
L.
Maier-Hein
,
C.
Rother
and
U.
Köthe
,
ArXiv
1808.04730v3 (
2018
).
17.
A.
Kirsch
,
An Introduction to the Mathematical Theory of Inverse Problems
. (
Springer Nature
,
Berlin
,
2021
).
18.
D.
Gamerman
and
H. F.
Lopes
,
Markov Chain Monte Carlo: Stochastic simulation for Bayesian inference.
(
Chapman and Hall/CRC
,
Florida, USA
,
2006
).
19.
M.
Sunnåker
,
A. G.
Busetto
,
E.
Numminen
,
J.
Corander
,
M.
Foll
and
C.
Dessimoz
,
PLoS Comput Biol
9
1
(
2013
).
20.
R. D.
Wilkinson
,
Statistical Applications in Genetics and Molecular Biology
12
(
2
),
129
141
(
2013
).
21.
J.
Lintusaari
,
M. U.
Gutmann
,
R.
Dutta
,
S.
Kaski
and
J.
Corander
,
Systematic Biology
66
(
1
),
1063
5157
(
2017
).
22.
M.
Mirza
and
S.
Osindero
,
ArXiv
1411.1784v1 (
2014
).
23.
K.
Sohn
,
H.
Lee
and
X.
Yan
,
Advances in Neural Information Processing Systems
28
(
2015
).
24.
E. G.
Tabak
and
E.
Vanden-Eijnden
,
Communications in Mathematical Sciences
8
(
1
),
217
233
(
2010
).
25.
E. G.
Tabak
and
C. V.
Turner
,
Pure and Applied Mathematics
66
(
2
),
145
164
(
2012
).
26.
L.
Dinh
,
D.
Krueger
and
Y.
Bengio
,
presented at the International Conference on Learning Representations (ICLR
),
San Diego, USA
,
2015
(unpublished).
27.
L.
Dinh
,
J.
Sohl-Dickstein
and
S.
Bengio
,
presented at the ICLR
,
Toulon, Frankreich
,
2017
(unpublished).
28.
D. P.
Kingma
and
P.
Dhariwal
,
presented at the 32nd Conference on Neural Information Processing Systems (NeurIPS 2018
),
Montréal, Kanada
,
2018
(unpublished).
29.
L.
Ardizzone
,
T.
Bungert
,
J.
Kruse
,
P.
Steinbach
and
A.
René
30.
L.
Ardizzone
,
C.
Lüth
,
J.
Kruse
,
C.
Rother
and
U.
Köthe
,
ArXiv
1907.02392v3 (
2019
).
31.
I.
Tolstikhin
,
O.
Bousquet
,
S.
Gelly
and
B.
Schölkopf
,
ArXiv
arXiv:1711.01558 (
2017
).
32.
J.
Kruse
, in
Visual Learning Lab
(
Heidelberg
, 23.02.
2021
).
33.
V. F.
Ksoll
,
L.
Ardizzone
,
R.
Klessen
,
U.
Koethe
,
E.
Sabbi
,
M.
Robberto
,
D.
Gouliermis
,
C.
Rother
,
P.
Zeidler
and
M.
Gennaro
,
Monthly Notices of the Royal Astronomical Society
499
(
4
),
5447
5485
(
2020
).
34.
J.
Bergstra
,
R.
Bardenet
,
Y.
Bengio
and
B.
Kégl
,
presented at the Proceedings of the 24th International Conference on Neural Information Processing Systems
,
Granada, Spanien
,
2011
(unpublished).
35.
T.
Akiba
,
S.
Sano
,
T.
Yanase
,
T.
Ohta
and
M.
Koyama
,
presented at the Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining
,
New York, USA
,
2019
(unpublished).
36.
P.
Isola
,
J. Y.
Zhu
,
T.
Zhou
and
A. A.
Efros
,
ArXiv
1611.07004v3 (
2018
).
37.
F.
Johannaber
and
W.
Michaeli
,
Handbuch Spritzgießen, 2. Auflage
. (
Carl Hanser Verlag
,
München, Wien
,
2004
).
38.
C.
Hopmann
,
G.
Menges
,
W.
Michaeli
and
P.
Mohren
,
Spritzgießwerkzeuge. Auslegung, Bau, Anwendung. 7 Auflage.
(
Carl Hanser Verlag
,
München, Wien
,
2018
).
39.
C.
Hopmann
,
S.
Jeschke
,
T.
Meisen
,
T.
Thiele
,
H.
Tercan
,
M.
Liebenberg
,
J.
Heinisch
and
M.
Theunissen
,
presented at the the 33rd International Conference of the Polymer Processing Society Cancun
,
Mexico
,
2019
(unpublished).
40.
J.
Heinisch
,
Y.
Lockner
and
C.
Hopmann
,
Journal of Manufacturing Processes
61
,
347
368
(
2021
).
41.
Y.
Lockner
and
C.
Hopmann
,
The International Journal of Advanced Manufacturing Technology
112
,
3501
ȓ
3513
(
2021
).
42.
N.
Sieber
and
H. J.
Sebastian
,
Spezielle Funktionen
. (
Springer Verlag
,
Berlin, Heidelberg
,
1977
).
This content is only available via PDF.
You do not currently have access to this content.