Several methods exist to establish relations between variables, the most common of which is the construction of models. These models enable the expression of physical relationships or laws between variables. Empirical experimentation, measurement and expert knowledge are alternate means to obtain data and establish said rules. Likewise, many methods have been developed for data representation. Underlying all of these methods are the same two pursuits for the best representation of the data, and the minimum time for execution of the method. The first pursuit seeks the achievement of two goals: a good representation of the data (i.e. minimum error) and a simple formulation (i.e. physically realistic). The second pursuit, a minimum time for execution, is desirable mainly for control processes. There is often however, an inverse relationship between minimum error and time of execution. This paper presents an alternative mathematical tool to yield a data model. The model utilizes fuzzy techniques to produce a data model for short-term load forecasting studies. The engineer responsible for this activity can also introduce his/her experience before the final forecasting. A user friendly interface has been developed to provide a man-machine interface. To build this interface, many suggestions from the operators have been incorporated. This paper also details this interface with all characteristics and features. Currently, this software runs at CEMIG Brazilian State Power Industry. The first impressions of the user are also presented in this paper.
[1]
M.s. Abou-Hussien,et al.
An Accurate Model for Short-Term Load Forecasting
,
1981,
IEEE Transactions on Power Apparatus and Systems.
[2]
A. P. Alves da Silva,et al.
A new constructive ANN and its application to electric load representation
,
1997
.
[3]
G. Gross,et al.
Short-term load forecasting
,
1987,
Proceedings of the IEEE.
[4]
A. P. Alves da Silva,et al.
Neural networks for fault location in substations
,
1996
.
[5]
Kishan G. Mehrotra,et al.
Forecasting the behavior of multivariate time series using neural networks
,
1992,
Neural Networks.
[6]
B. Valiquette,et al.
A fuzzy knowledge-based system for bus load forecasting
,
1992,
[1992 Proceedings] IEEE International Conference on Fuzzy Systems.
[7]
Saifur Rahman.
Formulation and analysis of a rule-based short-term load forecasting algorithm
,
1990
.