Comprehensive study & implementation of energy prognosis using neural net approach

According to current prevailing views, the field of energy will undergo significant structural changes in coming decades, making it radically different from what we know today. The classical approach to long term forecasting is often limited to the use of load and weather information occurring with monthly or annual frequency. So, in this work, the main objective of this work is prediction and forecast of historical energy data using ANN technique and curve fitting. It proposes a modern approach that takes advantage of daily information to create more accurate and defensible forecasts. The main scenarios are predictive modelling & scenario analysis. It uses ANN technique for predicting and optimizing the data. The other objective of this work is to minimize the error value up to 10^-5. It also uses time series analysis methods curve fitting and surface fitting methods. This research investigates the forecasting of residential energy consumption by applying the structural time series model to yearly data. It also provides power estimation at a particular time. The MATLAB software is used to set up a neural network model.