School of Electric Power, South China University of Technology, Guangzhou 510640, China

Massive electric vehicle (EV) aggregation control technology is the key to achieving efficient participation of large-scale EVs in vehicle-to-grid (V2G). However, existing aggregation methods face two major challenges: first, the mutually exclusive characteristics of EV charging and discharging lead to non-convexity in the power feasible region; second, the power constraint boundaries of EVs with different user preferences exhibit significant differences. The former will result in overestimation of the adjustable boundaries of EVs, while the latter makes the evaluation of adjustable boundaries of EVs tend to be conservative. To address the above problems, this paper first achieves convex characterization of the non-convex feasible region by reconstructing the mathematical expression of the EV power feasible region, effectively solving the modeling problem of mutually exclusive charging-discharging constraints. Second, a constraint set contraction based aggregation algorithm is designed, which significantly improves the evaluation accuracy of adjustable boundaries under boundary differences by enhancing the inner approximation degrees of freedom of traditional aggregation algorithms. Case study analysis demonstrates that the proposed method effectively improves the computational efficiency and aggregation accuracy of massive EV scheduling.