Artificial Neural Network Motor Control for Full-Electric Injection Moulding Machine

This paper proposes a new artificial neural network-based position controller for a full-electric injection moulding machine. Such a controller improves the dynamic characteristics of the positioning for hot runners, pin valve and the injection motors for varying moulding parameters. Practical exper...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE International Conference on Industrial Technology (Online) S. 60 - 65
Hauptverfasser: Veligorskyi, Oleksandr, Chakirov, Roustiam, Khomenko, Maksym, Vagapov, Yuriy
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 01.02.2019
Schlagworte:
artificial neural networks
brushless motors
Cavity resonators
control
Fasteners
Injection molding
injection moulding
motor drive
Permanent magnet motors
Pins
Plastics
Valves
ISSN:2643-2978
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:This paper proposes a new artificial neural network-based position controller for a full-electric injection moulding machine. Such a controller improves the dynamic characteristics of the positioning for hot runners, pin valve and the injection motors for varying moulding parameters. Practical experimental data and Matlab's System Identification Toolbox have been used to identify the transfer functions of the motors. The structure of the artificial neural network, which used positioning error and speed of error, was obtained by numerical modelling in Matlab/Simulink. The artificial neural network was trained using back-propagation algorithms to provide control of the motor current thus ensuring the required position and velocity. The efficiency of the proposed ANN-based controller has been estimated and verified in Simulink using real velocity data and the position of the injection moulding machine and pin valve motors.
ISSN:2643-2978
DOI:10.1109/ICIT.2019.8755023