Research
Short Term Load Forecasting of a Region of India using Generalized Regression Neural Network
In this paper the Generalized Regression Neural Network is used for short term load forecasting (STLF) of Rajasthan region, India. It is a powerful technique to schedule plant maintenance, power system control and load flow. Rajasthan state has rich cultural and geographical diversities. It is the biggest state of India and its land area is 342,239 km². The actual data of load and temperature have been collected from Load Dispatch Center, Rajasthan and Meteorological Center Jaipur, Rajasthan, for the duration from January 2008 to December 2008. Load is forecasted with help of Artificial Neural Network and Generalized Regression Neural Network (GRNN) based models for summer, monsoon and winter seasons. Last 24 hours load, maximum and minimum temperature, season code, day type and effect of social celebrations are used as input of the networks. Results have been obtained for different patterns of load. Results show that both models have good performance and reasonable prediction accuracy. Their comparison demonstrates that GRNN model is much faster, more reliable and accurate for effective STLF of Rajasthan region, India.
