Implementing a weighted least squares procedure in training a neural network to solve the short-term load forecasting problem

IEEE Transactions on Power SystemsPublished 1 January 1997

M.H. Choueiki, Clark A. Mount‐Campbell, Stan Ahalt

DOI10.1109/59.627877

Source URL

Citations29

SJR quartileQ1

SJR score3.63

SNIP2.76

TL;DR

It is recommended that the weighted least squares procedure be further studied by electric utilities which use neural networks to forecast their short-term load, and experience large variabilities in their hourly marginal energy costs during a 24-hour period.

Abstract

The use of a weighted least squares procedure when training a neural network to solve the short-term load forecasting (STLF) problem is investigated. Our results indicate that a neural network that implements the weighted least squares procedure outperforms a neural network that implements the least squares procedure during the on-peak period for the two performance criteria specified; MAE% and COST, during the entire period for the COST criterion. It is therefore, recommended that the weighted least squares procedure be further studied by electric utilities which use neural networks to forecast their short-term load, and experience large variabilities in their hourly marginal energy costs during a 24-hour period.

Keywords

Computer ScienceEngineering

The MIT Press eBooksParallel Distributed Processing

3,227 Citations1987James L. McClelland, David E. Rumelhart +1 more

IEEE Transactions on Power SystemsShort term load forecasting using a multilayer neural network with an adaptive learning algorithm

268 Citations1992Ku-Long Ho, Y.-Y. Hsu +1 more

A multilayer feedforward neural network is proposed for short-term load forecasting and it is found that, once trained by the proposed learning algorithm, the neural network can yield the desired hourly load forecast efficiently and accurately.

Artificial neural network for forecasting daily loads of a Canadian electric utility

48 Citations2002B. Kermanshahi, C. Hart Poskar +5 more

The results of the proposed ANN networks have been compared to those of the present system (multiple linear regression) and show an improved forecast capability.

IEEE Transactions on Power SystemsBuilding a 'quasi optimal' neural network to solve the short-term load forecasting problem

45 Citations1997M.H. Choueiki, Clark A. Mount‐Campbell +1 more

A comparison study demonstrates the superior performance of the 'quasi optimal' neural network over an automated Box-Jenkins seasonal ARIMA model in solving the STLF problem.

Forecasting economic turning points with neural nets

38 Citations2002Richard Hoptroff, M.J. Bramson +1 more

An approach to a difficult forecasting problem: predicting the turning points of the economic cycle, using a fully connected multilayer perceptron, or feedforward network for nonlinear multivariate forecasting.

Next day's peak load forecasting using an artificial neural network

14 Citations2002Takashi Onoda

The authors combine the DSC search method (Davis, Swann, Campey search method) with the backpropagation learning algorithm (Bp) to reduce the training time and avoid converging at local minima.

Another look at forecast accuracy of neural networks

13 Citations2002M. Casey Brace, V. Bui-Nguyen +1 more

This paper compares the ability of six artificial neural networks to predict hourly system load for the Puget Sound Power and Light Company, a major North American electric utility, for the period November 1, 1991 to March 31, 1992.

Short-term system load forecasting using an artificial neural network

13 Citations2002A. Papalexopoulos, S. Hao +1 more

A new, artificial neural network (ANN) based model for the calculation of next day's load forecasts that consistently outperforms the existing model in terms of both average errors over a long period of time and number of 'large' errors.

An artificial neural network based short term load forecasting with special tuning for weekends and seasonal changes

10 Citations2002N.S. Moharari, A. Debs

The artificial neural network technique is utilized for power electric load forecasting using the backpropagation algorithm developed by the authors to follow the load pattern, weather pattern, seasonal factors and to consider special events like weekends and holidays.