.png){kind=icon}
.png){kind=icon}
Return
Model-Free Neural Network Sliding Mode Control Method for Fault-Tolerant Permanent Magnet Vernier Rim-Driven Motor Drive System
.png){kind=icon}
PRE
AI
X
Z
J
Y
S
DOI：10.1007/s12555-026-00055-0.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
To address the performance degradation of fault-tolerant permanent magnet vernier rim-driven motors caused by parameter perturbations and unknown disturbances, this paper proposes a model-free neural network sliding mode control strategy. First, a mathematical model considering parameter variations and external disturbances is formulated. Based on the input-output characteristics of the vector control system, an ultra-local model of the speed loop is constructed, and a model-free neural network sliding mode controller is designed without relying on accurate motor parameters. To estimate unknown term in ultra-local model in real time, an extended nonsingular terminal sliding mode observer was developed, which can effectively suppress the chattering of the estimated disturbance compared with traditional sliding mode observers. Both simulation and experimental results demonstrate that, compared with traditional model-free sliding mode control and neural network sliding mode control, the proposed method achieves superior speed tracking, reduced torque ripple, and improved performance under parameter perturbations.
Keywords:
Neural network sliding mode control
Model-free control
Extended nonsingular terminal sliding mode observer
Journal
IF:
2.9
Papers:
170
Citations:
6.5K
Organization
Cited Papers