Important aspects of electricity system planning are overall yearly el
ectricity consumption and information about variations in consumption.
These variations will influence the capacity that has to be installed
, as well as the type and quantity of fuel that will be used. Using on
ly economic, socio-economic, technical and meteorological indicators,
we have developed a model that can predict the hourly load shape. This
model is based on a division of the total electricity-consumption int
o 60 economic sectors and electrical appliances. For each of these sec
tors, a load shape is formed. Summation over sectors then gives the to
tal hourly electricity consumption pattern. To test this model, the ho
ur-to-hour load shape of the Dutch public grid for 1980 was formulated
and compared with the actual load shape. The root-mean-square differe
nce between the estimated and actual load shapes is 5% of the mean hou
rly load. The accuracy of the estimated maximum load for every week is
of the same magnitude as that used for electricity-supply system plan
ning over the short term. We conclude that our model can be used for l
ong-term load shape forecasting.