Identification of Overcurrent Relay Faults Using Backpropagation Neural Network

Size: px
Start display at page:

Download "Identification of Overcurrent Relay Faults Using Backpropagation Neural Network"

Transcription

1 Identification of Overcurrent Relay Faults Using Backpropagation Neural Network Tey Ching Li College of Graduate Studies, Universiti Tenaga Nasional, Malaysia. Farah Hani Nordin Department of Electronics & Communication Engineering, University Tenaga Nasional, Malaysia. Tuan Ab Rashid Bin Tuan Abdullah Department of Electrical Power Engineering, University Tenaga Nasional, Malaysia. Abstract Overcurrent relays are considered as the backbone for any protection system and any faults in overcurrent relays may lead to power system breakdown. However, the actual fault of a relay is usually known after the faulty relay is sent for a Root Cause Analysis which is time consuming. Thus, the aim of this paper is to identify the overcurrent relay faults using Backpropogation Neural Network based on the general faults obtained by visual inspection without going through the Root Cause Analysis. A systematic system is created using neural network and Graphical User Interface (GUI) so that an inexperienced user will be able to predict the actual fault without sending the relay to the manufacturer. This will not only expedite repairing time, improve system availability but also will be able to reduce the operating cost. Keywords - Overcurrent Relay; Backpropagation Neural Network; Graphical User Interface (GUI) I. INTRODUCTI Power system protection plays an important role in maintaining a high degree of service reliability and security in present day. Nowadays, electricity supply is an essential service. Power outages and blackouts have become unacceptable disruptions to our daily lives and routine activities [1]. The world experienced at least thirteen major electrical power supply outage in year 1965 to 2012 [2]. Blackouts can be initiated by many causes, such as severe demand and generation imbalance, protection system failures and incorrect or slow actions of system operator. Hence, power system protection plays one of the important roles in maintaining the power system stability. Overcurrent relay is one of the protection devices in power system. Overcurrent relays are usually used for both primary and back-up protective relays, applied in every protective zone in the system. The basic principle is that when the current flowing into the overcurrent relay exceeds a predetermined amount, the relay operates with or without an intended time delay and trips the associated circuit breakers [3]. Several faults may cause overcurrent relay to malfunction. In the area of relay protection, faster response is needed while the malfunction of relay protection is not acceptable. When fault occurs at the overcurrent relay, the person in charge of the maintenance works will only be able to describe the faults that they see with their naked eyes. Only experienced workers will be able to predict the actual faults of the relay. Once a fault is observed, the faulty overcurrent relay will be sent to the manufacturer for the relay to go through a Root Cause Analysis to identify the actual fault. The process is time consuming and repairing works of the faulty overcurrent relay will only be made after the actual fault is identified. Thus, the aim of this paper is to discuss the development of a Graphical User Interface (GUI) that can be used to identify the actual fault of the overcurrent relay using Backpropagation Neural Network (BPNN). The GUI developed will be able to be used by inexperienced person to predict the actual fault of the overcurrent relay without having to wait for the results from the Root Cause Analysis. This will expedite repairing time, improve system availability and reduce operating cost. The paper starts with the introduction in Section I followed by Section II which explains the data of the overcurrent relay faults. Section III describes the training of the overcurrent relay faults using backpropagation neural network and Section IV explains the development of the GUI with neural network to identify the relay faults. Section V concludes this paper. II. OVERCURRENT RELAY FAULT DATA The overcurrent relay fault is collected and separated into two categories which are: i. the general overcurrent relay faults and ii. actual overcurrent relay faults. The general faults data is collected based on visual inspection on the overcurrent relay while actual faults are obtained after the Root Cause Analysis is conducted on the overcurrent relay. Both general and actual overcurrent relay faults are then converted into numerical data manually as shown in Appendix A. For instance the general fault in Appendix A of Display Error Code Software Failure is defined as number one while the corresponding actual fault, DSP Chip Failure, is defined as number tree. Then all general faults of Display Error Code Software Failure is defined as number one while all the actual faults of DSP Chip Failure is defined with number three. 78

2 Faults Categories The 3rd National Graduate Conference (NatGrad2015), Universiti Tenaga Nasional, Putrajaya Campus, 8-9 April In this paper, general overcurrent relay faults are presented as inputs to the neural network while the actual overcurrent relay faults are presented as output. All this data is then presented Fig. 1. Hence, neural network is used to perform the identification of the overcurrent relay faults. Before the neural network can be used to identify the overcurrent relay fault, it needs to be trained. Note that all the data in Appendix A is used to train the neural network. Once the neural network is trained only one value that represents the general fault is be entered to test and whether the neural network is able to identify the correct value which correspond the correct actual fault Overcurrent Relay Faults Data Input Relay Types of Faults Fig 1. Input Data and Output Data Output Relay III. TRANING RELAY FAULTS USING BPNN The basic element of a neural network is the processing node, which performs two functions. First it sums the values of its inputs. This sum is then passed through an arbitrary activation function to produce the node's output value as shown in Fig. 2. ( ) (1) Where NET is the sum of weighted inputs to the processing node The goal of the training is to minimize the overall error between the desired and actual outputs of the network. It has been proven that backpropagation learning with sufficient hidden layers can approximate any nonlinear functions to arbitrary accuracy. This makes backpropagation learning neural network a good candidate for signal prediction and system modeling [4]. MATLAB Neural Network/ Data Manager (nntool) is used to create the network. Firstly, the input which is the general faults and output which is the actual faults are loaded into the network as input and target data, respectively. Then, a network property is defined as the feedforward backpropagation. Neural network parameters such as number of layers, number of neurons and the type of transfer function needs also to be defined. Once defined, the network is now ready to be trained. The training of the network need to be conducted repeatedly since the network parameters are changed through trial and error. The network parameters that gives the minimum training Mean Squared Error (MSE) will be chosen and tested. The performance of the trained neural network can be determined through Mean Square Error (MSE) which is defined as ( ) (2) Where = vector of n prediction and = vector of true value. The training went through several combinations of network parameters and it was found that the network starts to show improvements in the performance when the number of hidden layer is set to two layers and the transfer function for both layers are defined as log-sigmoid (logsig). Table I listed ten of the training processes conducted with the two logsig hidden layers and uses purelin transfer function at the output layer. From Table I it can be seen that gives the minimum MSE of x with the 30 neurons are defined for the first hidden layer and fifteen neurons for the second hidden layer. Note that for the output layer, the number of neuron is one since there is only one output signal. With this trained neural network, the network is then exported to MATLAB workspace and simulink blocks diagram is generated. Both Simulink blocks diagram and the architecture of the network is shown in Figure 3a and Figure 3b. Fig 2. Neuron model For the backpropagation training algorithm, the activation function must be differentiable. The most common form is the sigmoid function which is defined as No. of Hidden Neuron (L1) TABLE I. PERFORMANCE OF NEURAL NETWORK No. of Hidden Neuron (L2) Transfer Function at L1 Transfer Functio n at L2 Transfer Function at Output Mean Square Error (MSE) 30 5 Logsig Logsig Purelin Logsig Logsig Purelin Logsig Logsig Purelin x Logsig Logsig Purelin Logsig Logsig Purelin x Logsig Logsig Purelin Logsig Logsig Purelin x Logsig Logsig Purelin Logsig Logsig Purelin x Logsig Logsig Purelin x

3 (a) the GUI component such as Push Button. In this paper, the Push Button is named as Simulate as shown in Fig.4. By using M-file Editor, code can be added to the call-back to perform the functions needed. When user selects the General Fault Categories in the Input List Box and click Simulate as shown in Fig. 4, a numerical input data representing the General Faults Categories will be generated and sent to the Simulink trained neural network to identify the corresponding numerical value of the actual fault. Once the neural network has simulated the numerical value of the actual fault it is then sent bank to the GUI and the numerical actual fault is transformed into the respective actual fault. Thus, not only the user does not have to know the neural network that runs with the GUI but also even an inexperienced user can identify the actual fault. General fault is choose. (b) Fig. 3. Simulink Block Diagrams of (a) Trained Fault Identification Neural Network (b) The architecture of the Trained Neural Network The trained neural network is then tested and see whether it is able to output the correct numerical value when a single numerical value is inserted. Table II shows some of the numerical inputs tested into the neural network ad ir shows that the neural network is able to give the corresponding numerical value. TABLE II. RESULTS VALIDATI Input Data Target Data Output Data Since the neural network can only performs in numerical value, thus it is more users friendly when it performs in graphical user interface (GUI). The general faults of overcurrent relay can choose directly from the GUI and simulated results also can be performing in statement form instead of numerical form. IV OVERCURRENT RELAY FAULT IDENTIFICATI WITH GUI In this paper, the trained neural network is presented in a graphical user interface (GUI) to make the system user friendly. Graphical user interface is a graphical display in one or more windows where it contains controls. GUIDE is providing the tools for creating GUIs. The GUIDE will automatically generate an M-file that controls the GUI operation after the GUI is saved. This M- file provide code to initialize the GUI. It also contains a framework for the GUI call-backs which is the routines that execute in response to user generated events interacts with Push Button Actual fault is displayed after pressing the simulate button. Fig 4. Example of simulate results V. CCLUSSI The overcurrent relay faults are identified using Backpropagation Neural Network where it consists of two hidden layers and one output layer with trained MSE equal to 0.422x The raw data of overcurrent relay faults is interpreted into numerical data so that it can be trained using neural network. A GUI is design based on the trained neural network where it is able to perform the results in statement forms instead of numerical form. Thus, when general fault of overcurrent relay is chosen in the system, the actual faults can be generated directly through this GUI. This developed GUI can be used even by non-experienced personal to identify the actual fault of the overcurrent relay. REFERENCES [1] S. P. A. K. Abdullah Asuhaimi Mohd Zin, "Protection System Analysis Using Fault Signatures in Malaysia," Electrical Power \& Energy Systems, vol. 45, pp

4 205, February [2] Blackout: World's 13 Biggest Power Outages, [3] A. K. R. Yin Lee Goh and F. H. N. Aidil Azwin Zainul Abidin, "Modelling of Overcurrent Relay Using Digital Signal Processor," in IEEE Symposium on Industrial Electronics and Applications (ISIEA 2010), [4] A. Abraham, Handbook of Measuring System Design, P. H. Sydenham and R. Thorn, Eds., John Wiley & Sons,

5 APPENDIX A General Overcurrent Relay Faults OVERCURRENT RELAY GENERAL AND ACTUAL FAULTS Actual Overcurrent Relay Faults Input Relay (General Overcurrent Relay Faults) Output Relay (Actual Overcurrent Relay Faults) Display Error Code_Software Failure DSP Chip Failure 1 3 Display Error Code_Relay Booted DSP Chip Failure Repeatedly 2 3 No Screen Display_Healthy & O/C LED DSP Chip Failure 4 3 Others_Same Termial Block Sticker Defect in Assembly 21 1 Others_Relay Tread Faulty Defect in Assembly 22 1 Defective Component_Defective RL2 Contact Defective Component 10 2 Relay Can't Power Up_Stop Working DSP Chip Failure After A While 28 4 Key Pad Malfunction_Keypad LEFT Faulty Defect in Assembly 12 1 Others_Hardware Alarm DSP Chip Failure 23 4 Faulty LCD Module Faulty LCD Module 11 5 Key Pad Malfunction_Keypad No DSP Chip Failure Respond 13 4 No Screen Display_Healthy & LED D8 DSP Chip Failure 5 4 Key Pad Malfunction_LCD Display "ALARM" Defective Component 14 2 Relay Booting Error_Healthy On & DSP Chip Failure Booted Repeatedly 26 3 Relay Booting Error_Healthy On & DSP Chip Failure Booted Repeatedly 26 3 No Screen Display_Healthy & LED DSP Chip Failure D8 Blinked 6 4 Hardware Failure Hardware Failure 18 6 Key Pad Malfunction_LED D8 Blinked & DSP Chip Failure Mini Relay Tripping

6 Relay Can't Power Up_Healthy & Warning LED & LCD Display Defective Component 29 2 "ALARM" In Beginning Hardware Failure Hardware Failure 18 6 Key Pad Malfunction_Healthy DSP Chip Failure 16 4 No Screen Display_Healthy & LCD Backlight Faulty LCD Module 7 5 No Screen Display_Healthy & O/C LED DSP Chip Failure 4 3 No Fault Found_Initially No Display No Fault Found 19 7 No Fault Found_L1 & L2 Display Value No Fault Found 20 7 Key Pad Malfunction_Front Port Failed Respond & No. 4 Configurable LED Blinking Key Pad Malfunction_Front Port Failed Respond & No. 4 Configurable LED Blinking Key Pad Malfunction_Front Port Failed Respond & No. 4 Configurable LED Blinking Normal Repair 17 8 Normal Repair 17 8 Normal Repair 17 8 No Screen Display_Healthy & O/C LED Normal Repair 4 8 No Screen Display_Relay No Display Normal Repair 8 8 Display Error Code_Display Booting Error Message Normal Repair 3 8 Relay Unable to Reset Normal Repair 30 8 Others_LED Warning Normal Repair 24 8 No Screen Display_Relay No Display Normal Repair 8 8 Others_No Software Normal Repair

MATLAB representation of neural network Outline Neural network with single-layer of neurons. Neural network with multiple-layer of neurons.

MATLAB representation of neural network Outline Neural network with single-layer of neurons. Neural network with multiple-layer of neurons. MATLAB representation of neural network Outline Neural network with single-layer of neurons. Neural network with multiple-layer of neurons. Introduction: Neural Network topologies (Typical Architectures)

More information

Neural Networks Laboratory EE 329 A

Neural Networks Laboratory EE 329 A Neural Networks Laboratory EE 329 A Introduction: Artificial Neural Networks (ANN) are widely used to approximate complex systems that are difficult to model using conventional modeling techniques such

More information

International Journal of Electrical and Computer Engineering 4: Application of Neural Network in User Authentication for Smart Home System

International Journal of Electrical and Computer Engineering 4: Application of Neural Network in User Authentication for Smart Home System Application of Neural Network in User Authentication for Smart Home System A. Joseph, D.B.L. Bong, and D.A.A. Mat Abstract Security has been an important issue and concern in the smart home systems. Smart

More information

Early tube leak detection system for steam boiler at KEV power plant

Early tube leak detection system for steam boiler at KEV power plant Early tube leak detection system for steam boiler at KEV power plant Firas B. Ismail 1a,, Deshvin Singh 1, N. Maisurah 1 and Abu Bakar B. Musa 1 1 Power Generation Research Centre, College of Engineering,

More information

Research on Evaluation Method of Product Style Semantics Based on Neural Network

Research on Evaluation Method of Product Style Semantics Based on Neural Network Research Journal of Applied Sciences, Engineering and Technology 6(23): 4330-4335, 2013 ISSN: 2040-7459; e-issn: 2040-7467 Maxwell Scientific Organization, 2013 Submitted: September 28, 2012 Accepted:

More information

International Journal of Advanced Research in Computer Science and Software Engineering

International Journal of Advanced Research in Computer Science and Software Engineering Volume 3, Issue 4, April 203 ISSN: 77 2X International Journal of Advanced Research in Computer Science and Software Engineering Research Paper Available online at: www.ijarcsse.com Stock Market Prediction

More information

Planar Robot Arm Performance: Analysis with Feedforward Neural Networks

Planar Robot Arm Performance: Analysis with Feedforward Neural Networks Planar Robot Arm Performance: Analysis with Feedforward Neural Networks Abraham Antonio López Villarreal, Samuel González-López, Luis Arturo Medina Muñoz Technological Institute of Nogales Sonora Mexico

More information

Review on Methods of Selecting Number of Hidden Nodes in Artificial Neural Network

Review on Methods of Selecting Number of Hidden Nodes in Artificial Neural Network Available Online at www.ijcsmc.com International Journal of Computer Science and Mobile Computing A Monthly Journal of Computer Science and Information Technology IJCSMC, Vol. 3, Issue. 11, November 2014,

More information

J. Electrical Systems 13-2 (2017): Regular paper

J. Electrical Systems 13-2 (2017): Regular paper Nor Hazwani Idris 1, *, Nur Ashida Salim 1, Muhammad Murtadha Othman 1 and Zuhaila Mat Yasin 1 Regular paper Prediction of Cascading Collapse Occurrence due to the Effect of Hidden Failure of a Protection

More information

Bus Protection Application Challenges

Bus Protection Application Challenges Bus Protection Application Challenges KN Dinesh Babu - Megger JC Theron, Lubomir Sevov GE Grid Solutions 2017 Texas A&M Protective Relay Conference Content Introduction Application Challenges Increase

More information

Software Based Transmission Line Fault Analysis

Software Based Transmission Line Fault Analysis International journal of scientific and technical research in engineering (IJSTRE) www.ijstre.com Volume 1 Issue 3 ǁ June 2016. Software Based Transmission Line Fault Analysis E. O. Ogunti 1 1 (Department

More information

A neural network that classifies glass either as window or non-window depending on the glass chemistry.

A neural network that classifies glass either as window or non-window depending on the glass chemistry. A neural network that classifies glass either as window or non-window depending on the glass chemistry. Djaber Maouche Department of Electrical Electronic Engineering Cukurova University Adana, Turkey

More information

Gupta Nikita $ Kochhar

Gupta Nikita $ Kochhar Volume 3, Issue 5, May 2013 ISSN: 2277 128X International Journal of Advanced Research in Computer Science and Software Engineering Research Paper Available online at: www.ijarcsse.com Congestion Control

More information

Neural Networks (Overview) Prof. Richard Zanibbi

Neural Networks (Overview) Prof. Richard Zanibbi Neural Networks (Overview) Prof. Richard Zanibbi Inspired by Biology Introduction But as used in pattern recognition research, have little relation with real neural systems (studied in neurology and neuroscience)

More information

CHAPTER 7 MASS LOSS PREDICTION USING ARTIFICIAL NEURAL NETWORK (ANN)

CHAPTER 7 MASS LOSS PREDICTION USING ARTIFICIAL NEURAL NETWORK (ANN) 128 CHAPTER 7 MASS LOSS PREDICTION USING ARTIFICIAL NEURAL NETWORK (ANN) Various mathematical techniques like regression analysis and software tools have helped to develop a model using equation, which

More information

Implementation of a Library for Artificial Neural Networks in C

Implementation of a Library for Artificial Neural Networks in C Implementation of a Library for Artificial Neural Networks in C Jack Breese TJHSST Computer Systems Lab 2007-2008 June 10, 2008 1 Abstract In modern computing, there are several approaches to pattern recognition

More information

Real Time Monitoring of

Real Time Monitoring of Real Time Monitoring of Cascading Events Mladen Kezunovic Nan Zhang, Hongbiao Song Texas A&M University Tele-Seminar, March 28, 2006 Project Reports (S-19) M. Kezunovic, H. Song and N. Zhang, Detection,

More information

Supervised Learning in Neural Networks (Part 2)

Supervised Learning in Neural Networks (Part 2) Supervised Learning in Neural Networks (Part 2) Multilayer neural networks (back-propagation training algorithm) The input signals are propagated in a forward direction on a layer-bylayer basis. Learning

More information

A Novel Supervisory Control for Residential Building Protection System

A Novel Supervisory Control for Residential Building Protection System A Novel Supervisory Control for Residential Building Protection System J. Rajaram, *N M Nor, T. Ibrahim, H. Daud, P. H. Shaikh * Electrical & Electronic Engineering Department, Universiti Teknologi PETRONAS,

More information

Finding Dominant Parameters For Fault Diagnosis Of a Single Bearing System Using Back Propagation Neural Network

Finding Dominant Parameters For Fault Diagnosis Of a Single Bearing System Using Back Propagation Neural Network International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:13 No:01 40 Finding Dominant Parameters For Fault Diagnosis Of a Single Bearing System Using Back Propagation Neural Network

More information

Artificial Neural Network Based Byzantine Agreement Protocol

Artificial Neural Network Based Byzantine Agreement Protocol P Artificial Neural Network Based Byzantine Agreement Protocol K.W. Lee and H.T. Ewe 2 Faculty of Engineering & Technology 2 Faculty of Information Technology Multimedia University Multimedia University

More information

ASPEN Software - Introduction

ASPEN Software - Introduction ASPEN Software - Introduction ASPEN is a very user-friendly software package that is capable of doing load-flow and shortcircuit studies, as well as relay application and coordination. Its simple graphical

More information

SECTION ENCLOSED SWITCHES AND CIRCUIT BREAKERS

SECTION ENCLOSED SWITCHES AND CIRCUIT BREAKERS SECTION 262816 - ENCLOSED SWITCHES AND CIRCUIT BREAKERS PART 1 - GENERAL 1.1 RELATED DOCUMENTS A. Drawings and general provisions of the Contract, including General and Supplementary Conditions and Division

More information

Development of Emulation Network Analyzer Tool for Computer Network Planning

Development of Emulation Network Analyzer Tool for Computer Network Planning Bulletin of Electrical Engineering and Informatics ISSN: 2302-9285 Vol. 5, No. 2, June 2016, pp. 219~229, DOI: 10.11591/eei.v5i2.557 219 Development of Emulation Network Analyzer Tool for Computer Network

More information

Date of Publication: THREE PHASE LINE FAULT DETECTION ON DISTRIBUTION LINES USING GSM TECHNIQUE

Date of Publication: THREE PHASE LINE FAULT DETECTION ON DISTRIBUTION LINES USING GSM TECHNIQUE International Journal of Emerging Technology and Innovative Volume 2, Issue 3, March 2016 (ISSN: 2394 6598) Date of Publication: 29.03.2016 THREE PHASE LINE FAULT DETECTION ON DISTRIBUTION LINES USING

More information

A Data Classification Algorithm of Internet of Things Based on Neural Network

A Data Classification Algorithm of Internet of Things Based on Neural Network A Data Classification Algorithm of Internet of Things Based on Neural Network https://doi.org/10.3991/ijoe.v13i09.7587 Zhenjun Li Hunan Radio and TV University, Hunan, China 278060389@qq.com Abstract To

More information

Power System Protection with Relay Co-Ordination

Power System Protection with Relay Co-Ordination Power System Protection with Relay Co-Ordination Mehulkumar D. Devdhariya 1, Vibhuti R. Adroja 2, Kajal M. Rafaliya 3, Prof. Manan M. Desai 4 1,2,3 Students, B.E. in Electrical Engineering, 4 Assistant

More information

COMBINING NEURAL NETWORKS FOR SKIN DETECTION

COMBINING NEURAL NETWORKS FOR SKIN DETECTION COMBINING NEURAL NETWORKS FOR SKIN DETECTION Chelsia Amy Doukim 1, Jamal Ahmad Dargham 1, Ali Chekima 1 and Sigeru Omatu 2 1 School of Engineering and Information Technology, Universiti Malaysia Sabah,

More information

1. Approximation and Prediction Problems

1. Approximation and Prediction Problems Neural and Evolutionary Computing Lab 2: Neural Networks for Approximation and Prediction 1. Approximation and Prediction Problems Aim: extract from data a model which describes either the depence between

More information

Machine Learning 13. week

Machine Learning 13. week Machine Learning 13. week Deep Learning Convolutional Neural Network Recurrent Neural Network 1 Why Deep Learning is so Popular? 1. Increase in the amount of data Thanks to the Internet, huge amount of

More information

ANN-Based Modeling for Load and Main Steam Pressure Characteristics of a 600MW Supercritical Power Generating Unit

ANN-Based Modeling for Load and Main Steam Pressure Characteristics of a 600MW Supercritical Power Generating Unit ANN-Based Modeling for Load and Main Steam Pressure Characteristics of a 600MW Supercritical Power Generating Unit Liangyu Ma, Zhiyuan Gao Automation Department, School of Control and Computer Engineering

More information

A Neuro-Fuzzy Application to Power System

A Neuro-Fuzzy Application to Power System 2009 International Conference on Machine Learning and Computing IPCSIT vol.3 (2011) (2011) IACSIT Press, Singapore A Neuro-Fuzzy Application to Power System Ahmed M. A. Haidar 1, Azah Mohamed 2, Norazila

More information

Blackstart Hardware-in-the-loop Relay Testing Platform

Blackstart Hardware-in-the-loop Relay Testing Platform 21, rue d Artois, F-75008 PARIS CIGRE US National Committee http : //www.cigre.org 2016 Grid of the Future Symposium Blackstart Hardware-in-the-loop Relay Testing Platform R. LIU R. SUN M. TANIA Washington

More information

Artificial Neural Network (ANN) Approach for Predicting Friction Coefficient of Roller Burnishing AL6061

Artificial Neural Network (ANN) Approach for Predicting Friction Coefficient of Roller Burnishing AL6061 International Journal of Machine Learning and Computing, Vol. 2, No. 6, December 2012 Artificial Neural Network (ANN) Approach for Predicting Friction Coefficient of Roller Burnishing AL6061 S. H. Tang,

More information

For Monday. Read chapter 18, sections Homework:

For Monday. Read chapter 18, sections Homework: For Monday Read chapter 18, sections 10-12 The material in section 8 and 9 is interesting, but we won t take time to cover it this semester Homework: Chapter 18, exercise 25 a-b Program 4 Model Neuron

More information

CHAPTER VI BACK PROPAGATION ALGORITHM

CHAPTER VI BACK PROPAGATION ALGORITHM 6.1 Introduction CHAPTER VI BACK PROPAGATION ALGORITHM In the previous chapter, we analysed that multiple layer perceptrons are effectively applied to handle tricky problems if trained with a vastly accepted

More information

Neuro-Fuzzy Computing

Neuro-Fuzzy Computing CSE531 Neuro-Fuzzy Computing Tutorial/Assignment 2: Adaline and Multilayer Perceptron About this tutorial The objective of this tutorial is to study: You can create a single (composite) layer of neurons

More information

SURGE PROTECTION USER MANUAL. Surge Protection Device PORTABLE POWER DISTRIBUTION SURGE PROTECTION DEVICE (SPD) B

SURGE PROTECTION USER MANUAL. Surge Protection Device PORTABLE POWER DISTRIBUTION SURGE PROTECTION DEVICE (SPD) B SURGE PROTECTION PORTABLE POWER DISTRIBUTION SURGE PROTECTION DEVICE (SPD) USER MANUAL Surge Protection Device B 01 156 9000 This Surge Protection Device is intended for professional use only Read this

More information

PERFORMANCE ANALYSIS AND VALIDATION OF CLUSTERING ALGORITHMS USING SOFT COMPUTING TECHNIQUES

PERFORMANCE ANALYSIS AND VALIDATION OF CLUSTERING ALGORITHMS USING SOFT COMPUTING TECHNIQUES PERFORMANCE ANALYSIS AND VALIDATION OF CLUSTERING ALGORITHMS USING SOFT COMPUTING TECHNIQUES Bondu Venkateswarlu Research Scholar, Department of CS&SE, Andhra University College of Engineering, A.U, Visakhapatnam,

More information

VAMP 221. Table of Contents

VAMP 221. Table of Contents 2 V221/EN M/E019 VAMP 221 Table of Contents Table of Contents 1. General... 7 1.1. VAMP 221 arc protection system components... 7 1.1.1. Central unit VAMP 221... 9 1.1.2. I/O units VAM 12L / VAM 12 LD,

More information

Use of Artificial Neural Networks to Investigate the Surface Roughness in CNC Milling Machine

Use of Artificial Neural Networks to Investigate the Surface Roughness in CNC Milling Machine Use of Artificial Neural Networks to Investigate the Surface Roughness in CNC Milling Machine M. Vijay Kumar Reddy 1 1 Department of Mechanical Engineering, Annamacharya Institute of Technology and Sciences,

More information

Identification of Multisensor Conversion Characteristic Using Neural Networks

Identification of Multisensor Conversion Characteristic Using Neural Networks Sensors & Transducers 3 by IFSA http://www.sensorsportal.com Identification of Multisensor Conversion Characteristic Using Neural Networks Iryna TURCHENKO and Volodymyr KOCHAN Research Institute of Intelligent

More information

CS6220: DATA MINING TECHNIQUES

CS6220: DATA MINING TECHNIQUES CS6220: DATA MINING TECHNIQUES Image Data: Classification via Neural Networks Instructor: Yizhou Sun yzsun@ccs.neu.edu November 19, 2015 Methods to Learn Classification Clustering Frequent Pattern Mining

More information

Week 3: Perceptron and Multi-layer Perceptron

Week 3: Perceptron and Multi-layer Perceptron Week 3: Perceptron and Multi-layer Perceptron Phong Le, Willem Zuidema November 12, 2013 Last week we studied two famous biological neuron models, Fitzhugh-Nagumo model and Izhikevich model. This week,

More information

Research of Fault Diagnosis in Flight Control System Based on Fuzzy Neural Network

Research of Fault Diagnosis in Flight Control System Based on Fuzzy Neural Network Available online at www.sciencedirect.com Procedia Engineering 5 (2 ) 7 74 Research of Fault Diagnosis in Flight Control System Based on Fuzzy Neural Network Peng Zhang a, Rui Zhang b, Xiao-xiao Ran b

More information

Power Load Forecasting Based on ABC-SA Neural Network Model

Power Load Forecasting Based on ABC-SA Neural Network Model Power Load Forecasting Based on ABC-SA Neural Network Model Weihua Pan, Xinhui Wang College of Control and Computer Engineering, North China Electric Power University, Baoding, Hebei 071000, China. 1471647206@qq.com

More information

Back propagation Algorithm:

Back propagation Algorithm: Network Neural: A neural network is a class of computing system. They are created from very simple processing nodes formed into a network. They are inspired by the way that biological systems such as the

More information

General characteristics of C-type relay modules

General characteristics of C-type relay modules General characteristics of C-type relay modules User s manual and Technical description B I > I >> Indicators for measured values I L1 I L2 I L3 IRF Self-supervision alarm indicator (Internal Relay Fault)

More information

An Algorithm For Training Multilayer Perceptron (MLP) For Image Reconstruction Using Neural Network Without Overfitting.

An Algorithm For Training Multilayer Perceptron (MLP) For Image Reconstruction Using Neural Network Without Overfitting. An Algorithm For Training Multilayer Perceptron (MLP) For Image Reconstruction Using Neural Network Without Overfitting. Mohammad Mahmudul Alam Mia, Shovasis Kumar Biswas, Monalisa Chowdhury Urmi, Abubakar

More information

NEURAL NETWORK BASED REGRESSION TESTING

NEURAL NETWORK BASED REGRESSION TESTING NEURAL NETWORK BASED REGRESSION TESTING Esha Khanna Assistant Professor, IT Department, D. A.V Institute of Management, (India) ABSTRACT Regression testing re-executes all the test cases in order to ensure

More information

Utility Implements Communications- Assisted Special Protection and Control Schemes for Distribution Substations

Utility Implements Communications- Assisted Special Protection and Control Schemes for Distribution Substations Utility Implements Communications- Assisted Special Protection and Control Schemes for Distribution Substations Michael. Duff City of College Station Payal Gupta, Dharmendra Prajapati, and Alex Langseth

More information

Table of Contents. Automatic Transfer Switch. User Manual ATS-16A/20A/30A/32A

Table of Contents. Automatic Transfer Switch. User Manual ATS-16A/20A/30A/32A Table of Contents 1. Important Safety Instructions... 2 2. Overview... 3 3. Product Functional Description... 4 3.1. Front Panel... 4 3.2. LCD Display Description... 5 3.3. Rear Panel... 6 3.4. Communication

More information

Instantaneously trained neural networks with complex inputs

Instantaneously trained neural networks with complex inputs Louisiana State University LSU Digital Commons LSU Master's Theses Graduate School 2003 Instantaneously trained neural networks with complex inputs Pritam Rajagopal Louisiana State University and Agricultural

More information

Optimizing Number of Hidden Nodes for Artificial Neural Network using Competitive Learning Approach

Optimizing Number of Hidden Nodes for Artificial Neural Network using Competitive Learning Approach Available Online at www.ijcsmc.com International Journal of Computer Science and Mobile Computing A Monthly Journal of Computer Science and Information Technology IJCSMC, Vol. 4, Issue. 5, May 2015, pg.358

More information

USER MANUAL. Uninterruptible Power Supply Line-interactive VCL Series UPS VA. GE Critical Power

USER MANUAL. Uninterruptible Power Supply Line-interactive VCL Series UPS VA. GE Critical Power Critical Power USER MANUAL Uninterruptible Power Supply Line-interactive VCL Series UPS 400 600 800 1000 1500 VA GE Consumer & Industrial SA General Electric Company CH 6595 Riazzino (Locarno) Switzerland

More information

Neural Network Neurons

Neural Network Neurons Neural Networks Neural Network Neurons 1 Receives n inputs (plus a bias term) Multiplies each input by its weight Applies activation function to the sum of results Outputs result Activation Functions Given

More information

Concept of Automation in Management of Electric Power Systems Richard Joseph, Nerey Mvungi

Concept of Automation in Management of Electric Power Systems Richard Joseph, Nerey Mvungi Concept of Automation in Management of Electric Power Systems Richard Joseph, Nerey Mvungi Abstract An electric power system includes a generating, a transmission, a distribution, and consumers subsystems.

More information

A Matlab Tool for Analyzing and Improving Fault Tolerance of Artificial Neural Networks

A Matlab Tool for Analyzing and Improving Fault Tolerance of Artificial Neural Networks A Matlab Tool for Analyzing and Improving Fault Tolerance of Artificial Neural Networks Rui Borralho*. Pedro Fontes*. Ana Antunes*. Fernando Morgado Dias**. *Escola Superior de Tecnologia de Setúbal do

More information

PowerStudio is simple, powerful and user-friendly software. Provides complete monitoring of energy in industrial and commercial buildings.

PowerStudio is simple, powerful and user-friendly software. Provides complete monitoring of energy in industrial and commercial buildings. Circutor - Founded in 1973 in Barcelona, Spain - Wide product portfolio with solutions for energy metering systems, equipment for protection and energy management - Solutions for EnMS, reactive power compensation,

More information

AV-4000 GSM AV-4000 DVR

AV-4000 GSM AV-4000 DVR E a s y L o a d e r AV-4000 GSM AV-4000 DVR (for LED keypads) Integrated Alarm Control panel & Communicator Series 4000 Appendix for GSM Alarm Panels Version 1.05 Edition I This product is subject to continuous

More information

Neural Networks. Lab 3: Multi layer perceptrons. Nonlinear regression and prediction.

Neural Networks. Lab 3: Multi layer perceptrons. Nonlinear regression and prediction. Neural Networks. Lab 3: Multi layer perceptrons. Nonlinear regression and prediction. 1. Defining multi layer perceptrons. A multi layer perceptron (i.e. feedforward neural networks with hidden layers)

More information

ASPEN Software - Introduction

ASPEN Software - Introduction ASPEN Software - Introduction ASPEN is a very user-friendly software package that is capable of doing load-flow and shortcircuit studies, as well as relay application and coordination. Its simple graphical

More information

THE discrete multi-valued neuron was presented by N.

THE discrete multi-valued neuron was presented by N. Proceedings of International Joint Conference on Neural Networks, Dallas, Texas, USA, August 4-9, 2013 Multi-Valued Neuron with New Learning Schemes Shin-Fu Wu and Shie-Jue Lee Department of Electrical

More information

COMP 551 Applied Machine Learning Lecture 14: Neural Networks

COMP 551 Applied Machine Learning Lecture 14: Neural Networks COMP 551 Applied Machine Learning Lecture 14: Neural Networks Instructor: (jpineau@cs.mcgill.ca) Class web page: www.cs.mcgill.ca/~jpineau/comp551 Unless otherwise noted, all material posted for this course

More information

Image Compression: An Artificial Neural Network Approach

Image Compression: An Artificial Neural Network Approach Image Compression: An Artificial Neural Network Approach Anjana B 1, Mrs Shreeja R 2 1 Department of Computer Science and Engineering, Calicut University, Kuttippuram 2 Department of Computer Science and

More information

ANGE Automatic Neural Generator

ANGE Automatic Neural Generator ANGE Automatic Neural Generator Leonardo Reis 1, Luis Aguiar 1, Darío Baptista 1, Fernando Morgado Dias 1 1 Centro de Competências de Ciências Exactas e da Engenharia, Universidade da Madeira Campus penteada,

More information

Busbar protection REB 670

Busbar protection REB 670 Gunnar Stranne Busbar protection REB 670 Rio de Janeiro April 23-25, 2006 ABB Power Technologies AB 2 3 4 REB 500 is enabled for use with IEC61850 communication Benefits of REB 670 outstanding features

More information

ISSN: (PRINT) ISSN: (ONLINE)

ISSN: (PRINT) ISSN: (ONLINE) IJRECE VOL. 5 ISSUE 2 APR.-JUNE. 217 ISSN: 2393-928 (PRINT) ISSN: 2348-2281 (ONLINE) Code Clone Detection Using Metrics Based Technique and Classification using Neural Network Sukhpreet Kaur 1, Prof. Manpreet

More information

Load forecasting of G.B. Pant University of Agriculture & Technology, Pantnagar using Artificial Neural Network

Load forecasting of G.B. Pant University of Agriculture & Technology, Pantnagar using Artificial Neural Network Load forecasting of G.B. Pant University of Agriculture & Technology, Pantnagar using Artificial Neural Network VIKRAM VEER SINGH 1 Electrical Engineering Department College of Technology GBPUAT, Pantnagar,

More information

Applications of Programmable Logic Controllers DG31 34

Applications of Programmable Logic Controllers DG31 34 Applications of Programmable Logic Controllers DG31 34 Purpose Unit purpose: This Unit is designed to introduce candidates to Programmable Logic Controllers (PLCs) and enable them to understand how PLCs

More information

PERMANENT DOCUMENT. Annex AU to Routine Test Requirements for manufacturers (as per Article 9 of the Agreement)

PERMANENT DOCUMENT. Annex AU to Routine Test Requirements for manufacturers (as per Article 9 of the Agreement) PERMANENT DOCUMENT ENEC 303 Annex AU Annex AU to Routine Test Requirements for manufacturers (as per Article 9 of the Agreement) Safety requirements of Audio/video, information and communication technology

More information

FAULT INDICATOR Type EFI/P

FAULT INDICATOR Type EFI/P APPLICATIONS Detect earth faults in medium voltage net Highlights Earth fault passage indicator combined Automatic and manual reset Front push button for test and reset Selectable earth fault trip currents

More information

Protection Redundancy Main I and Main II Security and Reliability

Protection Redundancy Main I and Main II Security and Reliability Protection Redundancy Main I and Main II Security and Reliability Roy Moxley, Siemens Abstract - Protection history and practice has long had two relays applied for protection of one line or piece of primary

More information

SNIWD: Simultaneous Weight Noise Injection With Weight Decay for MLP Training

SNIWD: Simultaneous Weight Noise Injection With Weight Decay for MLP Training SNIWD: Simultaneous Weight Noise Injection With Weight Decay for MLP Training John Sum and Kevin Ho Institute of Technology Management, National Chung Hsing University Taichung 4, Taiwan. pfsum@nchu.edu.tw

More information

REACTIVE ENERGY REGULATORS

REACTIVE ENERGY REGULATORS REACTIVE ENERGY REGULATORS computer MAX 6f 12Vdc computer MAX 12f 12Vdc Instruction Manual (M066B01-03-15B) - 1 - Table of Contents 1 INTRODUCTION AND SAFETY TIPS... 3 1.1 CHECKS TO BE CARRIED OUT WHEN

More information

INTELLIGENT SEISMIC STRUCTURAL HEALTH MONITORING SYSTEM FOR THE SECOND PENANG BRIDGE OF MALAYSIA

INTELLIGENT SEISMIC STRUCTURAL HEALTH MONITORING SYSTEM FOR THE SECOND PENANG BRIDGE OF MALAYSIA INTELLIGENT SEISMIC STRUCTURAL HEALTH MONITORING SYSTEM FOR THE SECOND PENANG BRIDGE OF MALAYSIA Reni Suryanita Faculty of Engineering Civil Engineering Department University of Riau, Pekanbaru reni.suryanita@lecturer.unri.ac.id

More information

GROUND FAULT CIRCUIT INTERUPTERS

GROUND FAULT CIRCUIT INTERUPTERS Title: GFCI Effective Date: 11/14/2014 Control Number: THG_0031 Revision Number: 1 Date: 10/21/2015 Annual Review Completed: 5/13/2015 GROUND FAULT CIRCUIT INTERUPTERS CONTENTS I. PURPOSE II. GENERAL REQUIREMENTS

More information

International Journal of Advance Engineering and Research Development. Flow Control Loop Analysis for System Modeling & Identification

International Journal of Advance Engineering and Research Development. Flow Control Loop Analysis for System Modeling & Identification Scientific Journal of Impact Factor(SJIF): 3.134 e-issn(o): 2348-4470 p-issn(p): 2348-6406 International Journal of Advance Engineering and Research Development Volume 2,Issue 5, May -2015 Flow Control

More information

An Artificial Intelligence Based Approach for Bus Bar Differential Protection Faults Analysis in Distribution Systems

An Artificial Intelligence Based Approach for Bus Bar Differential Protection Faults Analysis in Distribution Systems An Artificial Intelligence Based Approach for Bus Bar Differential Protection Faults Analysis in Distribution Systems By Eng. Sameh Moustafa Mohamed Abd Elgowad Elbana A thesis submitted to The Faculty

More information

Predictive Maintenance in a Mission Critical Environment

Predictive Maintenance in a Mission Critical Environment CSI 2130 Predictive Maintenance in a Mission Critical Environment This paper discusses the need to proactively protect mechanical and electrical support systems in data processing centers and data storage

More information

p. 1 p. 9 p. 26 p. 56 p. 62 p. 68

p. 1 p. 9 p. 26 p. 56 p. 62 p. 68 The Challenges and Opportunities Faced by Utilities using Modern Protection and Control Systems - Paper Unavailable Trends in Protection and Substation Automation Systems and Feed-backs from CIGRE Activities

More information

LENGTH-ANGLE-RADIUS-LOAD INDICATING SYSTEM

LENGTH-ANGLE-RADIUS-LOAD INDICATING SYSTEM www.patamerica.com PAT America, Inc. LENGTH-ANGLE-RADIUS-LOAD INDICATING SYSTEM EI65/0005 OPERATOR S MANUAL P/N 056-065-190-005, Rev. C, 6/15/00 Operator s Manual EI65 / 0005 NOTICE The information in

More information

Open Access Research on the Prediction Model of Material Cost Based on Data Mining

Open Access Research on the Prediction Model of Material Cost Based on Data Mining Send Orders for Reprints to reprints@benthamscience.ae 1062 The Open Mechanical Engineering Journal, 2015, 9, 1062-1066 Open Access Research on the Prediction Model of Material Cost Based on Data Mining

More information

Control Equipment Limited 96 Zone Discovery Panel, 32 Zone Voyager Panel and Integra Network Repeater 40- Character User Manual

Control Equipment Limited 96 Zone Discovery Panel, 32 Zone Voyager Panel and Integra Network Repeater 40- Character User Manual Created by: A.C. Revised by: G.G. Approved by: JBJ Control Equipment Limited 96 Zone Discovery Panel, 32 Zone Voyager Panel and Integra Network Repeater 40- Character User Manual Table of Contents 1. Introduction...3

More information

Important Considerations in Testing and Commissioning Digital Relays

Important Considerations in Testing and Commissioning Digital Relays Important Considerations in Testing and Commissioning Digital Relays Drew Welton, Beckwith Electric Co. Inc. Will Knapek, OMICRON electronics Corp. USA Justification Digital technology in protection relays

More information

Implementation of JIT Technique in a Mixed Model Assembly Line: A Case Study

Implementation of JIT Technique in a Mixed Model Assembly Line: A Case Study Volume-6, Issue-2, March-April 2016 International Journal of Engineering and Management Research Page Number: 485-489 Implementation of JIT Technique in a Mixed Model Assembly Line: A Case Study Himanshu

More information

FUSELESS ists C,H,K & G

FUSELESS ists C,H,K & G FUSELESS ists C,H,K & G How does it work? ists SYSTEM DIAGRAM ists C, H, K & G DETECTION The first step of the detection process involves attenuating the three waveforms Supply 1, Supply 2 and output supply

More information

Application of Nonlinear Later TV Edition in Gigabit Ethernet. Hong Ma

Application of Nonlinear Later TV Edition in Gigabit Ethernet. Hong Ma 3rd International Conference on Science and Social Research (ICSSR 2014) Application of Nonlinear Later TV Edition in Gigabit Ethernet Hong Ma Education and Training Department, Shaanxi Vocational and

More information

SIPROTEC 5 V7.8 Protection, automation and monitoring for digital substations

SIPROTEC 5 V7.8 Protection, automation and monitoring for digital substations SIPROTEC 5 V7.8 Protection, automation and monitoring for digital substations siemens.com/siprotec5 SIPROTEC 5 Table of content Introduction New functions of V7.8 SIPROTEC 5 - the core of Digital Substation

More information

MAXOUT DX CRANE LOAD LIMITER

MAXOUT DX CRANE LOAD LIMITER MAXOUT DX CRANE LOAD LIMITER Model MX10x, Version 2+ Model MX10xD Installation and User Manual V1: 31/03/2017 CASWA Pty Ltd 2017 CASWA Pty Ltd 2017 1 Page CONTENTS 1 OVERVIEW... 4 2 SPECIFICATIONS... 5

More information

BE1-BPR BREAKER PROTECTION RELAY ADVANTAGES

BE1-BPR BREAKER PROTECTION RELAY ADVANTAGES BE1-BPR BREAKER PROTECTION RELAY The BE1-BPR Breaker Protection Relay provides all the capabilities required for most breaker protection and monitoring functions. It incorporates a simple programming language

More information

Welcome. Hardeep Singh Service and Parts Director Trane Middle East & Africa

Welcome. Hardeep Singh Service and Parts Director Trane Middle East & Africa Welcome Congratulations and welcome to your new Elite Service membership. Packed with benefits to add value to your daily operations at no additional cost, this loyalty-rewards program has been developed

More information

DGSZV-EP DIGITAL GALVANIC LONGITUDINAL DIFFERENTIAL PROTECTION. Application field

DGSZV-EP DIGITAL GALVANIC LONGITUDINAL DIFFERENTIAL PROTECTION. Application field DGSZV-EP DIGITAL GALVANIC LONGITUDINAL DIFFERENTIAL PROTECTION The digital galvanic longitudinal differential protection of type DGSZV-EP is part of device family named EuroProt. This short description

More information

Improving Trajectory Tracking Performance of Robotic Manipulator Using Neural Online Torque Compensator

Improving Trajectory Tracking Performance of Robotic Manipulator Using Neural Online Torque Compensator JOURNAL OF ENGINEERING RESEARCH AND TECHNOLOGY, VOLUME 1, ISSUE 2, JUNE 2014 Improving Trajectory Tracking Performance of Robotic Manipulator Using Neural Online Torque Compensator Mahmoud M. Al Ashi 1,

More information

The use of Neural Networks in Text Categorization. Abhishek Sethi

The use of Neural Networks in Text Categorization. Abhishek Sethi The use of Neural Networks in Text Categorization Abhishek Sethi What is a Neural Network An attempt at creating a learning system that mimics the brain Brain consists of billions of neurons, interconnected,

More information

LCI User Manual mantracourt.com

LCI User Manual mantracourt.com LCI User Manual mantracourt.com LCI Load Cell Junction Box with Fault Monitor Contents Chapter 1 Introduction to the LCI... 2 Chapter 2 Installing the LCI... 3 Chapter 3 Setting up the LCI... 4 Sequence

More information

Switchgear Design Impacts the Reliability of Backup Power Systems

Switchgear Design Impacts the Reliability of Backup Power Systems Switchgear Design Impacts the Reliability of Backup Power Systems December 2010/AT307 by Reza Tajali, P.E. Square D Power Systems Engineering Make the most of your energy SM Revision #1 12/10 When a cascaded

More information

Channel Performance Improvement through FF and RBF Neural Network based Equalization

Channel Performance Improvement through FF and RBF Neural Network based Equalization Channel Performance Improvement through FF and RBF Neural Network based Equalization Manish Mahajan 1, Deepak Pancholi 2, A.C. Tiwari 3 Research Scholar 1, Asst. Professor 2, Professor 3 Lakshmi Narain

More information

Power System Network Simulator (PSNsim)

Power System Network Simulator (PSNsim) The PSNsim is hardware simulator that simulates all parts of power system from generation to utilization. The PSNsim is a scaled down model of the real power system that provides an experimental environment

More information

LECTURE NOTES Professor Anita Wasilewska NEURAL NETWORKS

LECTURE NOTES Professor Anita Wasilewska NEURAL NETWORKS LECTURE NOTES Professor Anita Wasilewska NEURAL NETWORKS Neural Networks Classifier Introduction INPUT: classification data, i.e. it contains an classification (class) attribute. WE also say that the class

More information