Weather Sensitive Short Term Load Forecasting using Non-fully connected Feed Forward Neural Network

Abstract

ABSTRACT Optimal daily operation of electric power generating plants is very essential to reduce input costs and possibly the prices of electricity in general. Load forecasting is extremely important for energy suppliers, financial institutions, and other participants in electric energy generation, transmission, distribution, and markets. An accurate and reliable electric load forecasting systems are absolutely required. Precise load forecasting helps the electric utility to make unit commitment decisions, reduce spinning reserve capacity and schedule device maintenance plan properly. Since in power systems the next day’s power generation must be scheduled every day, dayahead short-term load forecasting (STLF) is a necessary daily task for power dispatch. Its accuracy affects the economic operation and reliability of the system greatly. Under prediction of STLF leads to insufficient reserve capacity preparation and, in turn, increases the operating cost by using expensive peaking units. On the other hand, over prediction of STLF leads to the unnecessarily large reserve capacity, which is also related to high operating cost. This thesis presents a solution methodology using fuzzy logic approach and artificial neural network for short term load forecasting and is implemented on historical weather sensitive data i.e. temperature, humidity, wind speed and historical load data for forecasting the load. The proposed fuzzy logic approach is implemented on weather sensitive data and the accuracy of the result is compared using two different membership functions. Artificial neural network approach is implemented on the proposed non-fully connected neural network consists of five fully connected supporting networks representing weather variables, day type and load data as inputs. Jodhpur Vidyut Nigam hourly load data used for training and testing collected from State Load Dispatch and Communication Centre, Rajasthan Vidyut Parasaran Nigam. The results are obtained from two different approaches are compared and accuracy of neural network is reported