The conventional fuzzy modelling of short-term load forecasting has a drawb
ack in that the fuzzy rules or the fuzzy membership functions are determine
d by trial and error. An automatic model identification procedure is propos
ed to construct the fuzzy model for short-term load forecast. An analysis o
f variance is used to identify the influential variables of the system load
. To set up the fuzzy rules, a cluster estimation method is adopted to dete
rmine the number of rules and the membership functions of variables involve
d in the premises of the rules. A recursive least squares method is then us
ed to determine the coefficients in the concluding parts of the rules. None
of these steps involves nonlinear optimisation and all steps have well bou
nded computation time. This method was tested on the Taiwan Power Company's
(Taipower) load data and the performance of the proposed method is compare
d to those of Box-Jenkins (B-J) transfer function and artificial neural net
work (ANN) models.