This paper presents a systematic method to derive an optimal switching
plan to achieve energy loss minimization for short term and long term
operation of distribution systems. The short term optimal switching c
riterion is developed by binary integer programming with branch and bo
und technique. A overall optimization problem is formulated for the op
timal switching operation. It is further divided into several subprobl
ems to find the largest energy loss reduction among all possible switc
hing operation between two feeders. A quick method is applied to estim
ate the largest loss reduction for each feeder-pairs in the distributi
on system during short term switching operation. In this paper, the co
mposite load profile of each feeder is derived by the field test over
one year time period. According to the typical load patterns of each c
ustomer type, the hourly loading of each service section can be simula
ted more accurately. After determining the hourly optimal network conf
iguration by taking into account the typical daily load patterns for a
ll the distribution feeders, the long term optimal switching criterion
for each season is derived and the corresponding critical switches, w
hich should be operated for energy loss reduction as season changed fr
om summer to winter and vice versa, are also determined. During the co
mputer simulation, the installation and operation cost of sectionalizi
ng switches are included in the objective function as actual cost so t
hat the results obtained will be more practical. A practical Taipower
distribution system is selected for computer simulation to demonstrate
the efficiency and practicality of the proposed method to reduce the
energy loss by operating the critical switches derived.