PROBABILISTIC NEURAL NETWORK AND ITS ADAPTIVE VERSION — A STOCHASTIC APPROACH TO PD PATTERN CLASSIFICATION TASK

2005 ◽  
Vol 02 (04) ◽  
pp. 333-344 ◽  
Author(s):  
B. KARTHIKEYAN ◽  
S. GOPAL ◽  
S. VENKATESH
Keyword(s):  
Neural Network ◽  
Pattern Classification ◽  
Probabilistic Neural Network ◽  
Stochastic Approach ◽  
Diagnostic Techniques ◽  
Single Type ◽  
Destructive Testing ◽  
Theoretical Concepts

The quality of electrical insulation of any power apparatus is an indispensable requirement for its successful and reliable operation. Partial Discharge (PD) phenomenon serves as an effective Non Destructive Testing (NDT) technique and provides an index on the quality of the insulation. The innovative trend of using Artificial Neural Network (ANN) towards the classification of PD patterns is cogent and discernible. In this paper a novel method for the classification of the PD patterns using the original Probabilistic Neural Network (PNN) as well as its variation is elucidated. A preprocessing scheme that extracts pertinent features of PD from the raw data towards achieving a compact ANN has been employed. The classification of single-type insulation defects such as voids, surface discharges and corona has been taken up. The first part of the paper gives a brief on PD, various diagnostic techniques and interpretation. The second part deals with the theoretical concepts of PNN and its adaptive version. The last part provides details on various results and comparisons of the PNN and its adaptive version in PD pattern classification.

Application of Probabilistic Neural Network in Pattern Classification

2013 ◽  
Vol 441 ◽  
pp. 738-741 ◽  
Author(s):  
Shuo Ding ◽  
Xiao Heng Chang ◽  
Qing Hui Wu
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Pattern Classification ◽  
Probabilistic Neural Network ◽  
Back Propagation ◽  
Back Propagation Neural Networks ◽  
Training Samples ◽  
The One ◽  
Speed Accuracy

The network model of probabilistic neural network and its method of pattern classification and discrimination are first introduced in this paper. Then probabilistic neural network and three usually used back propagation neural networks are established through MATLAB7.0. The pattern classification of dots on a two-dimensional plane is taken as an example. Probabilistic neural network and improved back propagation neural networks are used to classify these dots respectively. Their classification results are compared with each other. The simulation results show that compared with back propagation neural networks, probabilistic neural network has simpler learning rules, faster training speed and it needs fewer training samples; the pattern classification method based on probabilistic neural network is very effective, and it is superior to the one based on back propagation neural networks in classifying speed, accuracy as well as generalization ability.

scholarly journals Pattern Classification of Fabric Defects Using a Probabilistic Neural Network and Its Hardware Implementation using the Field Programmable Gate Array System

2017 ◽  
Vol 25 (0) ◽  
pp. 42-48 ◽  
Author(s):  
Abul Hasnat ◽  
Anindya Ghosh ◽  
Amina Khatun ◽  
Santanu Halder
Keyword(s):  
Neural Network ◽  
Hardware Implementation ◽  
Probabilistic Neural Network ◽  
Clock Period ◽  
Data Set ◽  
Gate Arrays ◽  
Hardware System ◽  
Field Programmable ◽  
Programmable Gate Arrays

This study proposes a fabric defect classification system using a Probabilistic Neural Network (PNN) and its hardware implementation using a Field Programmable Gate Arrays (FPGA) based system. The PNN classifier achieves an accuracy of 98 ± 2% for the test data set, whereas the FPGA based hardware system of the PNN classifier realises about 94±2% testing accuracy. The FPGA system operates as fast as 50.777 MHz, corresponding to a clock period of 19.694 ns.

Classification of Power Quality Disturbances Using Parzen Kernels

2010 ◽  
Vol 11 (1) ◽  
Author(s):  
Mihir Narayan Mohanty ◽  
Vinay Kumar ◽  
Aurobinda Routray ◽  
Prithviraj Kabisatpathy
Keyword(s):  
Power Quality ◽  
Pattern Classification ◽  
Probabilistic Neural Network ◽  
Classification Problem ◽  
Good Choice ◽  
Gaussian Kernel ◽  
Unknown Parameters ◽  
Power Quality Disturbances ◽  
Non Parametric

A novel method is proposed in this paper for the classification of power quality disturbances using a Probabilistic Neural Network with a Parzen kernel. An attempt has been made to solve the problem as a pattern classification problem in which there is a normal class and a series of abnormal classes. The traditional parametric techniques of pattern classification can't be employed due to unknown parameters of the density functions displayed by the extracted features of the signal. Hence non-parametric pattern classification method was to be adopted and Parzen kernel being used. Parzen kernel is one of the most famous non-parametric techniques and has been a good choice for this purpose with its ease of implementation and good accuracy level. The time varying nature of the probability densities are adaptively identified by Parzen windows. Experimental results have been presented for establishing the efficacy of the method as a tool to automate the Power Quality Classification problem. Various kinds of signals such as Sag, Swell, Momentary flicker, Harmonics were generated and subjected to the above classification scheme. A detailed study on the accuracy and performance of the proposed algorithm has been made with variations in parameters such as the number of training samples and the variance of the Gaussian kernel used.

The Classification of Non-Asbestos Gasket Formulation Using Clustering Method

2012 ◽  
Vol 233 ◽  
pp. 388-391
Author(s):  
Mei Hong Liu ◽  
Zhen Hua Li ◽  
Yu Xian Li ◽  
Jun Ruo Chen
Keyword(s):  
Neural Network ◽  
Raw Materials ◽  
Probabilistic Neural Network ◽  
Well Performance ◽  
Clustering Problem ◽  
Network Method ◽  
Reduce Development Time ◽  
Som Neural Network ◽  
Optimal Formula

At present, study on the non-asbestos gasket materials is the hotspot research in static sealing field. The non-asbestos sealing gaskets research and development has made great strides into the practical phase. Formula is an important factor of material, which determines performance of material. In order to obtain well performance, it is needed to optimization formula to get optimal formula that not only improve performance of non-asbestos gasket, but also reduce development time accordingly reduce cost of non-asbestos gasket. Classification of raw materials can be transformed into a mathematical clustering problem. It means that according to some algorithm, there will be some sort of input values of similar links together. Many neural networks were widely used in the classification of different materials. A method of classification by using neural network to the known 15 kinds of the non-asbestos gaskets of formula data was proposed in this paper. By using the PNN (probabilistic neural network), LVQ(Learning Vector Quantization) neural network and SOM (Self-Organizing Feature Map) neural network respectively to classify the non-asbestos gaskets to find a suitable method in the classification of non-asbestos gaskets formula. The results indicated that PNN neural network and LVQ neural network method based on the data that provided in the paper both can effectively classify, while SOM neural network can not classify them ideally, thus it provides a new theoretical basis for the classification of the non-asbestos gaskets.

Sign in / Sign up

Export Citation Format

Share Document