A novel wavelet transform aided neural network based transmission line fault analysis method

Unsymmetrical fault diagnosis in transmission/distribution networks

“…Knowledge based fault detection [3][4][5][6][7][8][9][10][11][12][13][14][15][16] utilises prior knowledge of the system quantities (voltages and currents and waveforms) under different fault and system operating conditions. This knowledge is then used to train a learning system to identify abnormal conditions and classify them.…”

“…This knowledge is then used to train a learning system to identify abnormal conditions and classify them. Methods such as Artificial Neural Network [8,9,[11][12][13][14][15][16], Fuzzy theory [6], Fuzzy Neural Network [5], and Adaptive neuro-fuzzy inference systems [9], have widely been used for fault analysis in power systems. The above referenced approaches however suffer from the following limitations: (i) the dynamics of the excitation and speed governing systems are largely ignored.…”

Protection of Thyristor Controlled Series Compensated Transmission Lines using Support Vector Machine

“…BPNN (back propagation Neural Network), RBFNN [4] (radial basis function neural network), FNN (Fuzzy Neural network) are emp loyed for adaptive protection of such a line where the protection philosophy is viewed as a pattern classification problem [5]. Furthermore, co mb ined techniques have already been used, such as wavelet transform and fu zzy logic [6,7]; neural network and fuzzy logic [8,9]; neural network and wavelet transform [10].and neural network and total least square estimat ion of signal parameters via the rotational invariance [11]. The networks generate the trip or block signals using a data window of voltages and currents at the relaying point.…”

Wavelet entropy based algorithm for fault detection and classification in FACTS compensated transmission line

“…Different types of neural networks (NN) based pattern recognition procedures [7][8][9] were proposed which large training need set generation, large training time and design of a new neural network for each transmission line. Different attempts have been made for fault location and classification using numerical methods, wavelet transform, S-transform, TT-transform, fuzzy logic systems and support vector machines [3][4][5][6][7][8][9][10][11][12][13][14][15]. Most of these attempts were trying to classify the fault and identify the faulted section in a transmission line compensated either by series capacitor protected by metal-oxide varistor (MOV) or compensated by thyristor-controlled series compensators (TCSCs) protected by MOV or compensated by both.…”