Fault Identification in Doubly fed Induction Generator using FFT and Neural Networks
- PDF / 1,554,512 Bytes
- 10 Pages / 595.276 x 790.866 pts Page_size
- 44 Downloads / 175 Views
Fault Identification in Doubly fed Induction Generator using FFT and Neural Networks Marcelo Patrício de Santana1 · José Roberto Boffino de Almeida Monteiro1 · Fabbio Anderson Silva Borges1 · Geyverson Teixeira de Paula1 · Thales Eugênio Portes de Almeida1 · William César de Andrade Pereira1 · Allan Gregori de Castro1
Received: 17 September 2015 / Revised: 1 October 2016 / Accepted: 20 November 2016 / Published online: 7 February 2017 © Brazilian Society for Automatics–SBA 2017
Abstract This paper presents a fault identification system for doubly fed induction generator. The proposed system is designed to identify single-phase faults and load switching events on an isolated load. Firstly, the system preprocess the stator line current data by the fast Fourier transform (FFT). In order to reduce the dimensionality of the FFT output data, the principal component analysis method is used. The fault identification stage is based on artificial neural network (ANN). Also, a post-processing is employed in order to increase the network reliability, which reduces the error of ANN. The proposed system is simulated and experimentally validated on different voltage, speed and load conditions. Keywords Neural network · Fault identification · Fast Fourier transform
1 Introduction Modern high-power wind turbines operate under variable mechanical speed. Compared to constant speed generation, this topology offers flicker reduction and four-quadrant active and reactive power capabilities (Alaya et al. 2011). The doubly fed induction generator (DFIG) is widely used in variable speed generation. The main advantage is that the power converter in this system is connected to rotor winding and has
B
Marcelo Patrício de Santana [email protected] José Roberto Boffino de Almeida Monteiro [email protected]
1
Department of Electrical Engineering, University of São Paulo - USP/EESC/SEL, Av. Trabalhador São-carlense, 400, Arnold Schimidt, São Carlos, SP CEP 13566-590, Brazil
123
about 25–30% of machine rated power. This fact leads to a overall cost reduction when compared to systems where converters must have the same power as the generator (Boldea 2006). Fault situations on electrical system, like short circuits, can damage the generator itself and the electrical equipment as well. Therefore, in these situations, a fault identification system plays an important role since protection actions can be promptly taken, increasing the generator lifetime and reducing preventive maintenance frequency. Furthermore, the standards for electric power system in numerous countries around the world require that the DFIG still connected to the grid even during fault conditions. Therefore, in order to comply with these standards the fault identification process must be considered to address the properly action according to the identified fault (Justo et al. 2015). Some papers have been studying fault identification in doubly fed induction generator recently (Justo et al. 2015; Stojcic et al. 2013; Attoui and Omeiri 2014; El-Naggar and Erlich 2016). The
Data Loading...
