Position Control of PMBLDC Motor Using SVR- and ANFIS-Based Controllers

In this letter, support vector regression (SVR)- and adaptive neuro-fuzzy inference system (ANFIS)-based controllers are implemented for position control of a three-phase permanent magnet brushless DC (PMBLDC) motor. The performance of proposed control schemes is compared with the conventional PI controller for different angular positions of the rotor. The simulation results show the effectiveness of the proposed schemes in terms of rise time $$(t_\mathrm{r})$$ and steady-state error $$(e_\mathrm{ss})$$ with ANFIS showing an improvement of 99.2% and SVR showing 90.6% for steady-state error in comparison with the conventional PI approach. The improvement for rise time is 4% and 1.4% by ANFIS and SVR, respectively, in comparison with the conventional PI approach.