Authors:
A. Rizzi
;
F. Possemato
;
S. Caschera
;
M. Paschero
and
F. M. Frattale Mascioli
Affiliation:
SAPIENZA University of Rome, Italy
Keyword(s):
Distribution Feeder Reconfiguration, Power Factor Correction, Power Losses Minimization, Smart Grid, Graph Theory, Optimization, Genetic Algorithm.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Computational Intelligence
;
Evolutionary Computing
;
Evolutionary Multiobjective Optimization
;
Genetic Algorithms
;
Informatics in Control, Automation and Robotics
;
Intelligent Control Systems and Optimization
;
Knowledge Discovery and Information Retrieval
;
Knowledge-Based Systems
;
Machine Learning
;
Soft Computing
;
Symbolic Systems
Abstract:
The power loss reduction is one of the main targets for any electrical energy distribution company. In this paper the problem of the joint optimization of both topology and network parameters in a real Smart Grid is faced. A portion of the Italian electric distribution network managed by the ACEA Distribuzione S.p.A. located in Rome is considered. It includes about 1200 user loads, 70 km of Medium Voltage (MV) lines, 6 feeders, a Thyristor Voltage Regulator (TVR) and 6 distributed energy sources (5 generator sets and 1 photovoltaic
plant). The power factor correction (PFC) is performed tuning the 5 generator sets and setting the state of the breakers in order to perform the distributed feeder reconfiguration (DFR). The joint PFC and DFR problem is faced by considering a suited objective function and by adopting a genetic algorithm. In this paper we present a heuristic method to compare the graphs of two admissible topologies, such that similar graphs are characterized by close active
power loss values. This criterion is used to define a suited ordering of the list of admissible configurations, aiming to improve the continuity of the fitness function to the variation of the configurations parameter. Tests are performed by feeding the simulation environment with real data concerning dissipated and generated active and reactive power values. Preliminary results are very interesting, showing that, for the considered real network, the proposed ordering criteria for admissible network configurations can facilitate the optimization process.
(More)