Classification of Faults in a Distributed Generator Connected Power System Using Artificial Neural Network

Singh, Ritu; Pattanaik, Smruti Rekha; Bhuyan, Anshuman; Panigrahi, Basanta K.; Shukla, Jyoti; Shukla, Soorya Prakash

doi:10.1109/i-smac47947.2019.9032448

Cited by 2 publications

(1 citation statement)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Even though non-stationary power signals include all of the information in terms of current and voltage, intelligent approaches for interpreting potentially valuable information are highly challenging. Furthermore, the fault classification approach employs a feature extraction technique to obtain decreased dimensions and information data [7]. The fault detector and classifier modules utilize this module to extract numerous aspects that characterize the signal.…”

Section: Introductionmentioning

confidence: 99%

Fault Classification in a DG Connected Power System using Artificial Neural Network

Bhuyan

Panigrahi

Pati

2022

IJEEI

Self Cite

View full text Add to dashboard Cite

Distributed generation is playing a major role in the power system to meet the growing load demand. Integration of Distributed Generator (DG) to grid leads to various issues of protection and control of power system structure. From the different fault issues that occur in a distributed generator integrated power system, classification of fault remains as one of the most vital issues even after years of in-depth research. Because researchers are attempting to detect and diagnose these faults as soon as possible in order to avoid financial losses, this work aims to investigate the sort of fault that happened in the hybrid system. This paper proposed artificial neural network-based approaches for fault disturbances in a microgrid made up of wind turbine generators, fuel cells and diesel generator. The voltage signal is retrieved at the point of common coupling (PCC). The extracted data are used for training and testing purposes. Artificial neural network technique is utilized for the classification of faults in the simulated model. Furthermore, performance indices (PIs) such as standard deviation and skewness are calculated for reduction of data size and better accuracy. Both the fault and parameters are varied to check the usefulness of the proposed method. Finally, the results are discussed and compared with different DG penetration.

show abstract

Section: Introductionmentioning

confidence: 99%