Simplified adaptive Fractional-Order fuzzy logic control for power quality Enhancement in microgrids with high penetration of electric vehicle charging stations

With the increasing deployment of microgrids and the growing integration of electric vehicles, maintaining power quality has become a significant challenge. Distribution Static Compensators (DSTATCOMs) are widely used for fast-acting reactive power compensation and voltage support, making them well-suited to the dynamic operating conditions of microgrids and renewable-integrated systems. However, conventional controllers, such as proportional-integral (PI) and fractional-order PI (FOPI) controllers, often exhibit limited dynamic response. Similarly, traditional Fuzzy Logic Controllers (FLCs) are constrained by the size and complexity of their rule base. To overcome these limitations, this study introduces an optimized Simplified Fractional-Order Fuzzy Logic Controller (SFO-FLC), designed to reduce computational complexity by minimizing the number of fuzzy inference rules without compromising control performance. A weighted-sum approach, implemented using the Tornado optimization algorithm with Coriolis force modeling, enables the transformation of a conventional two-input FLC into an efficient single-input structure. Moreover, the proposed controller enhances robustness by incorporating the fractional-order derivative of the input signal and the fractional-order integral of the output signal into the fuzzy inference system. An adaptive version of the proposed controller, termed the Simplified Adaptive Fractional-Order FLC (SAFO-FLC), is also presented to further improve performance under varying system conditions. The simulation results demonstrated that the SFO-FLC achieved a significantly faster dynamic response with zero overshoots, requiring approximately 24 % − 85 % of the settling time of the PI, FOPI, and cascaded PI + PI controllers, and about 32 % − 85 % of the settling time of FOPI + FOPI controller. Both the classical adaptive FO-FLC and the proposed SAFO-FLC delivered comparable performance with faster recovery, requiring approximately 48 % − 88 % of the settling time of the SFO-FLC. Nevertheless, the SFO-FLC remained superior in minimizing overshoot. Notably, the proposed SAFO-FLC required only 20 % of the memory used for fuzzy rules in the classical FO-FLC, highlighting its efficiency and structural simplicity.