Design and realization of an auto-tuned modified neuro-fuzzy sliding-mode-based IM drive deploying feedback linearization

This paper employs an intuitive feedback linearization (FBL) controlled induction motor (IM) drive incorporated with the modified neuro-fuzzy sliding-mode control (NFSMC) based speed and torque compensators. The proposed method substantially reduces the torque and speed chattering with the optimal drive performance under system uncertainties and disturbance. An intuitive feedback technique-based linearization of IM is designed and simulated in the stationary reference frame. The modified neuro-fuzzy control (NFC) has the benefit of enhanced computational efficiency over classical NFC as the single input introduced here as an error instead of two inputs error and change in error as in classical NFC. Therefore, it is suitable for real-time applications. The parameters of the simple modified NFC is updated by an adaptive mechanism based on sliding-mode control (SMC). The effectiveness of the proposed method using FBL-based IM drive is verified in simulation as well as experiment and is evident from the comparative results that the system performance is not declined using proposed single-input simple NFSMC as compared to the conventional two-input NFSMC, rather it shows enhanced performance over PI-controller based drive. Extensive experimental results have been provided to justify the potentiality of the proposed techniques using DSP2812.