Technology research on high-precision control of A-axis in efficient and powerful milling machine for blisk manufacturing

Pengbing Zhao, Yaoyao Shi, Liqun Ning

Research output: Contribution to journalArticlepeer-review

6 Scopus citations

Abstract

The control precision of the A-axis as a key functional component of a five-axis CNC machine tool directly affects the machining accuracy and surface quality of the blisks to be manufactured on it. Taking into consideration the influences of friction, backlash, parameter perturbation, measurement noise and other nonlinear disturbances on the control precision of an A-axis servo system, a robust control algorithm (LQSMC) based on the integration of linear quadratic optimal control (LQC) and sliding mode control (SMC) is proposed. Based on the system state space expression and LQC, this method improves the system state space model and defines a new sliding mode surface equation by introducing a state estimation based on the Kalman filter and the control input, and the improved control algorithm not only approaches the performance of LQC but also inhibits the chattering of SMC. Simulation analysis and experimental results show that LQSMC possesses high control precision, strong robustness, strong interference suppression ability and other advantages. It can effectively improve the positioning and tracking precision of the A-axis servo system, ensure blade machining precision and surface consistency, and significantly reduce surface roughness.

Original languageEnglish
Pages (from-to)1706-1715
Number of pages10
JournalHangkong Xuebao/Acta Aeronautica et Astronautica Sinica
Volume34
Issue number7
DOIs
StatePublished - Jul 2013

Keywords

  • A-axis
  • High-precision control
  • Kalman filter
  • Linear quadratic optimal control
  • Nonlinear disturbance
  • Sliding mode control

Fingerprint

Dive into the research topics of 'Technology research on high-precision control of A-axis in efficient and powerful milling machine for blisk manufacturing'. Together they form a unique fingerprint.

Cite this