---

In this article, an efficient method to minimize energy losses is presented. The proposed method uses intermittent load conditions over a future time interval instead of an instantaneous network condition. This method obtains the optimal condition during the given period according to the current value of the condition. A given time interval is divided into many smaller subintervals. By increasing the number of subintervals or load profiles, the dimensions of the problem increase, for which an optimal value must be obtained. In this method, the variables are divided into the group of continuous and discrete control variables. While only continuous control variables are allowed to change in each sub-interval, continuous and discrete variables are set at the beginning of each time interval. This problem is solved by using the GBD general bend decomposition method. Using this method, the load conditions for each subinterval in the NLP subproblem are solved. Then, the results of the NLP subproblem are used in the main subproblem. As shown in the simulation results, the proposed method not only improves the voltage profile but also reduces the total energy wasted in the desired period.