Coordinated EV Charging and Discharging Strategy Using a Mamdani MIMO Fuzzy Controller: Application to the Onitsha 11-kV Network

Abstract

Modern electricity distribution networks are facing serious operating difficulties as a result of the explosive proliferation of electric vehicles (EVs), especially in areas like Nigeria's Onitsha distribution network that have little grid flexibility. Severe voltage imbalance, frequency aberrations, and higher power losses can result from uncoordinated grid-to-vehicle (G2V) charging and vehicle-to-grid (V2G) discharging. In order to effectively coordinate EV charging and discharging while maintaining grid stability, this work creates a Multi-Input Multi-Output (MIMO) Mamdani fuzzy logic controller. The controller generates two outputs: charging rate and a quantitative grid stability index by integrating four real-time indicators: State of Charge (SOC), Time Remaining (TR), grid frequency, and grid voltage. To capture expert operating logic, eighty-one fuzzy rules were created, allowing for the simultaneous optimization of charging performance and stability support. The Onitsha 11 kV network model, comprising 29 load centers and two injection substations, was used to validate the controller’s performance across five critical operating scenarios ranging from charging-priority to grid-critical and rapid-discharging conditions. Comparative evaluation with a conventional PI controller and a DISO fuzzy controller showed that the MIMO controller consistently achieved superior grid stability indices (0.826– 0.846 in normal conditions) and provided effective protective behavior during grid-critical states by entering safe idle mode. Frequency deviation was also reduced by up to 22% compared to baseline controllers. Statistical significance tests further confirmed the superiority of the MIMO controller in specific pairwise comparisons. Overall, the proposed MIMO Mamdani controller presents a robust, adaptive, and intelligent solution for coordinated EV integration in weak distribution networks.