Downloads & Free Reading Options - Results

 In order to maintain the quality power supply and reliable operation of power delivery service to the utilities with minimum interruption, it is extremely important that the transmission line faults need to be detected and located in a reliable and accurate manner. To meet these requirement an approach based on discrete wavelet transform using neural network (NN) is proposed to estimate the fault location of the three phase transmission line and the results have been compared with the conventional method (measuring the sequence components of voltage and current). The proposed approach is based on the measurements of the three phase fault current signals at one terminal of the transmission line (at relaying point), created for different fault operating conditions using MATLAB software. To accomplish the task, features of these measured signals are extracted in terms of summation of ninth level detail coefficients using multi-resolution analysis (MRA) based on discrete wavelet transform, selecting daubechies4 (db4) as mother wavelet at 100 kHz sampling frequency. These features are then used as an input to train and test the feed forward neural network (FFNN) to provide fault location from the relaying point on the transmission line. The NN estimates the location of the fault with a significant reduction in percentage error compared to the conventional method of fault location for wide variations of operating condition. Hence, the use of multi-resolution analysis (MRA) based on discrete wavelet transform in combination with NN is quite promising for fault location estimation of the transmission line.

In order to maintain the quality power supply and reliable  operation of power delivery service to the utilities with  minimum interruption,  it is extremely important that the  transmission line faults need to be detected and located in a  reliable and accurate manner. To meet these requirement an  approach based on discrete wavelet transform using neural  network (NN) is proposed to estimate the fault location of  the three phase transmission line and the results have been  compared with the conventional  method  (measuring the  sequence components of voltage and current). The proposed  approach is based on the measurements of the three phase  fault current signals at one terminal of the transmission line  (at relaying point), created for different fault operating  conditions using MATLAB software. To accomplish the task,  features of these measured signals are extracted in terms of  summation of ninth level detail coefficients using multiresolution analysis (MRA) based on discrete wavelet  transform, selecting daubechies4 (db4) as mother wavelet at  100 kHz sampling frequency. These features are then used as  an input to train and test the feed forward neural network  (FFNN) to provide fault location from the relaying point on  the transmission line. The NN estimates the location of the  fault with a significant reduction in percentage error  compared to the conventional method of fault location for  wide variations  of operating condition. Hence, the use of  multi-resolution analysis (MRA) based on discrete wavelet  transform in combination with NN is quite promising for  fault location estimation of the transmission line.