In this paper Artificial Neural Networks (ANN) are used to predict the energy consumption in an air conditioned administrative building. The model is constructed by a simulation package called Design Builder where EnergyPlus is the core program which is provided with the weather data for Cairo and simulated to produce the system performance hourly data (Total energy consumption of the Building, Chiller energy consumption, etc). Different experiments and cases are done to evaluate the performance of the prediction at different types of inputs introduced to the ANN. Different combinations of Input parameters and different structures of ANN are used. A feed forward backpropagation structure is used for modeling which is trained by Levenberg Marquardt algorithm. It was found that changing the input data which is introduced to the ANN has a great effect on its predicting ability. Different performance parameters are defined for assessment of the ability of the trained neural network such as Root Mean Square Error and coefficient of variation where at the best case they became 4.014KW and 0.042. This will introduce the ANN capability and will show the effect of different parameters and conditionson its predictability. This model can also be potentially used for optimization of building control systems.