Network reconfiguration results in voltage deviations because of the time-varying nature of loads. Nowadays, with the integration of renewable energy sources with the power grid, the optimal configuration should be determined taking into account the variations in loads and distributed generations. In this study, a long-term reconfiguration method named the long-term extended flow algorithm, which improves the timeliness of configuration behavior on time-varying loads, is proposed to substitute for the previous short-term ones. Moreover, optimal reconfiguration instants are achieved by a novel approach of optimal daily schedule called accumulation of unbalanced load distribution based on the trade-off between switching operations costs and reduction in voltage deviations. The proposed methods were tested by case studies of two test distribution systems under realtime measured data with the interconnection of photovoltaic generators in the MATLAB environment. By applying the proposed methods, the total operating cost of the network is reduced significantly within reasonable computation time, and its efficiency is also improved compared to fixed configuration, online reconfiguration policies, and daily schedule proposed by previous authors.