Please use this identifier to cite or link to this item:
http://acervodigital.unesp.br/handle/11449/76325
- Title:
- Multilayer perceptron neural networks training through charged system search and its Application for non-technical losses detection
- Universidade Estadual Paulista (UNESP)
- Polytechnic Institute of Porto-IPP
- Universidade de São Paulo (USP)
- Research Executive Agency
- REA: 318912
- The non-technical loss is not a problem with trivial solution or regional character and its minimization represents the guarantee of investments in product quality and maintenance of power systems, introduced by a competitive environment after the period of privatization in the national scene. In this paper, we show how to improve the training phase of a neural network-based classifier using a recently proposed meta-heuristic technique called Charged System Search, which is based on the interactions between electrically charged particles. The experiments were carried out in the context of non-technical loss in power distribution systems in a dataset obtained from a Brazilian electrical power company, and have demonstrated the robustness of the proposed technique against with several others nature-inspired optimization techniques for training neural networks. Thus, it is possible to improve some applications on Smart Grids. © 2013 IEEE.
- 26-Aug-2013
- 2013 IEEE PES Conference on Innovative Smart Grid Technologies, ISGT LA 2013.
- Charged System Search
- Neural Networks
- Nontechnical Losses
- Charged system searches
- Competitive environment
- Meta-heuristic techniques
- Multi-layer perceptron neural networks
- Non-technical loss
- Optimization techniques
- Power distribution system
- Trivial solutions
- Electric load distribution
- Electric utilities
- Privatization
- Smart power grids
- Neural networks
- http://dx.doi.org/10.1109/ISGT-LA.2013.6554383
- Acesso restrito
- outro
- http://repositorio.unesp.br/handle/11449/76325
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.