A Bi‐Level Optimization Method for Multi‐Objective Siting of Grid‐Side Energy Storage Based on an Improved Sparrow Search Algorithm

To address the nonlinear, multi‐objective, and computational complexity issues of grid‐side energy storage siting under high renewable energy penetration, this paper proposes a bi‐level optimization method using an improved multi‐objective Sparrow Search Algorithm (TAL‐MOSSA) and Mixed‐Integer Quadratic Programming (MIQP). First, a bi‐level model is constructed to minimize nodal voltage deviation and balance substation load rates. The lower‐level operation model is transformed into an MIQP problem via convex transformation and constraint linearization, ensuring fast and accurate solving. Second, for upper‐level siting, TAL‐MOSSA is proposed. It integrates Tent chaotic map initialization, nonlinear adaptive weights, and a Lévy flight perturbation strategy. Combined with an external archive and an adaptive grid mechanism to maintain the Pareto front, it overcomes the standard SSA's susceptibility to local optima and uneven solution distribution. Finally, simulations on an actual 220 kV grid demonstrate that the proposed scheme effectively improves voltage quality and load balancing. Furthermore, TAL‐MOSSA exhibits significant advantages in convergence accuracy, Pareto front uniformity, and computational efficiency.