Data-Driven Denoising Technique for IFOC Induction Machine Drives to Improve Torque Response

High-performance induction machine drives can employ field-oriented control which relies on healthy sensor feedback. Noise in the environment causes problems in the transient response and causes torque ripple. Filtering measured quantities helps solve the ripple and transient problems but may increase memory requirements and generate filter delays. This paper proposes a data-driven delay-free denoising technique to improve torque response. Simulation and experimental results and analyses show that the proposed method reduces torque ripple amplitudes without increasing memory requirements and without adding sensor delays.