Enhanced direct torque control based on intelligent approach for doubly-fed induction machine fed by three-level...

Area of Science:

• Electrical Engineering
• Control Systems
• Artificial Intelligence

Background:

• Doubly-fed induction machines (DFIMs) are crucial in variable-speed applications.
• Traditional direct torque control (DTC) methods often suffer from torque and flux ripple, and harmonic distortion.
• Improving the performance of DFIM drives requires advanced control strategies to mitigate these issues.

Purpose of the Study:

• To develop and evaluate an adaptive neuro-fuzzy inference system (ANFIS) based 24-sector direct torque control (DTC) for DFIMs.
• To minimize torque and flux fluctuations and stator current total harmonic distortion (THD) in DFIM drives.
• To compare the performance of the proposed ANFIS-DTC approach against conventional DTC methods.

Main Methods:

• Implemented a 24-sector DTC strategy for a DFIM using a 3-level neutral point clamped inverter.
• Replaced traditional hysteresis controllers for flux and torque with an ANFIS controller.
• Utilized MATLAB simulations to test the proposed ANFIS-DTC and conventional DTC under various operating conditions.

Main Results:

• The ANFIS-based DTC significantly reduced stator current THD by up to 81.43%.
• Torque undulation was decreased by up to 59.31% compared to conventional DTC.
• Speed overshoot was minimized by up to 98.83%, demonstrating superior dynamic performance.

Conclusions:

• The proposed 24-sector DTC-ANFIS offers superior performance in minimizing torque and flux ripple and stator current THD for DFIM drives.
• The ANFIS-based control strategy proves highly effective and efficient in enhancing DFIM drive characteristics.
• This advanced control approach presents a promising solution for electrical applications requiring high-performance motor control.