One of the most fundamental and significant issues for the operation and management of any electric power grid is the balance of power supply and demand. The supply-demand balance can be ensured in a number of ways, but in this study we concentrate on one particular method that makes it easier, namely the forecasting of energy consumption, which is extensively employed by utility companies or system operators. It is well known that this prediction is difficult for a variety of reasons, including inaccurate weather predictions, ambiguous customer behaviour, etc. As a result, it's possible that analytical and linear models of electricity use can't adequately address these problems. The improved radial basis function neural network (iRBF-NN), whose inputs are time sampling points, temperature, and humidity related to the consumption, is therefore proposed in this research as a machine learning-based strategy to predict electricity consumption. By resolving an optimization issue using four different cost functions and comparing their performances and computing costs, it is possible to determine the parameters of this iRBF-NN. Then, using hourly projections of temperature and humidity, the resultant model is used to forecast future electricity demand. To demonstrate the effectiveness of the suggested approach, simulated results for real data from Tokyo are shown in the final section.
Components that we will be provide are.
1.complete documentation support
2.complete working hardware/software implemented in students environment
3.classes will held accordingly.