Under the large-scale access of distributed energy resources to the distribution network, the stochasticity and volatility of energy generation have gradually become the main difficulties in the optimized and stable operation of system voltage and reactive power. Based on the structure and characteristics of active distribution network, this paper takes network loss, voltage deviation and economic index of equipment regulation as the objective function of distribution network optimization, and selects current, node voltage and energy storage as the constraints to form a multi-objective reactive power optimization model of active distribution network. At the same time, Circle chaotic initialization, sinusoidal cosine algorithm and Cauchy variational perturbation are used to improve the standard sparrow search (SSA) algorithm, and the improved sparrow search (ISSA) algorithm based on multi-strategy fusion is proposed. By applying the model and solution method proposed in this paper to the scenery scenario, it can assist in decreasing the network loss expense by up to 36.53% and voltage offset by up to 24.34%.