Analysis of IEC 61850-9-2LE Measured Values Using a Neural Network

Process bus communication has an important role to digitalize substations. The IEC 61850-9-2 standard specifies the requirements to transmit digital data over Ethernet networks. The paper analyses the impact of IEC 61850-9-2LE on physical protections with (analog-digital) input data of voltage and current. With the increased interaction between physical devices and communication components, the test proposes a communication analysis for a substation with the conventional method (analog input) and digital method based on the IEC 61850 standard. The use of IEC 61850 as the basis for smart grids includes the use of merging units (MUs) and deployment of relays based on microprocessors. The paper analyses the merging unit's functions for relays using IEC 61850-9-2LE. The proposed method defines the sampled measured values source and analysis of the traffic. By using neural net pattern recognition that solves the pattern recognition problem, a relation between the inputs (number of samples/ms—interval time between the packets) and the source of the data is found. The benefit of this approach is to reduce the time to test the merging unit by getting the feedback from the merging unit and using the neural network to get the data structure of the publisher IED. Tests examine the GOOSE message and performance using the IEC standard based on a network traffic perspective.

Download Full-text

Interactive neural-network simulation: A textbook pattern-recognition problem yields to art-enhanced counterpropagation

Computer Applications in Engineering Education ◽

10.1002/cae.6180030203 ◽

1995 ◽

Vol 3 (2) ◽

pp. 79-88

Author(s):

Granino A. Korn

Keyword(s):

Neural Network ◽

Pattern Recognition ◽

Network Simulation ◽

Recognition Problem ◽

Pattern Recognition Problem ◽

Neural Network Simulation

Download Full-text

Automatic Operation of a Mobile Robot Using an RCE Network

Journal of Robotics and Mechatronics ◽

10.20965/jrm.1990.p0303 ◽

1990 ◽

Vol 2 (4) ◽

pp. 303-307

Author(s):

Hisato Kobayashi ◽

◽

Katsuhiko Inagaki

Keyword(s):

Neural Network ◽

Pattern Recognition ◽

Mobile Robot ◽

Initial Point ◽

Coulomb Energy ◽

Neural Net ◽

Automatic Operation ◽

Characteristic Equations ◽

Arbitrary Initial Point ◽

Energy Network

This article describes an example using a neural net as a method of mobile robot operation. The method eliminates the need for characteristic equations of a mobile robot, but requires an exercise to some extent using adequate ""patterns."" For accumulating experience of this practice, an RCE (restricted coulomb energy) network, or a pattern recognition-use neural network, is used. A simulation is conducted by driving a car into a garage. A man drives a car into a garage to create pattern data, which an RCE net is made to learn. After learning to some extent, it is allowed to put a car into a garage from an arbitrary initial point. The following are the descriptions of the results.

Download Full-text

A color pattern recognition problem based on the multiple classes random neural network model

Neurocomputing ◽

10.1016/j.neucom.2004.03.005 ◽

2004 ◽

Vol 61 ◽

pp. 71-83 ◽

Cited By ~ 5

Author(s):

Jose Aguilar

Keyword(s):

Neural Network ◽

Pattern Recognition ◽

Network Model ◽

Neural Network Model ◽

Color Pattern ◽

Recognition Problem ◽

Pattern Recognition Problem ◽

Random Neural Network

Download Full-text

FORMING EVOLUTIONARY DESIGN OF NEURAL NETWORKS WITH DIFFERENT NODES

International Journal of Computing ◽

10.47839/ijc.8.1.652 ◽

2014 ◽

pp. 16-23

Author(s):

Eva Volna

Keyword(s):

Neural Network ◽

Neural Networks ◽

Pattern Recognition ◽

Artificial Neural Network ◽

Feedforward Neural Network ◽

Evolutionary Design ◽

Pattern Recognition Problem ◽

Multilayer Feedforward Neural Network ◽

Artificial Neural ◽

Fully Connected

Evolution in artificial neural networks (e.g. neuroevolution) searches through the space of behaviours for a network that performs well at a given task. Here is presented a neuroevolution system evolving populations of neurons that are combined to form the fully connected multilayer feedforward neural network with fixed architecture. In this article, the transfer function has been shown to be an important part of architecture of the artificial neural network and have significant impact on an artificial neural network’s performance. In order to test the efficiency of described method, we applied it to the pattern recognition problem and to the alphabet coding problem.

Download Full-text

An Attention-Based Spiking Neural Network for Unsupervised Spike-Sorting

International Journal of Neural Systems ◽

10.1142/s0129065718500594 ◽

2019 ◽

Vol 29 (08) ◽

pp. 1850059 ◽

Cited By ~ 9

Author(s):

Marie Bernert ◽

Blaise Yvert

Keyword(s):

Neural Network ◽

Pattern Recognition ◽

Signal To Noise Ratio ◽

Spike Timing ◽

Spiking Neural Network ◽

Spike Sorting ◽

Adaptation Mechanism ◽

Pattern Recognition Problem ◽

Output Spike ◽

Few Data

Bio-inspired computing using artificial spiking neural networks promises performances outperforming currently available computational approaches. Yet, the number of applications of such networks remains limited due to the absence of generic training procedures for complex pattern recognition, which require the design of dedicated architectures for each situation. We developed a spike-timing-dependent plasticity (STDP) spiking neural network (SSN) to address spike-sorting, a central pattern recognition problem in neuroscience. This network is designed to process an extracellular neural signal in an online and unsupervised fashion. The signal stream is continuously fed to the network and processed through several layers to output spike trains matching the truth after a short learning period requiring only few data. The network features an attention mechanism to handle the scarcity of action potential occurrences in the signal, and a threshold adaptation mechanism to handle patterns with different sizes. This method outperforms two existing spike-sorting algorithms at low signal-to-noise ratio (SNR) and can be adapted to process several channels simultaneously in the case of tetrode recordings. Such attention-based STDP network applied to spike-sorting opens perspectives to embed neuromorphic processing of neural data in future brain implants.

Download Full-text

The Principle of Homology Continuity and Geometrical Covering Learning for Pattern Recognition

International Journal of Pattern Recognition and Artificial Intelligence ◽

10.1142/s0218001418500428 ◽

2018 ◽

Vol 32 (12) ◽

pp. 1850042 ◽

Cited By ~ 3

Author(s):

Xin Ning ◽

Weijun Li ◽

Jiang Xu

Keyword(s):

Neural Network ◽

Pattern Recognition ◽

Fundamental Property ◽

Experimental Results ◽

New Method ◽

Recognition Problem ◽

Pattern Learning ◽

Learning Method ◽

Pattern Recognition Problem ◽

Traditional Pattern

Homology Continuity is a fundamental property of the nature, but few of the traditional pattern recognition algorithms were aware of it. Firstly, this paper gives a brief description to the Principle of Homology Continuity (PHC), and tries to mathematically redefine it. Then, we introduce a PHC-based pattern learning method — Geometrical Covering Learning (GCL), following the Hyper sausage neural network as an instance of GCL. Lastly, we propose a GCL solution to the “two-spirals” pattern recognition problem. The final experimental results show that the new method is feasible and efficient.

Download Full-text