Since distribution substations connect transmission lines and distribution system, planning these substations is a very important step in scheduling process of power system and optimal placement of distribution transformers for reducing the involved loss is of concern. In this paper a new model is [proposed for solving optimal substation planning of distribution networks. In the objective function, we have included both capital and operation (loss) costs. The optimization problem is solved by genetic algorithm (GA). In order to evaluate the effects of load imbalance in low-voltage system, a group of balanced one-phase and three-phase and imbalanced three-phase in 4-wire system are used. As a case study, the optimization model is tested on the 33-bus system. Simulation results show that the total cost is reduced by balancing the load of different phases. Also, the results confirm a 48 % reduction in loss and an 11 % reduction in costs of the network planning. Moreover influence of energy cost change and load change on distribution network planning is evaluated by investigating on 33-bus system.