.png){kind=icon}
.png){kind=icon}
Return
Neural network-based adaptive integral sliding mode control for PMSM with command filter under uncertainties and time-varying loads
.png){kind=icon}
PRE
AI
Z
M
L
H
DOI：10.1007/s11071-026-12941-7.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The position tracking control of a permanent magnet synchronous motor (PMSM) subject to uncertainties and time-varying disturbances is studied in this paper. Based on the command filter, second- and third-order sliding mode surfaces are designed to enhance the robustness of the derived control input voltages. A decoupling term with a scale transformation coefficient is embedded into the control input voltage to tackle the coupling caused by the backstepping technique, which ensures reliable hardware operation of control signals. Thereafter, a radial basis function neural network with a scale factor is constructed to approximate state-dependent uncertainty. Additionally, an adaptive law with a scale factor is also developed to estimate the time-varying disturbance. The stability of the closed-loop PMSM system is then examined using the Lyapunov function. Finally, the effectiveness and advantages of the proposed control strategy for PMSM are verified via comparative experiments based on TMS320F28379D DSP. Results demonstrate that the designed control strategy enhances both transient and steady-state position tracking performance, with the integral of time-weighted squared error even reduced by $$17.3258\%$$ .
Keywords:
Permanent magnet synchronous motor
Command filter
Scale factor
Sliding mode controller
Neural network
Adaptive control
Journal
IF:
6
Papers:
1.4W
Citations:
4.1W
Organization
Cited Papers