A denoising-classification neural network for power transformer protection

Artificial intelligence(AI)can potentially improve the reliability of transformer protection by fusing multiple features.However,owing to the data scarcity of inrush current and internal fault,the existing methods face the problem of poor generalizability.In this paper,a denoising-classification neural network(DCNN)is proposed,one which inte-grates a convolutional auto-encoder(CAE)and a convolutional neural network(CNN),and is used to develop a reli-able transformer protection scheme by identifying the exciting voltage-differential current curve(VICur).In the DCNN,CAE shares its encoder part with the CNN,where the CNN combines the encoder and a classifier.Based on the inter-action of the CAE reconstruction process and the CNN classification process,the CAE regards the saturated features of the VICur as noise and removes them accurately.Consequently,it guides CNN to focus on the unsaturated features of the VICur.The unsaturated part of the VICur approximates an ellipse,and this significantly differentiates between a healthy and faulty transformer.Therefore,the unsaturated features extracted by the CNN help to decrease the data ergodicity requirement of AI and improve the generalizability.Finally,a CNN which is trained well by the DCNN is used to develop a protection scheme.PSCAD simulations and dynamic model experiments verify its superior performance.

Knowledge-based Convolutional Neural Networks for Transformer Protection

Deep learning based transformer protection has attracted increasing attention.However,its poor generalization abilities hinder the application of deep learning in the power system owing to the limited training samples.In order to improve its generalization abilities,this paper proposes a knowledge-based convolutional neural network(CNN)for the transformer protection.In general,the power experts can reliably discriminate between faulty transformers and healthy transformers only through the unsaturated parts of equivalent magnetization curve(voltage of magnetizing branch-differential current curve)but deep learning intends to focus on the combined features of saturated and unsaturated parts.Inspired by the identification process of power experts,CNN adopted a specially designed loss function in this paper which is used to identify the running states of power transformers.Specifically,the presented Restrictive Weight Sparsity substitutes a special regularization term for the common LI regularization.The presented Adaptive Sample Weight Adjustment endows the softmax loss of each sample with the optimizable weight the softmax loss of each sample with the optimizable weights to increase the impact of more-difficult-to-identify cases on the training process.With the modified loss function,the knowledge is abstractly introduced into the training process of CNN so as to successfully imitate the identification process of power experts.Accordingly,the proposed knowledge-based CNN will pay more attention to the unsaturated parts of equivalent magnetization curve even if only limited samples are included in the training process.The results of simulations and dynamic model experiments reveal that the knowledge-based CNN exhibits an improved generalization ability and the knowledge-based deep learning algorithm is a promising research direction.

Waveform feature monitoring scheme for transformer differential protection

We propose a new scheme for transformer differential protection. This scheme uses different characteristics of the differential currents waveforms (DCWs) under internal fault and magnetizing inrush current conditions. The scheme is based on choosing an appropriate feature of the waveform and monitoring it during the post-disturbance instants. For this purpose, the signal feature is quantified by a discrimination function (DF). Discrimination between internal faults and magnetizing inrush currents is carried out by tracking the signs of three decision-making functions (DMFs) computed from the DFs for three phases. We also present a new algorithm related to the general scheme. The algorithm is based on monitoring the second derivative sign of DCW. The results show that all types of internal faults, even those accompanied by the magnetizing inrush, can be correctly identified from the inrush conditions about half a cycle after the occurrence of a disturbance. Another advantage of the proposed method is that the fault detection algorithm does not depend on the selection of thresholds. Furthermore, the proposed algorithm does not require burdensome computations.

IEC61850 standard-based harmonic blocking scheme for power transformers

Transformer Differential and overcurrent schemes are traditionally used as main and backup protection respectively. The differential protection relay (SEL487E) has dedicated harmonic restraint function which blocks the relay tripping during the transformer magnetizing inrush conditions. However, the backup overcurrent relay (SEL751A) applied to the transformer protection does not have harmonic restraint element and trip the overcurrent relay during the inrush conditions. Therefore, major contribution of this research work is the developed harmonic blocking scheme for transformer which uses element (87HB) of the transformer differential relay (SEL487E) to send an IEC61850 GOOSE-based harmonic blocking signal to the backup overcurrent relay (SEL751A) to inhibit from tripping during the transformer magnetizing inrush current conditions. The simulation results proved that IEC61850 standard-based protection scheme is faster than the hardwired signals. Therefore, the speed and reliability of the transformer scheme are improved using the IEC61850 standard-based GOOSE applications.