In the post availability based tariff (ABT) scenario in India, the introduction of frequency linked pricing mechanism has made the prediction of system mean block frequency a key component of the daily operation and planning activities of an electric utility. It helps the utilities and system operators to take decisions for better scheduling and operation of the power system. An artificial neural network (ANN) based model to predict short term system mean block frequency (hour ahead and day ahead) in the ABT regime is developed in this paper. The data obtained from NRLDC (Northern Regional Load Dispatch Center), BTPS (Badarpur Thermal Power Station) and IMD (India Meteorological Department) for the period from March 2005 to April 2006 have been used for training, validating and testing the ANN models. The results have been analyzed using error indices. The application of predicted day ahead system mean block frequency for optimal declaration of available capacity of gas turbine stations by a genco has been presented. Expected net savings in money terms has also been computed.
[1]
S. R. Narasimhan,et al.
Significance of Unscheduled Interchange Mechanism in the Indian Electricity Supply Industry
,
2006
.
[2]
Simon Haykin,et al.
Neural Networks: A Comprehensive Foundation
,
1998
.
[3]
Carlos E. Pedreira,et al.
Neural networks for short-term load forecasting: a review and evaluation
,
2001
.
[4]
Marcus O'Connor,et al.
Artificial neural network models for forecasting and decision making
,
1994
.
[5]
S. J. Kiartzis,et al.
A neural network short term load forecasting model for the Greek power system
,
1996
.
[6]
Steven C. Wheelwright,et al.
Forecasting methods and applications.
,
1979
.
[7]
Dominic Maratukulam,et al.
ANNSTLF-a neural-network-based electric load forecasting system
,
1997,
IEEE Trans. Neural Networks.