Wavelet neural network based on islanding detection via inverter-based DG

摘要: In this article, a passive neurowavelet based on islanding detection technique for grid-connected inverter-based distributed generation has been developed. Connecting generator to the distribution network many benefits such as increasing capacity of grid and enhancing power quality. However, it gives rise problems. This is mainly due fact that networks are designed without any units at level. Hence, integrating generators into existing not problem-free. Unintentional one encountered Islanding situation where system containing both loads separated from main result reasons electrical faults their subsequent switching incidents, equipment failures, or preplanned events like maintenance. The proposed method utilizes combines wavelet analysis artificial neural detect islanding. Discrete transform capable decomposing signals different frequency bands. It can be utilized in extracting discriminative features acquired voltage signals. Passive schemes have large nondetection zone NDZ concern raised active its degrading quality effect. emphasis scheme reduce close possible keep output unchanged. simulations results, performed by MATLAB/Simulink, shows small NDZ. Also, detecting accurately within minimum standard time. © 2014 Wiley Periodicals, Inc. Complexity 21: 309-324, 2015