Iterative Learning-Based Negative Effect Compensation Control of Disturbance to Improve the Disturbance Isolation of System

Sensors (Basel). 2022 May 2;22(9):3464. doi: 10.3390/s22093464.

Abstract

At present, the cogging torque of permanent magnet synchronous motors (PMSM) seriously limits the Los pointing accuracy of aviation photoelectric stabilization platforms based on PMSM, which also restricts the requirements of ultra-long-distance and high-precision aviation reconnaissance and detection. For this problem, an off-line iterative learning control (ILC) was designed, and on this basis, a control method of negative effect compensation of disturbance (NECOD) is proposed. Firstly, the "dominant disturbance torque" in the system, that is, the cogging torque with the characteristics of position periodicity, was suppressed by off-line ILC according to different positions. Then, for the "residual disturbance" after compensation, NECOD was used to suppress it. In the constant speed scanning experiment of the aviation photoelectric stabilization platform, the method of combining the off-line iterative learning controller and the negative effect compensation of disturbance (NECOD + ILC) proposed in this paper significantly improved the Los control accuracy of the platform when compared with the classical active disturbance rejection control (ADRC) and ADRC + ILC methods, and the Los pointing error of the constant speed scanning process had only increased by less than 5% when the system had ±15% parameter perturbation. In addition, NECOD + ILC has fewer parameters and is easy to adjust, which is conducive to engineering application and promotion.

Keywords: negative effect compensation of disturbance; off-line iterative learning control; stability control.

MeSH terms

  • Algorithms*
  • Aviation*
  • Magnets
  • Torque

Grants and funding

This research received no external funding.