Analysis and forecasting of randomly varying parameters of power system

Careful planning of the electrical power sector is of great importance since the decisions to be taken involves the commitment of large resources, with potentially serious economic risks for the electrical utility and the economy as a whole. There are different types of techniques available for analysis and prediction of randomly varying parameters. They are classified as statistical, intelligent systems, time series, fuzzy logic, neural networks. In this paper the Weibull density function, Beta Density function and arithmetic mean method has been used to estimate the load demand. The results are compared to determine the most efficient method. Another issue of great importance is that day by day fossil fuels are getting depleted. Another option for conventional sources of energy is increase in generation of renewable sources of energy. Wind generation forecasting is necessary as large intermittent generations have influence on the grid security, system operation, and market economics. Although wind energy may not be dispatched, the cost impacts of wind can be substantially reduced if the wind energy can be scheduled using accurate wind speed forecasting. In this paper Statistical Method is used for analysis of load demand of power system and Artificial Neural Network (ANN) is used for wind speed forecasting.