Artificial neural network-based peak load forecasting using conjugate gradient methods

IEEE Transactions on Power Systems - Tập 17 Số 3 - Trang 907-912 - 2002
L. Mohan Saini1, M. Kumar Soni1
1Department of Electrical Engineering, National Institute of Technology, Kurukshetra, Haryana, India.

Tóm tắt

The daily electrical peak load forecasting (PLF) has been done using the feed forward neural network (FFNN)-based upon the conjugate gradient (CG) back-propagation methods, by incorporating the effect of 11 weather parameters, the previous day peak load information, and the type of day. To avoid the trapping of the network into a state of local minima, the optimization of user-defined parameters, namely, learning rate and error goal, has been performed. The training dataset has been selected using a growing window concept and is reduced as per the nature of the day and the season for which the forecast is made. For redundancy removal in the input variables, reduction of the number of input variables has been done by the principal component analysis (PCA) method of factor extraction. The resultant dataset is used for the training of a 3-layered NN. To increase the learning speed, the weights and biases are initialized according to the Nguyen and Widrow method. To avoid over fitting, an early stopping of training is done at the minimum validation error.

Từ khóa

#Artificial neural networks #Load forecasting #Gradient methods #Neural networks #Input variables #Principal component analysis #Feeds #Feedforward neural networks #Character generation #Weather forecasting

Tài liệu tham khảo

10.1093/comjnl/7.2.149 polak, 1971, Computational Methods in Optimization a Unified Approach 10.1007/BF01593790 10.1016/S0893-6080(05)80056-5 scales, 1985, Introduction to Non-Linear Optimization, 10.1007/978-1-349-17741-7 10.1147/rd.164.0431 10.1007/BF01580369 saini, 2001, Some aspects of machine learning using artificial neural network harman, 1976, Modern Factor Analysis 10.1109/IJCNN.1990.137819 10.1049/ip-gtd:19951602 10.1109/59.331456 10.1109/59.466468 dash, 1995, power-demand forecasting using a neural network with an adaptive learning algorithm, Generation Transmission and Distribution IEE Proceedings-, 142, 560, 10.1049/ip-gtd:19952245 10.1109/59.736285 10.1109/59.627852 10.1109/59.221223 charalambous, 1992, conjugate gradient algorithm for efficient training of artificial neural networks, Circuits Devices and Systems IEE Proceedings G, 139, 301, 10.1049/ip-g-2.1992.0050 10.1109/59.76685 sarle, 0, stopped training and other remedies for over fitting, Proc 1995 27th Symp Interface