Resumo:
Due to the constant evolution of society, with the fast technologies and processes modernization,
energy dealerships suffer constant demands to improve their services, culminating
in investments in their infrastructure, manpower and product quality. Studies are generally
carried out to allocate protective devices to reduce pre-established service quality
levels.
Short-Duration Voltage Variations (VTCDs) are events that cause problems for the various
dealerships’ consumers. As they have random occurrence, the capture of the associated
parameters becomes difficult, due to the difficulty in having measurement equipment at
different points in the network. In this scenario, methodologies for faults simulation are
important to estimate the behavior of such variables.
This master’s thesis proposes a methodology for reclosers placement in distribution systems
when considering the voltage sags’ impacts in sensitive consumer.
In order to estimate the various events, the performance of the State Enumeration Method
and Monte Carlo Simulation, the most common short circuit simulation methodologies,
were compared in a modified IEEE test system, when monitoring four buses at different
points in the network .
This work presents the results obtained from the comparison between the short circuit
simulation methodologies and their variables’ computational impacts. Monte Carlo Simulation
is an effective and practical method to estimate events.
The technique that provided the best cost-benefit was used to estimate the average voltage
sags’ values, now in an modified IEEE test system with double feeder, as input data to the
optimization routine, via Genetic Algorithm, for the recloser placement. Real coordinated
and selective adjustments were applied to verify the impact of the series installation of
automatic reclosure devices.
Also, the impacts arising from the optimal reclosers placement are presented. The adoption
of real adjustments shows the differences imposed on the behaviors in the coordinated and
selective philosophies adopted. The methodology proves to be robust to capture these
peculiarities during optimization.
