Load forecast is an indispensable part for distribution network planning. Accurate load forecasting can give effective guidance for grid investment, especially in the distribution network. With the increase of the factors that should be considered in the forecast process, the existing load forecast approach, such as trend extrapolation method is hard to give an effective forecasting. Therefore, an innovative AI-based load forecast method is proposed in this paper. The Artificial Neural Network (ANN) is used in this paper. Finally, the effectiveness of the proposed method is verified on the practical case. The results indicate that this method is more advanced than other approaches. Thus, it is of great importance to give an accurate load forecast when planning the distribution network.