Intelligent torque observer combined with backstepping sliding-mode control for two-mass systems

Vo Thanh Ha (University of Transport and Communications)
Pham Thi Giang (University of Economics-Technology for Industries)

International Journal of Power Electronics and Drive Systems (IJPEDS)
Vol 13, No 4: December 2022

The construction of a backstepping-sliding mode control using a high-gain observer's neural network for torque estimation is presented in this research. The correctness of the load torque data is crucial to solving the two-mass system control issue. The article suggests a radial basis function neural network topology to handle load torque estimation. When a non-rigid drive shaft is present, the predicted value is merged with backstepping-sliding mode control to ensure speed tracking performance. The closed-stability loop is demonstrated analytically and quantitatively to prove it. Additionally, a high-gain observer-based structure is used to compare the effectiveness of the proposed control. The effectiveness of the proposed control structure is demonstrated by MATLAB simulation.