This paper presents a novel signal localized convolution neural network (SLCNN) for the power transformer differential protection. The distinct signal localization is performed with the convolution process sequentially on the frequency and time coefficients which are obtained from the wavelet decomposition of the differential current signal. The SLCNN is trained with a modified back-propagation algorithm according to SLCNN's architecture. Three power transformer test systems are considered for evaluation of the proposed SLCNN. The training patterns of each transformer are generated for various operating conditions. The SLCNN for each transformer is trained, validated and tested using its corresponding patterns. Then the performance of SLCNN is evaluated through confusion matrix analysis and is also compared with long short-term memory (LSTM) deep neural network, support vector machine (SVM), conventional back-propagation neural network (CBPNN) and conventional biased restraint second harmonic (CBSH) blocking method.

This paper presents the testing of power transformer differential relay in a hardware-in-the loop using a Real Time Digital Simulator (RTDS) and Digital Signal Processor (DSP) for all possible operating conditions of a power transformer. The Empirical Fourier Transform-based transformer differential relay (EFTDR) algorithm is implemented in DSP - TMS320F28335 which is interfaced with RTDS through its analog and digital hardware circuit. The modeling of power transformer and simulation are carried out using RSCAD. The real-time analog current signal corresponding to the simulated current signal is generated using RTDS and is fed to the analog channel of TMS320F28335. Similarly, the response of the proposed relay is fed back to the digital channel of RTDS. The relay response enables the trip of corresponding circuit breakers in the RSCAD simulation circuit. Here, the performance evaluation of EFTDR is carried out for various cases of internal faults, inrush current, current transformer saturation and possible combinations of simultaneous occurrences of the aforesaid events of power transformer in hardware-in-the loop testing using RTDS and DSP.

This paper presents a novel approach for the power transformer differential protection based on empirical Fourier transform (EFT). The EFT is a novel transform technique, which is derived from discrete Fourier transform (DFT) with certain modifications based on the nature of current waveforms during internal fault, inrush, and current-transformer (CT) saturation. The fundamental component estimated by EFT is equal to DFT estimation for internal fault currents and zero or a very low value for inrush currents and CT saturation currents. The fundamental component estimated by EFT is used in the biased restraint characteristic with a deviation factor (DF) to make tripping decisions for relays. The DF describes the waveform deviation rate of differential current waveform from a pure sinusoidal waveform. The proposed EFT-based differential protection algorithm (EFT-DPA) is validated and compared with the conventional DFT-based differential protection algorithm (DFT-DPA) through modeling for an existing real-time power transformer in Tamil Nadu Transmission Corp. Ltd. (TANTRANSCO), Tamilnadu, India. Also, the performance of the EFT-DPA is investigated with the fault recorder data taken from the field differential relay of the same power transformer. The power transformer modeling is carried using PSCAD and the EFT-DPA is implemented in MATLAB.

This study presents a novel differential protection algorithm (DPA) for power transformer using chirplet transform (ChT). The proposed method combines the features of biased restraint characteristic (BRC) of the conventional differential relay and out-turn of ChT in a two-stage algorithm. In the first stage, the BRC plane is divided into three zones: namely, high-set (HS), non-trip and vulnerable zones. The tripping decisions are carried out in the first two zones based on differential and biased current. However, if the operating condition of the power transformer falls in the vulnerable zone, then there is an ambiguity in discriminating internal fault, inrush current and current transformer saturation cases. Therefore, in the second stage, ChT is applied to differential current signal to obtain an energy distribution on the time-frequency plane with respect to time, frequency and chirp rate. Then, using the mean and standard deviation of the normalised energy, power transformer operating conditions are classified. Also, most of the DPAs available in the literature are system dependent. However, the proposed novel DPA can be effectively used for any system. The proposed scheme is validated for two power transformer systems using PSCAD to simulate various operating conditions and MATLAB to implement the algorithm.

This paper investigates the performance of empirical Fourier transform (EFT) based differential protection algorithm (DPA) for power transformer protection. The performance indices such as sensitivity, specificity and accuracy are evaluated for EFT-DPA using confusion matrix analysis. The performance indices of the EFT-DPA are also compared with the existing conventional Discrete Fourier transform (DFT) DPA. Though DFT-DPA is well accepted in power utilities, the limitations such as maloperation for typical cases of simultaneous occurrences of inrush current, current transformer saturation, internal/external fault and cross country fault need to be addressed. Therefore, the recently developed EFT-DPA has to be thoroughly validated. Here, transformer modeling and DPA implementation is carried using PSCAD and MATLAB code, respectively. The performance analysis is statistically assessed for an existing real time power auto transformer in Tamil Nadu Transmission Corporation Limited (TANTRANSCO), Tamilnadu, India and two other power transformer systems available in the literature.