The increasing penetration of uncontrollable distributed generators (NDGs) exacerbates the risk of voltage violations in active distribution networks (ADNs). Soft open point (SOP) can realize accurate power flow control and continuous voltage regulation. Centralized control strategies of SOP are difficult to meet the requirement of fast voltage and reactive power control because of the heavily computational and communicational burdens. Local voltage control based on the real-time measurements has a fast response to the frequent distributed generators (DGs) fluctuations. In this paper, an optimization method of local control strategy of SOP based on kriging metamodel is proposed. Taking minimum network losses as the objective function, the metamodel for local control of SOP is constructed based on the kriging methods. Then, the operation strategies of SOP are developed by calculating the optimal weighted vector based on real-time measurements. The proposed local voltage control strategies of SOP based on only local measurements can quickly response to the frequent DGs fluctuations and lighten the computation burdens with large networks. Case studies under different scenarios on the IEEE 33-node system are conducted to verify the effectiveness of the proposed method, and the results show that the proposed method can effectively solve the problems of voltage deviation and voltage fluctuation caused by high penetration of DGs.