Академический Документы
Профессиональный Документы
Культура Документы
Abstract--This paper presents the development of an The ANNs described in this paper are multilayer
artificial neural network based short-term load forecasting perceptron structures designed to forecast the hourly load for
model. The model can forecast daily load profiles with a load 24 hours ahead. The back propagation has been for ANN
time of one day for next 24 hours. In this method can divide days
training. The purpose of this study is to forecast load
of year with using average temperature. Groups make according
linearity rate of curve. Ultimate forecast for each group obtain accurately, using actual data from the same period of previous
with considering weekday and weekend. 24 hours of a day several years are training data in order to expand learning
divided to 3 groups at 8 hours, network for every each of eight pattern. With the proposed method, the mean absolute
varieties must interpolate. This paper investigates effects of forecasting error was below 2%. Moreover, the forecasting
temperature and humidity on consuming curve. For forecasting errors of the methods in this paper have became smaller than
load curve of holidays at first calculate pick and valley and then
the errors of the method using data only previous year of the
the neural network forecast is re-shaped with the new data. The
networks are trained using hourly historical Load data and daily target year as ANN training data. This confirms the
historical max/min temperature and humidity data. The results effectiveness of the training using actual data from several
of testing the system on data from yazd utility are reported. years. This study used actual power load and weather
conditions during the one at yazd electric power co., yazd,
Index Terms--Artificial neural network, Holiday forecasting, Iran.
pick and valley load forecasting , Short-term load forecasting.
II. ANN
I. INTRODUCTION A generic feed forward neural network (FNN) is shown in
Day Actual Peak Forecasting Peak %error Hour Forecasted Load Actual Load %error
Fig. 3. Comparison of forecasted hourly loads (+...) and actual loads (---) for
15 th shaban (20 September 2005).
X. CONCLUSIONS
The system load forecasting model is a critically important
decision support tool for operating electric power systems. An
artificial neural network-based approach has been applied to
forecast the hourly electric load of the power system of yazd.
The ANN multi-layer structure has been trained using the
back propagation technique. For load forecasting must divide
days of year with using average temperature, Groups make
according linearity rate of curve. Ultimate forecast for each
group obtain with considering weekday & weekend. 24 hours
Fig. 2. Comparison of forecasted hourly loads (+...) and actual loads (---) for
1
13 th June 2005. Shaban is a lunar month and 15th shaban is the one of the religious
holiday in Iran.
of a day divided to 3 groups at 8 hours, for each groups there XII. BIOGRAPHIES
are one network. For forecasting load curve of holidays, high
and low load is first estimated then the neural network forecast
is re-shaped with the new data.
Reza Afkhami-Rohani received his BS and MS
degrees in Electrical Engineering from Sharif
XI. REFERENCES University of Technology, Tehran, Iran, in 1989 and
[1] W. R. Christiannse, "Short-term load forecasting using general 1992. He was an instructor in Sharif University in
exponential smoothing," IEEE Transactions PAS, vol. 90, no. 2, pp. 1992. Reza received his Ph.D. in the same field from
900-911, March/April, 1971. Southern Methodist University, Dallas, Texas with
[2] J. H. Park, Y. M. Park, and K. Y. Lee, "Composite Modeling for outstanding award in 1998. He has over 12 years of
Adaptive Short-Term Load Forecasting," IEEE Transactions on Power experience in the field of intelligent systems and
Systems, vol. 6, pp. 450-457, May. 1991. their applications to energy forecasting problems.
[3] S. Rahman, and R. Bhatnagar, "An Expert System Based Algorithm for Currently, he is a director as well as senior R&D
Short-Term Load Forecast," IEEE Transactions on Power Systems, vol. engineer at PRT Inc. at Dallas. His research interests are in signal processing
3, pp. 392-399, May. 1988. and neural networks and their applications in modeling and forecasting.
[4] T. S. Dillon, S. Sestito, and S. Leung, "An Adaptive Neural Network
Approach in A Power System," IEEE Proc. of 1991 ANNPS, pp. 17-21,
July. 1991.
[5] K. Y. Lee, Y. T. Cha, and C. C. Ku, "A Study on Neural Networks for Fatemah Mosalman-Yazdi was born in yazd, Iran,
Short-Term Load Forecasting," Proc. of Applications of Neural in 1980. She received the B.S. degree in Electrical
Networks to Power Systems 1991, pp. 26-30, Seattle, WA, July. 1991. Engineering from Islamic Azad University of Yazd,
[6] C. C. Ku, and K. Y. Lee, "Diagonal Recurrent Neural Network Based Iran, in 2002, the M.S. degree in Electrical
Control Using Adaptive Learning Rates," Proc. 31st IEEE Conf. on Engineering from Nagaf abad University, Esfahan,
Decision and Control Tucson, Arizona, pp. 3485-3490, Dec. 16-18, Iran, in 2005. She has been working at Islamic Azad
1992. University of Mehriz. Her main research interests are
[7] A. Khotanzad, M.H. Davis, A. Abaye, and D.J. Martukulam, "An in the area of artificial intelligent methods and their
Artificial Neural Network Hourly Temperature Forecaster with application in power systems.
Applications in Load Forecasting," IEEE Trans. PWRS, vol. 11, pp. 870-
876, May. 1996.