Model Predictive Control in Milling based on Support Vector Machines

被引：13

作者：

Ay, Muzaffer ^{[1
]}

Stemmler, Sebastian ^{[1
]}

Schwenzer, Max ^{[2
]}

Abel, Dirk ^{[1
]}

Bergs, Thomas ^{[2
]}

机构：

[1] Rhein Westfal TH Aachen, Inst Automat Control, Campus Blvd 30, D-52074 Aachen, Germany

[2] Rhein Westfal TH Aachen, Lab Machine Tools & Prod Engn, Campus Blvd 30, D-52074 Aachen, Germany

来源：

IFAC PAPERSONLINE | 2019年 / 52卷 / 13期

关键词：

Milling; Model Predictive Controller; Support Vector Machines; Model-based Control; Machine Learning; Internet of Production; TUTORIAL;

D O I：

10.1016/j.ifacol.2019.11.462

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

Today's manufacturing systems are either optimized for flexible or individualized manufacturing. The machine operator determines the optimal setup for the machine variables that are accurately implemented by the machine controllers. However, the overall objective is productivity under the restriction of product quality, where a model-based predictive controller is used to rather control the process than machine settings. This approach requires an accurate model of the dynamic behavior of the machine tool. Therefore, the Support Vector Machines algorithm is applied to identify and model the dynamic behavior under unknown nonlinearities. This model is compared to a classical modeling approach to predict the future system behavior in a model-based predictive controller. Based on the prediction, an optimization problem is solved in order to determine an optimal feed velocity. The presented approach outperforms the previous control strategy with 15 % shorter manufacturing time. Although the identified nonlinear SVM model is far more accurate for the analyzed system than previous models, further research has to be conducted regarding the application within a model predictive control strategy. (C) 2019, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.

引用

页码：1797 / 1802

页数：6

共 15 条

[1] [Anonymous], 2000, NATURE STAT LEARNING, DOI DOI 10.1007/978-1-4757-3264-1
[2] Ay M, 2018, MED C CONTR AUTOMAT, P891, DOI 10.1109/MED.2018.8442437
[3] Bemporad A., 2007, 45 IEEE C DEC CONTR, P6678, DOI DOI 10.1109/CDC.2006.377490
[4] Brecher C, 2017, INTEGRATIVE PRODUCTION TECHNOLOGY: THEORY AND APPLICATIONS, P1, DOI 10.1007/978-3-319-47452-6_1
[5] Practical selection of SVM parameters and noise estimation for SVM regression
Cherkassky, V
Ma, YQ
[J]. NEURAL NETWORKS, 2004, 17 (01) : 113 - 126
[6] A Novel Online LS-SVM Approach for Regression and Classification
Damen, Erdem
Beyhan, Selami
[J]. IFAC PAPERSONLINE, 2017, 50 (01): : 8642 - 8647
[7] Espinoza M., 2008, EUR S TIM SER PRED
[8] qpOASES: a parametric active-set algorithm for quadratic programming
Ferreau, Hans Joachim
Kirches, Christian
Potschka, Andreas
Bock, Hans Georg
Diehl, Moritz
[J]. MATHEMATICAL PROGRAMMING COMPUTATION, 2014, 6 (04) : 327 - 363
[9] Maciejowski J.M., 2000, Predictive control: with constraints, Vfirst
[10] Tutorial overview of model predictive control
Rawlings, JB
[J]. IEEE CONTROL SYSTEMS MAGAZINE, 2000, 20 (03): : 38 - 52

← 1 2 →