The use of distribution automation systems has grown significantly in electricity companies in recent years. The main reason for this is the need to have a smarter network, in order to reduce the time of power interruption. For such applications, communication using radio frequency is preferred because this solution is more reliable than 3G/4G and cheaper than fiber optics. In this context, it is of utmost importance to have a more efficient base transceiver station (BTS) that can cover the communication of a larger area. In this paper, the mathematical concepts of an intelligent antenna array, as well as the logical operation of an intelligent radiation system controller, are presented. Such system receives as inputs the geographic coordinates of network elements and automatically feeds an intelligent Yagi–Uda antenna array with the appropriate parameters, in order to optimize the radiation pattern into the desired directions. The presented model uses a stochastic optimization method to automatically achieve a set of optimal electrical parameters to excite the array and efficiently direct its beams in a fully controlled way. Thus, the results obtained indicate that the proposed intelligent scheme allows the energy optimization of the antenna system, reducing in 61% the number of BTS needed to cover the same area, when compared to traditional collinear antenna systems.