Optimal centralized scheduling strategy for bidirectional charging of PEV fleets in low-voltage microgrids

Abstract

Rapid growth of plug-in electric vehicles (PEVs) is reshaping demand in low-voltage microgrids where voltage stability and power-quality margins are tight. Uncoordinated charging deepens evening peaks, stresses feeder limits, and constrains renewable hosting. This paper proposes a centralized, optimization-based scheduling strategy for bidirectional charging coordinating grid-to-vehicle (G2V) and vehicle-to-grid (V2G) dispatch to jointly minimize energy cost and enhance voltage stability. A linear programming (LP) model optimizes charging/discharging over discrete intervals subject to realistic constraints: charger power limits, state-of-charge (SoC) bounds, nodal-voltage regulation, and line-flow limits. The optimization is embedded in a forward-backward sweep load-flow loop to respect feeder physics. Using the IEEE European LV 8-bus system, we evaluate five scenarios single tariff, time-of-use (ToU) tariff, holiday load growth, ToU under holiday load, and photovoltaic (PV) integration. Relative to an uncontrolled baseline, the centralized strategy shifts demand off-peak, reduces peaks by up to 40% (12.0 to 7.2 kW), lowers energy cost by up to 25% (₹192.0 to ₹144.0), and improves minimum node voltages to 400–407 V; with PV, energy cost reaches ₹96.0 and minimum voltage rises to 412 V, all within EN 50,160 (±10%) bounds. These results validate a practical, scalable demand-side management (DSM) approach that improves reliability, reduces operating cost, and facilitates renewable integration; extensions to real-time, data-driven, or decentralized variants for larger fleets are outlined.