Long-term energy demand predictions based on short-term measured data

In order to obtain long-term predictions based on short-term data, a neural network model was developed. The model parameters are indoor and outdoor temperature difference and energy for heating and internal use. For purposes of training the neural network model a method for extending the measured data to represent an annual variation is proposed. The method has been applied on six single-family buildings. Based on access to data from 2 to 5 weeks, the deviation between predicted and measured diurnal energy demand on an annual basis was about 4% with a correlation of 90–95%, when access to the indoor and outdoor temperature difference was assumed. For models based on access to data from the warmest periods with a very small heating demand, the deviation was about 2–4 times larger.