Load forecasting of high-rise buildings based on BP neural network

Accurate prediction of power load of high-rise buildings lays an important foundation for safe and efficient management of power system and rational distribution of power resources. However, it is a difficult problem to predict the power load of super high-rise buildings by combining the external factor data of high-rise buildings. According to the structural parameters of the building and the influence of external factors on the load, a load prediction model based on multi-input and single-output BP neural network is proposed. By using the load power consumption of super high-rise buildings, fixed structural parameters such as floor area and height, as well as the uncertain factor information of the number of users, the load forecast of similar buildings is carried out. The experimental results show that the accuracy of BP neural network for building load prediction can fully meet our daily demand for building electrical load prediction, and provide effective data support for the overall load design of building, the planning of power supply and distribution lines, the selection of substation equipment and the distribution of power supply load.