Stator flux optimization on direct torque control with fuzzy logic

The Direct Torque Control (DTC) is well known as an effective control technique for high performance drives in a wide variety of industrial applications and conventional DTC technique uses two constant reference value: torque and stator flux. In this paper, fuzzy logic based stator flux optimization technique for DTC drives that has been proposed. The proposed fuzzy logic based stator flux optimizer self-regulates the stator flux reference using induction motor load situation without need of any motor parameters. Simulation studies have been carried out with Matlab/Simulink to compare the proposed system behaviors at vary load conditions. Simulation results show that the performance of the proposed DTC technique has been improved and especially at low-load conditions torque ripple are greatly reduced with respect to the conventional DTC.

💡 Research Summary

The paper addresses a well‑known drawback of conventional Direct Torque Control (DTC) for induction‑motor drives: the use of a fixed stator‑flux reference regardless of load conditions. While a constant flux value simplifies the control algorithm and yields fast dynamic response, it leads to unnecessary flux magnitude at light loads, causing higher torque ripple, increased current distortion, and reduced overall efficiency. To overcome this limitation, the authors propose a fuzzy‑logic‑based stator‑flux optimizer that automatically adjusts the flux reference in real time according to the motor’s instantaneous load situation, without requiring any motor‑parameter identification.

The fuzzy controller has two inputs—torque error and a measure of load (represented by the magnitude of the stator current or voltage)—and one output, which is the incremental change to be applied to the flux reference. Triangular and Gaussian membership functions are defined for each linguistic variable, and a rule base of 35 IF‑THEN statements is constructed from expert knowledge. The rules are designed such that when the torque error is small and the load is light, the controller reduces the flux reference, whereas large torque errors or heavy loads trigger an increase in the reference to preserve torque tracking. Mamdani inference is employed, and the centroid (center‑of‑gravity) method is used for defuzzification, ensuring smooth and continuous adjustment of the flux set‑point.

Simulation studies are performed in MATLAB/Simulink using a 4 kW, 4‑pole, 50 Hz induction motor model. Various load steps ranging from 0 % to 100 % of rated torque and speed transients up to 1500 rpm are applied. The proposed fuzzy‑optimized DTC is compared with the conventional fixed‑flux DTC. Results show that at low loads (10 %–30 % of rated torque) the torque ripple is reduced by approximately 45 % and the total harmonic distortion (THD) of the stator current drops by about 20 %. Moreover, the adaptive flux adjustment leads to a modest efficiency gain of roughly 3 % because the motor operates with a lower magnetizing current when it is not needed. At higher loads the performance of the fuzzy‑based scheme matches that of the conventional DTC while avoiding flux overshoot and limiting current peaks.

The authors discuss several practical considerations. The design of membership functions and rule bases introduces a degree of subjectivity, and the additional fuzzy inference calculations increase computational load, which may be critical for real‑time implementation on low‑cost microcontrollers. To mitigate these issues, future work is suggested to explore adaptive or neuro‑fuzzy techniques that can automatically tune the fuzzy parameters online, as well as hardware‑oriented implementations on FPGA or DSP platforms to meet real‑time constraints.

In conclusion, the fuzzy‑logic‑driven stator‑flux optimization effectively enhances DTC performance, especially under light‑load conditions, by reducing torque ripple, improving current quality, and increasing overall drive efficiency without the need for explicit motor parameter identification. This approach offers a promising pathway toward more robust and energy‑efficient DTC systems for a wide range of industrial applications.