Optimized allocation of fault passage fndicators in electrical power distribution systems using adaptive genetic algorithm

• Grant number: 15/06543-1
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: September 01, 2015
• End date: July 31, 2016
• Field of knowledge: Engineering - Electrical Engineering - Power Systems
• Principal Investigator: Fabio Bertequini Leao
• Grantee: Guilherme Hideki Ogura
• Host Institution: Faculdade de Engenharia (FEIS). Universidade Estadual Paulista (UNESP). Campus de Ilha Solteira. Ilha Solteira , SP, Brazil

Abstract

The maintenance and improvement of distribution networks reliability indices were always goals to be achieved by electric utilities due mainly to maintain power quality to consumers and pre-set by regulatory agencies limits. To achieve these goals considering the expansion and modernization of the network, utilities must invest in technology that enables monitoring and operating the network strategically. On the other hand it is necessary to reduce the maximum operating and investment costs in the network. One of the technologies available for the monitoring and operation of the network are the Fault Passage Indicators (FPI) sensors. These can be used to reduce the time for fault location and network repair time, and therefore reduce the Cost of Energy Not Supplied (CENS) improving network reliability. On the other hand, one should consider that there is an investment cost for allocation of these sensors, and it is not acceptable to install sensors in all branches of the system due to the high cost of investment. This research project will address the problem of FPI sensors allocation in distribution systems as a combinatorial optimization problem, considering two conflicting goals: CENS and cost of investment in sensor allocation. To solve the mathematical model will be developed and implemented an Adaptive Genetic Algorithm. The algorithm has recombination and mutation rates automatic and dynamically calibrated, based on the saturation of the current population, having an immediate response to possible premature convergence to local optima. Experimental tests will be carried out with the method proposed using a test system of literature.