Name: ALINNE PEREIRA DE OLIVEIRA

Publication date: 20/04/2017
Advisor:

Name	Role
HELDER ROBERTO DE OLIVEIRA ROCHA	Advisor *
JAIR ADRIANO LIMA SILVA	Co-advisor *

Examining board:

Name	Role
HELDER ROBERTO DE OLIVEIRA ROCHA	Advisor *
JAIR ADRIANO LIMA SILVA	Co advisor *
JOSE LEANDRO FÉLIX SALLES	Internal Examiner *
REGINALDO BARBOSA NUNES	External Examiner *

Summary: Power line communications (PLC) appears as a good alternative for data communication due to the already installed and pervasive power distribution infrastructure. It is based on this features that, a methodology for resilient allocation of PLC equipments for data communication in medium voltage (MV) networks is proposed in this work. However, because electrical networks were not customized for data transmission purposes, destructive interference exists in the medium to degrades high data rate communications. In the frequencies above 1 MHz, the PLC medium is characterized as a multipath channel caused
by the derivations and impedance mismatches that exists along the main transmission segment. Normally called interfering branches, the ramification lengths around 10 % of the main segment have a destructive influence on the PLC signal. Therefore, a methodology to strategically allocate the lowest quantity of PLC equipments in the MV considering the presence or absence of interfering branches is proposed. Moreover, due to changes in the topology of the electrical MV access network during its operation and/or mantainence, the proposed method should allocate repeaters to fit all the topological
scenarios in order to prevent interruptions in the data communications. Another problem contemplated is the loss of any repeater without the powerline communications being impaired or even interrupted.
The methodology proposed in this work uses metaheuristics based on a Genetic Algorithm and on a new Binary Particle Swarm optimization (NBPSO) to generate good solutions to the particular allocation problem addressed in this work. The same amount of PLC equipments installed in a real network was achieved after the application of the proposed methodology. Using the proposed Genetic Algorithm, the method also successfully designated 14 PLC repeaters for the MV located in Rio Preto do Oeste in S˜ao Mateus-ES, which should be addressed with technology in the near future.
Keywords: Power Line Communication (PLC), PLC equipment allocation, Genetic Algorithm, New Binary Particle Swarm Optimization, Resilience.

Access to document