Identification of variable displacement axial-piston pump with proportional valve control

The classical approach for regulation the displacement volume of axial-piston pumps is installation of hydraulic controller such as DR, DFR etc. which implement a control law of the pressure and flow rate. The electronic controllers for the axial-piston pumps commonly rely on the swash plate swivel...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Mitov, Alexander, Kralev, Jordan, Slavov, Tsonyo
Format: Tagungsbericht
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The classical approach for regulation the displacement volume of axial-piston pumps is installation of hydraulic controller such as DR, DFR etc. which implement a control law of the pressure and flow rate. The electronic controllers for the axial-piston pumps commonly rely on the swash plate swivel angle feedback. A specific aspect of our custom developed system is that it does not use a sensor for the swivel angle, but relies on the measurement of the flow rate. The designed test bench include A10VSO pump equipped with a proportional valve VT-DFP and electronic amplifier VT-5041. The software is implemented as a hardware in the loop solution, where the controller is running as a blocking Simulink® model in the host workstation, and the drivers and communication services are embedded in the PLC. The real-time synchronization between both systems is guaranteed by a custom protocol implementation. The main goal of the present study is to obtain a mathematical model of the pump dynamics using the black box system identification methods. The identification technique is applied in order to obtain a state-space parametric model. The model parameters are estimated by minimization of prediction error. The state-space model order is selected by comparing the Hankel singular values of the different order models. The 6th order model is reflecting to best balance between FIT to experimental data and model complexity.
ISSN:0094-243X
1551-7616
DOI:10.1063/5.0198991