Artificial Neural Networks - Industrial and Control Engineering Applications

ARTIFICIAL NEURAL NETWORKS INDUSTRIAL AND CONTROL ENGINEERING APPLICATIONS

Kenji Suzuki

Sign up for access to the world's latest research

checkGet notified about relevant papers

checkSave papers to use in your research

checkJoin the discussion with peers

checkTrack your impact

Abstract

Artificial neural networks may probably be the single most successful technology in the last two decades which has been widely used in a large variety of applications. The purpose of this book is to provide recent advances of artificial neural networks in industrial and control engineering applications. The book begins with a review of applications of artificial neural networks in textile industries. Particular applications in textile industries follow. Parts continue with applications in materials science and industry such as material identification, and estimation of material property and state, food industry such as meat, electric and power industry such as batteries and power systems, mechanical engineering such as engines and machines, and control and robotic engineering such as system control and identification, fault diagnosis systems, and robot manipulation. Thus, this book will be a fundamental source of recent advances and applications of artificial neural networks in industrial and control engineering areas. The target audience includes professors and students in engineering schools, and researchers and engineers in industries.

Figures (404)

Table 1. Properties of jute, polypropylene and polyester fibres

Artificial Neural Networks - Industrial and Control Engineering Applications

tig. 1. Neural architecture of the fabric property

Table 3. Fabric samples for development of Emperical and ANN models

Modelling of Needle-Punched Nonwoven Fabric Properties Using Artificial Neural Network

Modelling of Needle-Punched Nonwoven Fabric Properties Using Artificial Neural Network
Table 7. Experimental and predicted initial modulus values by empirical and ANN models — Modelling of Needle-Punched Nonwoven Fabric Properties Using Artificial Neural Network Table 7. Experimental and predicted initial modulus values by empirical and ANN models

Table 8. Experimental verification of predicted results (tenacity and initial modulus)

Artificial Neural Networks - Industrial and Control Engineering Applications
Table 8. Experimental verification of predicted results (tenacity and initial modulus) — Artificial Neural Networks - Industrial and Control Engineering Applications Table 8. Experimental verification of predicted results (tenacity and initial modulus)

Table 11. Experimental verification of predicted results of air permeability values

Artificial Neural Networks - Industrial and Control Engineering Applications
Table 11. Experimental verification of predicted results of air permeability values — Artificial Neural Networks - Industrial and Control Engineering Applications Table 11. Experimental verification of predicted results of air permeability values

Fig. 1. Experimental configuration of a LIBS system.

Fig. 3. Examples of LIBS spectra for materials with similar composition.

Fig. 4. Basic structure of an artificial neural network.

Artificial Neural Networks for Material Identification, Mineralogy and
Analytical Geochemistry Based on Laser-Induced Breakdown Spectroscop
Fig. 7. Natural rock & mineral samples and their powder tablets counterparts. — Artificial Neural Networks for Material Identification, Mineralogy and Analytical Geochemistry Based on Laser-Induced Breakdown Spectroscop Fig. 7. Natural rock & mineral samples and their powder tablets counterparts.

Fig. 7. Natural rock & mineral samples and their powder tablets counterparts.

more detailed quantitative analysis, possibly, in conjunction with data from other sensors.

Artificial Neural Networks - Industrial and Control Engineering Applications
Fig. 15. A comparison of the validation performance between a typical ANN with
conventional training (a) and the ANN with constructive training (b). — Artificial Neural Networks - Industrial and Control Engineering Applications Fig. 15. A comparison of the validation performance between a typical ANN with conventional training (a) and the ANN with constructive training (b).

For the calibration curve method, the deviation 1s mainly due to the serious matrix effects of

\rtificial Neural Networks for Material Identification, Mineralogy and
\nalytical Geochemistry Based on Laser-Induced Breakdown Spectroscopy — \rtificial Neural Networks for Material Identification, Mineralogy and \nalytical Geochemistry Based on Laser-Induced Breakdown Spectroscopy

Fig. 6. Regression analysis of of approximated K' based on Ea. 7 for test data.

Artificial Neural Networks - Industrial and Control Engineering Applications
Table 1. The compositions and mechanical properties of ZrO2/TiBz/ Al,O3 ceramic material — Artificial Neural Networks - Industrial and Control Engineering Applications Table 1. The compositions and mechanical properties of ZrO2/TiBz/ Al,O3 ceramic material

Table 1. The compositions and mechanical properties of ZrO2/TiBz/ Al,O3 ceramic material

Table 4. The predicted results of standard BP algorithm in compositions optimization

Artificial Neural Networks - Industrial and Control Engineering Applications
Table 4. The predicted results of standard BP algorithm in compositions optimization — Artificial Neural Networks - Industrial and Control Engineering Applications Table 4. The predicted results of standard BP algorithm in compositions optimization

Table 5. The predicted results of GA-BP II algorithm in compositions optimization

Fig. 3. The curve of square error and fitness of GA-BP II in compositions optimization

Artificial Neural Networks - Industrial and Control Engineering Applications
Fig. 3. The curve of square error and fitness of GA-BP II in compositions optimization — Artificial Neural Networks - Industrial and Control Engineering Applications Fig. 3. The curve of square error and fitness of GA-BP II in compositions optimization

idicates that the BP neural network has 8 iterations convergence to the specified accuracy.
he elapsed-time is 129.939s and MSE is 0.1491. — idicates that the BP neural network has 8 iterations convergence to the specified accuracy. he elapsed-time is 129.939s and MSE is 0.1491.

Optimum Design and Application of Nano-Micro-Composite Ceramic Tool and
Die Materials with Improved Back Propagation Neural Network — Optimum Design and Application of Nano-Micro-Composite Ceramic Tool and Die Materials with Improved Back Propagation Neural Network

Artificial Neural Networks - Industrial and Control Engineering Applications
Fig. 3. Schematic architecture of Artificial Neural Network model. — Artificial Neural Networks - Industrial and Control Engineering Applications Fig. 3. Schematic architecture of Artificial Neural Network model.

Fig. 4. One sample of data, microstructure of produced steel with ASTM grain size no. 9.0.

Application of Bayesian Neural Networks to
Predict Strength and Grain Size of Hot Strip Low Carbon Steel:
Table 1. Input and output parameter information — Application of Bayesian Neural Networks to Predict Strength and Grain Size of Hot Strip Low Carbon Steel: Table 1. Input and output parameter information

Table 1. Input and output parameter information

Fig. 5. Behavior of tensile strength model on (a) training data (b) test data.

Adaptive Neuro-Fuzzy Inference System Prediction
of Calorific Value Based on the Analysis of U.S. Coal:
Table 1. Number of samples and range of GCV (as-received) for different U.S. states — Adaptive Neuro-Fuzzy Inference System Prediction of Calorific Value Based on the Analysis of U.S. Coal: Table 1. Number of samples and range of GCV (as-received) for different U.S. states

Table 1. Number of samples and range of GCV (as-received) for different U.S. states
Table 2. Ranges of proximate and ultimate analyses of coal samples (as-received) — Table 1. Number of samples and range of GCV (as-received) for different U.S. states Table 2. Ranges of proximate and ultimate analyses of coal samples (as-received)

Estimated deviations of GCV from target values for equations (4) through (6) are shown il
Table 3. — Estimated deviations of GCV from target values for equations (4) through (6) are shown il Table 3.

Adaptive Neuro-Fuzzy Inference System Prediction
of Calorific Value Based on the Analysis of U.S. Coals — Adaptive Neuro-Fuzzy Inference System Prediction of Calorific Value Based on the Analysis of U.S. Coals

Table 6. Estimation of deviations of GCV from target values for neuro-fuzzy models

Artificial Neural Network Applied for Detecting the
Saturation Level in the Magnetic Core of a Welding Transforme — Artificial Neural Network Applied for Detecting the Saturation Level in the Magnetic Core of a Welding Transforme

Table 1. Parameters of RSWS dynamic model

Artificial Neural Network Applied for Detecting the
Saturation Level in the Magnetic Core of a Welding Transforme
Table 2. Nonlinear characteristics of the output rectifier diodes Dj - (dip1) and D2 - (dip2) — Artificial Neural Network Applied for Detecting the Saturation Level in the Magnetic Core of a Welding Transforme Table 2. Nonlinear characteristics of the output rectifier diodes Dj - (dip1) and D2 - (dip2)

Table 3. Parameters for symmetrical behaviour of the resistance spot welding system

Fig. 3. Symmetrical behaviour of the resistance spot welding system

Artificial Neural Networks - Industrial and Control Engineering Applications
Fig. 3. Symmetrical behaviour of the resistance spot welding system — Artificial Neural Networks - Industrial and Control Engineering Applications Fig. 3. Symmetrical behaviour of the resistance spot welding system

Fig. 6. The dynamic inductance and maenetically nonlinear iron core characteristic

Fig, 3. Experimental and predicted values for recovery of fiber from DSF

Table 2. Range of variables selected for study of cellulase & xylanase production

Table 3. ANN based _ predicted combinations for optimized production of enzymes by T. reesei
Table 4. ANN based predicted combinations for optimized roduction of enzymes by A.niger, — Table 3. ANN based _ predicted combinations for optimized production of enzymes by T. reesei Table 4. ANN based predicted combinations for optimized roduction of enzymes by A.niger,

Table 3. ANN based _ predicted combinations for optimized production of enzymes by T. reesei

Application of Artificial Neural Networks to Food and Fermentation Technology

Radial basis function networks (RBF networks) represent a special type of ANN, which are
closely related to density estimation methods. A thorough mathematical description of RBF
Fig. 1. The structure of the a) Kohonen ANN, b) CP-ANN, c) BP-ANN and d) RBF network — Radial basis function networks (RBF networks) represent a special type of ANN, which are closely related to density estimation methods. A thorough mathematical description of RBF Fig. 1. The structure of the a) Kohonen ANN, b) CP-ANN, c) BP-ANN and d) RBF network

Application of Artificial Neural Networks in Meat Production and Technology

Table 4. Other applications of ANN in meat science and technology

State of Charge Estimation of Ni-MH battery pack by using ANN

Fig. 6. The output result waveform of the checking sample is input.
Table 1. Different network’s average erro! — Fig. 6. The output result waveform of the checking sample is input. Table 1. Different network’s average erro!

State of Charge Estimation of Ni-MH battery pack by using ANN
Fig. 6. The output result waveform of the checking sample is input. — State of Charge Estimation of Ni-MH battery pack by using ANN Fig. 6. The output result waveform of the checking sample is input.

Fig. 8. Actual structure of neural network

‘ia, 9. The forecasting result of ANN as set the different SOC initial value

Figure 1 depicts the ADALINE structure to track fundamental frequency which is a
proposed in this section.
\ssume that the voltage waveform of power system comprises unknown fundamental,
larmonics and decaying DC offset components as: — Figure 1 depicts the ADALINE structure to track fundamental frequency which is a proposed in this section. \ssume that the voltage waveform of power system comprises unknown fundamental, larmonics and decaying DC offset components as:

Artificial Neural Networks - Industrial and Control Engineering Applications
Fig. 2. Impulse response of the Hamming window with 20 Hz cut frequency — Artificial Neural Networks - Industrial and Control Engineering Applications Fig. 2. Impulse response of the Hamming window with 20 Hz cut frequency

The weight matrix W in this case has only one row. The network output is

have estimation with less than one percent error 1s represented in lable I.
Fig. 8. Average convergence time (cycles) to track static frequency changes — have estimation with less than one percent error 1s represented in lable I. Fig. 8. Average convergence time (cycles) to track static frequency changes

Table I Samples needed to estimate with 1 percent error for 50-70 frequency range

Fig. 13. Tracked frequency (Hz), case V.C.

Fig. 15. Complex normalized rotating state vector (An, ), case V.C.

Main information on the Marvdasht 230 /66 (kV) station and other substation supplied by
this station is given in Tables II and III.
Table Il Marvdasht substation capacities — Main information on the Marvdasht 230 /66 (kV) station and other substation supplied by this station is given in Tables II and III. Table Il Marvdasht substation capacities

Table Il Marvdasht substation capacities

Application of ANN to Real and Reactive Power Allocation Scheme
Fig. 1. Flow chart of the proposed real and reactive power allocation method — Application of ANN to Real and Reactive Power Allocation Scheme Fig. 1. Flow chart of the proposed real and reactive power allocation method

Table 1. Description of inputs and outputs of the training data for each ANN for real power

Application of ANN to Real and Reactive Power Allocation Scheme

Table 3. The number of samples for training, validation and test set

Fig. 7. Real power allocation method daily load curves for different buses
Fic. 8. Reactive power allocation method daily load curves for different buses — Fig. 7. Real power allocation method daily load curves for different buses Fic. 8. Reactive power allocation method daily load curves for different buses

Fic. 8. Reactive power allocation method daily load curves for different buses

Artificial Neural Networks - Industrial and Control Engineering Applications
Fig. 7. Real power allocation method daily load curves for different buses — Artificial Neural Networks - Industrial and Control Engineering Applications Fig. 7. Real power allocation method daily load curves for different buses

function of individual generators voltage.

Fig. 14. Distribution of reactive power from generator at bus 14 to loads within 168 hours
Table 7. Analysis of reactive power allocation for the 25-bus equivalent system — Fig. 14. Distribution of reactive power from generator at bus 14 to loads within 168 hours Table 7. Analysis of reactive power allocation for the 25-bus equivalent system

Application of ANN to Real and Reactive Power Allocation Scheme
Fig. 14. Distribution of reactive power from generator at bus 14 to loads within 168 hours — Application of ANN to Real and Reactive Power Allocation Scheme Fig. 14. Distribution of reactive power from generator at bus 14 to loads within 168 hours

Fig. 19. Distribution of reactive power from generator at bus 69 to loads within 24 hours
Table 12. Analysis of reactive power allocation for selected generators in the IEEE 118 bu
system — Fig. 19. Distribution of reactive power from generator at bus 69 to loads within 24 hours Table 12. Analysis of reactive power allocation for selected generators in the IEEE 118 bu system

Fig. 19. Distribution of reactive power from generator at bus 69 to loads within 24 hours

Artificial Neural Networks - Industrial and Control Engineering Applications
Table 13. Comparative computational time for MNE, ANN, and RBFEN methods for different
bus system — Artificial Neural Networks - Industrial and Control Engineering Applications Table 13. Comparative computational time for MNE, ANN, and RBFEN methods for different bus system

Artificial Neural Networks - Industrial and Control Engineering Applications
he NLARX model training can be cast as a non-linear unconstrained optimization problem: — Artificial Neural Networks - Industrial and Control Engineering Applications he NLARX model training can be cast as a non-linear unconstrained optimization problem:

Fie. 3. Data set I- NO, emission output generated in NRTC mode

Fig. 5. Pre-processed NO, output signal. Rearranged and composed training and validation set

Fig. 5. Pre-processed NO, output signal. Rearranged and composed training and validation set
‘ig. 6. Data set II training cycle of NO, target outpt — Fig. 5. Pre-processed NO, output signal. Rearranged and composed training and validation set ‘ig. 6. Data set II training cycle of NO, target outpt

Fig. 7. Correlation of measured NO, output with predicted neural network signal

tie. 8. Correlation between measured target output and predicted output with an R2 = 0.95

Fig. 10. Random signal of SOI for training and validation

Table 1. Results for NLARX models for random signal training
Fig. 12. Correlation of engine exhaust temperature with predicted neural network signal — Table 1. Results for NLARX models for random signal training Fig. 12. Correlation of engine exhaust temperature with predicted neural network signal

Fig. 13. Correlation of engine compressor mass-air flow with predicted neural network signal

Fig. 13. Correlation of engine compressor mass-air flow with predicted neural network signal
Fig, 14. Correlation of engine exhaust pressure with predicted neural network signal — Fig. 13. Correlation of engine compressor mass-air flow with predicted neural network signal Fig, 14. Correlation of engine exhaust pressure with predicted neural network signal

Artificial Neural Networks - Industrial and Control Engineering Applications
Fig. 15. Correlation of engine NOx with predicted neural network signal — Artificial Neural Networks - Industrial and Control Engineering Applications Fig. 15. Correlation of engine NOx with predicted neural network signal

Fig. 16. Schematic drawing of a diesel engine and auxiliaries tig. 17. Valve-Lift profile for inlet and exhaust valve

‘ia, 18. Correlation of engine intake manifold pressure with predicted neural network signal

The Applications of Artificial Neural Networks to Engines
Fig. 21. Single NLARX model: measured output signal correlated versus predicted output
signal — The Applications of Artificial Neural Networks to Engines Fig. 21. Single NLARX model: measured output signal correlated versus predicted output signal

The Applications of Artificial Neural Networks to Engines

Artificial Neural Networks - Industrial and Control Engineering Applications
Fig. 1a. Structure of SFF model
Fig. 1b. Structure of ANN model — Artificial Neural Networks - Industrial and Control Engineering Applications Fig. 1a. Structure of SFF model Fig. 1b. Structure of ANN model

Table 1. Fuzzy expressions for inputs and outputs

Fig. 4. Soeed membership function for HSS tool

Fig. 3. Depth of cut membership function
Fig. 4. Soeed membership function for HSS tool — Fig. 3. Depth of cut membership function Fig. 4. Soeed membership function for HSS tool

Artificial Neural Networks - Industrial and Control Engineering Applications
tig, 2. Hardness membership functior — Artificial Neural Networks - Industrial and Control Engineering Applications tig, 2. Hardness membership functior

Fig. 7. Feed membership function (BHN=175-275, HSS)

Fig. 6. Feed membership function (BHN=85-175,HSS)
Fig. 7. Feed membership function (BHN=175-275, HSS) — Fig. 6. Feed membership function (BHN=85-175,HSS) Fig. 7. Feed membership function (BHN=175-275, HSS)

Fig. 6. Feed membership function (BHN=85-175,HSS)

Rule 36: IF hardness is very hard AND depth of cut is very deep THEN speed is very slow and very fast.

Fig. 12. User-friendly viewer for the SFF model (from MATLAB)

A Comparison of Speed-Feed Fuzzy Intelligent
System and ANN for Machinability Data Selection of CNC Machine:
Table 5. Comparison of the results from SFF model with MDH for carbide too. — A Comparison of Speed-Feed Fuzzy Intelligent System and ANN for Machinability Data Selection of CNC Machine: Table 5. Comparison of the results from SFF model with MDH for carbide too.

Fig. 13. Cutting speed for high speed steel

Fig. 13. Cutting speed for high speed steel
Fig, 14. Feed for high speed steel — Fig. 13. Cutting speed for high speed steel Fig, 14. Feed for high speed steel

A Comparison of Speed-Feed Fuzzy Intelligent
System and ANN for Machinability Data Selection of CNC Machine — A Comparison of Speed-Feed Fuzzy Intelligent System and ANN for Machinability Data Selection of CNC Machine

Fig. 17. Performance curves for best tested ANN models

Artificial Neural Networks - Industrial and Control Engineering Applications
Table 11. Comparison of outputs for HSS tool — Artificial Neural Networks - Industrial and Control Engineering Applications Table 11. Comparison of outputs for HSS tool

Table 11. Comparison of outputs for HSS tool

Fig. 20. Comparison of speed values between SFF, ANN and MDH
Fig, 21. Comparison of feed values between SFF, ANN and MDH — Fig. 20. Comparison of speed values between SFF, ANN and MDH Fig, 21. Comparison of feed values between SFF, ANN and MDH

Artificial Neural Networks - Industrial and Control Engineering Applications
Fig. 20. Comparison of speed values between SFF, ANN and MDH — Artificial Neural Networks - Industrial and Control Engineering Applications Fig. 20. Comparison of speed values between SFF, ANN and MDH

Artificial Neural Networks - Industrial and Control Engineering Applications
Fig. 20. Comparison of speed values between SFF, ANN and MDH
Fig, 21. Comparison of feed values between SFF, ANN and MDH — Artificial Neural Networks - Industrial and Control Engineering Applications Fig. 20. Comparison of speed values between SFF, ANN and MDH Fig, 21. Comparison of feed values between SFF, ANN and MDH

presented in [Rivera et al., 1986]. IMC for nonlinear systems is introduced in [Economou et al.,
1986] and IMC with neural networks is described e.g. in [Norgaard et al., 2000]. — presented in [Rivera et al., 1986]. IMC for nonlinear systems is introduced in [Economou et al., 1986] and IMC with neural networks is described e.g. in [Norgaard et al., 2000].

Artificial Neural Networks - Industrial and Control Engineering Applications
Output prediction y(k+i) is possible to calculate by multiplying the equation (9) by the
inverse matrix A -1, equation (10). — Artificial Neural Networks - Industrial and Control Engineering Applications Output prediction y(k+i) is possible to calculate by multiplying the equation (9) by the inverse matrix A -1, equation (10).

Artificial Neural Network — Possible Approach to Nonlinear System Contro

Control law can be transformed to equation (16).

tig. 4. Controller tuning online using neural network

Artificial Neural Network — Possible Approach to Nonlinear System Contro
Fig. 5. Control performance with PI controller - Training set experiment — Artificial Neural Network — Possible Approach to Nonlinear System Contro Fig. 5. Control performance with PI controller - Training set experiment

Fig. 5. Control performance with PI controller - Training set experiment

Artificial Neural Network — Possible Approach to Nonlinear System Control

Fig. 10. Neural IMC control response
Fig. 11. Classical IMC control response — Fig. 10. Neural IMC control response Fig. 11. Classical IMC control response

Artificial Neural Network — Possible Approach to Nonlinear System Contro
Fig. 10. Neural IMC control response — Artificial Neural Network — Possible Approach to Nonlinear System Contro Fig. 10. Neural IMC control response

Fig. 12. Neural Network Predictive control response
Fig. 13. Piecewise Linearized Model Predictive control response — Fig. 12. Neural Network Predictive control response Fig. 13. Piecewise Linearized Model Predictive control response

Artificial Neural Networks - Industrial and Control Engineering Applications
Fig. 12. Neural Network Predictive control response — Artificial Neural Networks - Industrial and Control Engineering Applications Fig. 12. Neural Network Predictive control response

Artificial Neural Network — Possible Approach to Nonlinear System Contro
Fig. 14. Discrete controller tuned online — Artificial Neural Network — Possible Approach to Nonlinear System Contro Fig. 14. Discrete controller tuned online

Direct Neural Network Control via Inverse Modelling: Application on Induction Motors

Fig. 12. Speed error due to a step reference signal
tig. 13. System response to a square wave reference signa — Fig. 12. Speed error due to a step reference signal tig. 13. System response to a square wave reference signa

Fig. 12. Speed error due to a step reference signal

Fig. 14. System response to a sine wave reference signal

Fig. 16. Speed response under load torque conditior

System Identification of NN-based Model Reference Control of RUAV during Hover
Fig. 2. UAV Architecture: Hardware-in-the-loop Simulation — System Identification of NN-based Model Reference Control of RUAV during Hover Fig. 2. UAV Architecture: Hardware-in-the-loop Simulation

Artificial Neural Networks - Industrial and Control Engineering Applications
Fig. 5. Flow of Control System Design with System Identification — Artificial Neural Networks - Industrial and Control Engineering Applications Fig. 5. Flow of Control System Design with System Identification

Fig. 5. Flow of Control System Design with System Identification

Fig. 7. Control of a MIMO plant using NN controller

System Identification of NN-based Model Reference Control of RUAV during Hover

Artificial Neural Networks - Industrial and Control Engineering Applications
Table 1. Model states. — Artificial Neural Networks - Industrial and Control Engineering Applications Table 1. Model states.

System Identification of NN-based Model Reference Control of RUAV during Hove!
The descriptions of all parameters used in Eq. 19-20 are shown in Table 3 — System Identification of NN-based Model Reference Control of RUAV during Hove! The descriptions of all parameters used in Eq. 19-20 are shown in Table 3

The descriptions of all parameters used in Eq. 19-20 are shown in Table 3

Table 4. Eignevalues of the helicopter system

System Identification of NN-based Model Reference Control of RUAV during Hover
Fig. 13. Output response with network response in Longitudinal dynamics mode — System Identification of NN-based Model Reference Control of RUAV during Hover Fig. 13. Output response with network response in Longitudinal dynamics mode

ANG O;,;, Ong OUt the Gependency rate 1s very less.

System Identification of NN-based Model Reference Control of RUAV during Hover
The histogram of prediction error is shown in Fig. 18. — System Identification of NN-based Model Reference Control of RUAV during Hover The histogram of prediction error is shown in Fig. 18.

Artificial Neural Networks - Industrial and Control Engineering Applications
Fig, 18. Histogram of Prediction errors in Longitudinal Mode — Artificial Neural Networks - Industrial and Control Engineering Applications Fig, 18. Histogram of Prediction errors in Longitudinal Mode

Intelligent Vibration Signal Diagnostic System Using Artificial Neural Network

Intelligent Vibration Signal Diagnostic System Using Artificial Neural Network
Table 1. Test ball bearings — Intelligent Vibration Signal Diagnostic System Using Artificial Neural Network Table 1. Test ball bearings

Fig. 8. AR parameters patterns of defective bearings

(Adapted from Carpenter and Grossberg, 1987b

Conditioning Monitoring and Fault Diagnosis for a
Servo-Pneumatic System with Artificial Neural Network Algorithn
Fig. 5. Gripper Place Sensor — Conditioning Monitoring and Fault Diagnosis for a Servo-Pneumatic System with Artificial Neural Network Algorithn Fig. 5. Gripper Place Sensor

Fig. 10. The FFN type ANN estimations for the training cases

Fig. 11. The FFN type ANN estimations for the test cases.

Conditioning Monitoring and Fault Diagnosis for a
Servo-Pneumatic System with Artificial Neural Network Algorithrr
Table 2. The estimated categories with the ART 2 algorithm
ART2 - Number of Categories at different vigilances — Conditioning Monitoring and Fault Diagnosis for a Servo-Pneumatic System with Artificial Neural Network Algorithrr Table 2. The estimated categories with the ART 2 algorithm ART2 - Number of Categories at different vigilances

Table 2. The estimated categories with the ART 2 algorithm
ART2 - Number of Categories at different vigilances
Fig. 13. The graphical presentation of the ART2 results in the Table 1. — Table 2. The estimated categories with the ART 2 algorithm ART2 - Number of Categories at different vigilances Fig. 13. The graphical presentation of the ART2 results in the Table 1.

Fuzzy ART - Number of Categories at different vigilances
Fig. 13. The estimations of the Fuzzy ART — Fuzzy ART - Number of Categories at different vigilances Fig. 13. The estimations of the Fuzzy ART

Fig. 2. Schematic diagram for the biological neuron

Neural Networks’ Based Inverse Kinematics Solution
for Serial Robot Manipulators Passing Through Singularities
Fig. 2. Schematic diagram for the biological neuron — Neural Networks’ Based Inverse Kinematics Solution for Serial Robot Manipulators Passing Through Singularities Fig. 2. Schematic diagram for the biological neuron

Artificial Neural Networks - Industrial and Control Engineering Applications
Fig, 4. Schematic diagram of a multilayer feedforward neural network — Artificial Neural Networks - Industrial and Control Engineering Applications Fig, 4. Schematic diagram of a multilayer feedforward neural network

Neural Networks’ Based Inverse Kinematics Solution
for Serial Robot Manipulators Passing Through Singularities — Neural Networks’ Based Inverse Kinematics Solution for Serial Robot Manipulators Passing Through Singularities

Fig. 6. Information flow through a backpropagation network

Fig. 8. The learning curve for the first configuration

Artificial Neural Networks - Industrial and Control Engineering Applications
‘ig. 9. The topology of the second configuration network — Artificial Neural Networks - Industrial and Control Engineering Applications ‘ig. 9. The topology of the second configuration network

‘ig. 9. The topology of the second configuration network

Artificial Neural Networks - Industrial and Control Engineering Applications
Fig. 12. Trajectory tracking for both configurations compared to each other after the training
was finished for the Y coordinate — Artificial Neural Networks - Industrial and Control Engineering Applications Fig. 12. Trajectory tracking for both configurations compared to each other after the training was finished for the Y coordinate

Fig. 12. Trajectory tracking for both configurations compared to each other after the training
was finished for the Y coordinate — Fig. 12. Trajectory tracking for both configurations compared to each other after the training was finished for the Y coordinate

Table 3 shows the percentages of error for the testing data set for each joint during
simulation stage.
Table 3. Error percentages obtained for testing data through simulation stage — Table 3 shows the percentages of error for the testing data set for each joint during simulation stage. Table 3. Error percentages obtained for testing data through simulation stage

Artificial Neural Networks - Industrial and Control Engineering Applications
Fig. 14. Predicted trajectory for the Y coordinate — Artificial Neural Networks - Industrial and Control Engineering Applications Fig. 14. Predicted trajectory for the Y coordinate

Fig. 14. Predicted trajectory for the Y coordinate
Fig. 15. Predicted trajectory for the Z coordinate — Fig. 14. Predicted trajectory for the Y coordinate Fig. 15. Predicted trajectory for the Z coordinate

The error percentages in the experimental data are shown in table 4.
Table 4. Error percentages obtained for testing data through experimental stage — The error percentages in the experimental data are shown in table 4. Table 4. Error percentages obtained for testing data through experimental stage

Nirmal Yadav

International Journal of Current Microbiology and Applied Sciences

View PDFchevron_right

ARTIFICIAL NEURAL NETWORKS FOR SOLVING TEXTILE PROBLEMS: REVIEW

Ahmed Samar, Magdi El Messiry

The textile process involves the interaction of a large number of variables. The relations between these variables and the product properties could not be established conclusively. Different techniques have been suggested to determine these relations but with limited success. Artificial Neural Networks (ANN) represent a promising step in this field. In this research, a survey of the application of ANN from fibers through yarn; fabric, garments as well as finishing processes is performed. A comparison between the traditional techniques and ANN in textiles is conducted and a preview for the expected future of ANN in the textile field is also presented.

View PDFchevron_right

Scope for Artificial Neural Network in Textiles

Sentthil Kumar

2015

The artificial neural network (ANN) is increasingly used as a powerful tool for many real world problems. ANN has proved its usefulness for resolving many problems in textiles such as prediction of yarn properties, analysis of fabric defects, process optimization etc. The power of neural networks lies in their ability to represent complex relationships and learn them directly from the data being modeled. The prediction of properties or performance of a process in advance is required to minimize the setup cost and time. The ability to predict these properties accurately has become a challenge due to highly non-linear and interactive behaviour of textile materials. This paper presents basics of ANN and its applications in different textile domains.

View PDFchevron_right

APPLICATION OF NEURAL NETWORKS IN FABRIC ENGINEERING

GOPALAKRISHNAN DURAISAMY

Neural Network is an intelligent system that finds widespread applications in many research and engineering fields. Artificial neural network systems is one of the hopes available to textile industry to integrate the elements such as production, quality, cost, information, statistical process control, just-in-time manufacturing computer integrated manufacturing etc,. Artificial Neural Network (ANN) is a field of computer science that seeks to understand and implement computer based technology that can simulate characteristics of human intelligence (that include learning, adapting, reasoning, self-correction and automatic improvement) and human sensory capabilities. Artificial Neural Networks (ANN) are finding applications in textile industry in areas such as fabric engineering, lay planning in garment industry, market prediction, fashion prediction etc The Basics involved in the understanding of the Neural Networks, their properties, construction, Classification - based on application, advantages of neural computing. Next to the basics, Applications of Neural Networks in the area of fabric engineering is covered extensively. To have a clear-cut idea on exactly how the Neural Network works, a real time expert system (using NN) used in Prediction of Fabric end-uses is presented with the principle involved, Prediction network modeling, Training the network, the logic behind the prediction technique etc. Then, the application list extends and covers the use of NN in detecting the fabric faults, grading the color fastness of the fabric, Measurement / Prediction of comfort properties of the fabric.

View PDFchevron_right

NEURAL NETWORKS IN TEXTILE ENGINEERING

Milin Patel

Neural network is a computational structure. A neural network is a massively parallel distributed processor made up of simple processing units, which has a neural propensity for storing experiential knowledge and making it available for use. It has capability to organize its structural constituents, known as “Neurons”. It can perform certain computations (e.g. fabric pattern, fabric defects, fabric wrinkle, etc.) many times faster than the fastest computer. Artificial Neural Network (ANN) is a field of computer science that seeks to understand and implement computer based technology. The Neural Network is understood on the basis of their properties, construction, classification - based on application, advantages of neural computing. Fabric engineering needs a thorough understanding of the functional properties and their construction parameters. When the relationship between a set of interrelated properties goes beyond the complete comprehension of the human brain, neural networks (NNs) could be used to find the unknown function. Neural network describes the method, for the prediction of both construction and performance parameters of fabrics. Accordingly, constructional parameters are used as input for predicting the performance parameter in forward engineering, and the parameters are reversed for the reverse engineering prediction. Comparison between actual results and predicted results is made. An expert system with an embedded artificial neural network (ANN) is also used, with its functionality toward engineered fabric manufacturing.

View PDFchevron_right

Neural networks and its application in engineering

Olawale Jegede

… Science & IT …, 2009

Neural Network (NN) has emerged over the years and has made remarkable contribution to the advancement of various fields of endeavor. The purpose of this work is to examine neural networks and their emerging applications in the field of engineering, focusing more on Controls. In this work, we have examined the various architectures of NN and the learning process. The needs for neural networks, training of neural networks, and important algorithms used in realizing neural networks have also been briefly discussed. Neural network application in control engineering has been extensively discussed, whereas its applications in electrical, civil and agricultural engineering were also examined. We concluded by identifying limitations, recent advances and promising future research directions.

View PDFchevron_right

Theory and applications of neural networks for industrial control systems

Toshio Fukuda

IEEE Transactions on Industrial Electronics, 1992

Abstruct-This paper describes the theory and the applications of artificial neural networks, especially in a control field. Artificial neural networks try to mimic the nerve system in a mammalian brain into a mathematical model. Therefore, neural networks have some desirable characteristics and capabilities similar to the brain system, such as parallel processing, learning, nonlinear mapping, and generalization. Recently, many researchers have developed neural networks as new tools in many fields such as pattern recognition, information processing, design, planning, diagnosis, and control. We survey hybrid systems of the neural networks, fuzzy sets, and Artificial Intelligence (AI) technologies. Fuzzy sets and technologies have been also implemented as new tools in many fields and shown to be useful. Therefore, we deal with the hybrid systems as key technologies in the future.

View PDFchevron_right

Application of Artificial Neural Network into Manufacturing Processes

Ernest Mnkandla

Indonesian Journal of Computer Science, 2022

The neural network model is an advanced and effective tool aims at simulating the manufacturing operations. An important number of researchers have utilized artificial neural network (ANN) to optimising multiple response metrics in manufacturing applications. In the majority of situations, the use of ANN enables the prediction of the mechanical and physical properties of manufacturing goods based on provided technical data. To this end, the deployment of ANN in manufacturing sector is tremendously significant in terms of cost and material resource savings. Thus, Artificial neural network as a key component regarding the optimization of the manufacturing processes.

View PDFchevron_right

A comprehensive review for industrial applicability of artificial neural networks

Marcelo Simoes

2003

This paper presents a comprehensive review of the industrial applications of artificial neural networks (ANNs), in the last 12 years. Common questions that arise to practitioners and control engineers while deciding how to use NNs for specific industrial tasks are answered. Workable issues regarding implementation details, training and performance evaluation of such algorithms are also discussed, based on a judiciously chronological organization of topologies and training methods effectively used in the past years. The most popular ANN topologies and training methods are listed and briefly discussed, as a reference to the application engineer. Finally, ANN industrial applications are grouped and tabulated by their main functions and what they actually performed on the referenced papers. The authors prepared this paper bearing in mind that an organized and normalized review would be suitable to help industrial managing and operational personnel decide which kind of ANN topology and training method would be adequate for their specific problems.

View PDFchevron_right

Neural networks applications in manufacturing processes

Godwin Udo

Computers & Industrial Engineering, 1992

Advanced manufacturing systems are becoming too complex and dynamic for traditional rule-based decision support systems. Neural network technology application is gathering momentum in all aspects of manufacturing processes. This paper overviews the technology and surveys the literature on its application in manufacturing processes.

View PDFchevron_right

A SURVEY OF ARTIFICIAL NEURAL NETWORKS APPLICATIONS IN WEAR AND MANUFACTURING PROCESSES

Minodora Ripa

2000

The potential of the artificial neural networks in the field of wear and manufacturing processes is presented in this paper. Their properties of learning and nonlinear behavior make them useful to model complex nonlinear processes, better than the analytical methods. The neural structures, considered appropriate for such models, are presented. The applications found in the referenced papers mainly consist of prediction and classification. They present some common points, specific to the field: wear processes and particles, manufacturing processes, friction parameters, faults in mechanical structures. The results obtained by the quoted authors, in their interdisciplinary research are described, proving that neural networks are an useful tool during the design stage as well as the running or operation stage.

View PDFchevron_right

A New Classification Method for Predicting the Output of Dye Process in Textile Industry by Using Artificial Neural Networks

Sencer Sultanoglu

Advances in Computational Intelligence

In this paper, a new approach is proposed which predicts the output of the dyeing process in textile industry by using input data consisting of the alarms and/or the interventions during the process. Back-propagation algorithms and radial basis functions are utilized to form the neural networks in predicting whether the dye process is carried out correctly or not before an operator checks it manually. Industrial data are used to test the efficiency of the proposed concept which demonstrates that the success rate is over 85%. Keywords: Artificial Neural Network (ANN) • Dye process • Textile industry • Levenberg-Marquardt (LM) • General regression Neural Network (GRNN) • Gradient Descent with an Adaptive Learning Rate (GDA)

View PDFchevron_right

Application of Neural Networks for Assisting, Controlling and Designing Technological Processes

Krzysztof Żaba

Proceedings of 10th World Congress on Computational Mechanics, 2014

The methodology and examples of utilising neural networks for assisting, controlling and designing technological processes is presented in the paper. Examples concern the continuous melting, casting and rolling of aluminium and alloy rod intended for drawing for electrical wires. Neural networks allow to build dependencies between the performance parameter set, chemical composition of the processed material, and product properties. These functions are used for the process control and for determining the optimal work point of the technological line. The initial data constitute the analytical results of the metal chemical composition and process data collected by the specially built system of canvassing realisation parameters of all unit processes. The effectiveness and efficiency of the system was assessed.

View PDFchevron_right

A Research on Estimation of the Weave Fabric Properties with the Artificial Neural Networks

Erkan Türker

Tekstil Ve Konfeksiyon, 2017

In this study, a feedforward backpropagation artificial neural network (ANN) software was developed and it was tested with two different fabric group. While the former was consisted of 29 cotton samples produced under different conditions, the latter was consisted of 49 polyester samples produced under the same conditions. Each group is divided into training and control groups. The result values of the control groups produced by the trained neural network were obtained. The results of the linear regression for both groups with the same method were also obtained. While the fabric weight, thickness, warp and weft tensile strength for the cotton samples are being examined, the air and water permeability values for the polyester samples were examined. Artificial neural network values for all samples showed better fit than linear regression values. ANN and linear regression results were produced closer results for the polyester samples produced under controlled conditions. While ANN results on cotton samples produced under uncontrolled conditions showed good agreement, the fit of the linear regression results deteriorated. This study was supported by the BAP unit of Uşak University.

View PDFchevron_right

Experimental applications of artificial neural networks in Engineering Processing System

Samad Dadvandipour

2014

Artificial Neural Networks along with Image Processing Systems have proven to be successful, particularly in the domains of mathematics, science and technology. They have gained quite notable advantages beyond classical learning, as their usable engagement are observable in many fields of scientific environment related to the relevant systems. This paper presents a model for identifying the small components parts. The model may be significant in various industries mainly in engineering processing system areas. The objective of the study is to apply Artificial Neural Networks (ANN) in Image Processing System (IPS) with feed forward structure to detect, and recognize different parts or any other environment products on a moving conveyor bel. In the proposed model, we have used appropriate method of edge detection. The edge detection realizes artificial neural network with noise. The paper emphasizes the implementation of the model considering functionality, parts images, accurate dete...

View PDFchevron_right

Defects Identification in Textile by Means of Artificial Neural Networks

Giuseppe Mastronardi

Lecture Notes in Computer Science, 2008

In this paper we use a neural network approach for defects identification in textile. The images analyzed came from an artificial vision system that we used to acquire and memorize them in bitmap file format. The vision system is made of two grey scale line scan camera arrays and each array is composed of four CCD cameras with a sensor of 2048 pixels. Every single camera has a field of view of 600mm. The big amount of pixels to be studied to determine whether the texture is defective or not, requires the implementation of some encoding technique to reduce the number of the significant elements. The artificial neural networks (ANN) are manipulated to compress a bitmap that may contain several defects in order to represent it with a number of coefficients that is smaller than the total number of pixel but still enough to identify all kinds of defects classified. An error back propagation algorithm is also used to train the neural network. The proposed technique includes, also, steps to break down large images into smaller windows or array and eliminate redundant information.

View PDFchevron_right

Artificial Neural Network System for Prediction of Dimensional Properties of Cloth in Garment Manufacturing: Case Study on a T-Shirt

Mihriban Kalkancı

Fibres and Textiles in Eastern Europe, 2017

The purpose of the present study was to estimate dimensional measure properties of T-shirts made up of Single Jersey and interlock fabrics through Artificial Neural Networks (ANN). To that end, 72 different types of t-shirts were manufactured under 2 different fabric groups, each was consisting of 2 groups: one with elastane and the other without. Each of these groups were manufactured from six different materials in three different densities through two different knitting techniques of single jersey and interlock. For estimation of dimensional changes in these T-shirts, models including feed-forward, back-propagated, the momentum learning rule and sigmoid transfer function were utilized. As a result of the present study, the ANN system was found to be successful in estimation of pattern measures of garments. The prediction of dimensional properties produced by the neural network model proved to be highly reliable (R2> 0.99).

View PDFchevron_right

Use of input selection techniques to improve the performance of an artificial neural network during the prediction of yarn quality properties

Josphat Igadwa

Journal of Applied Polymer Science, 2008

The performance of an artificial neural network (ANN) is affected by the number and types of inputs. The aim of this article is to study the performance of ANN algorithms, used for the prediction of cotton yarn strength, elongation, and evenness, as the input units are subtracted (skeletonized) and added to the input layer. Nineteen factors, consisting of fiber properties, processing parameters, and yarn quality properties, were used as the main source of inputs. The initial sets of inputs, which were selected on the basis of their relationship with the output factors, were 13, 13, and 12 for yarn strength, elongation, and evenness, respectively. The final sets of inputs were 14 factors for the three yarn quality properties being predicted, and the new ANN algorithms showed performance improvement of 40, 37, and 47% for strength, elongation, and evenness, respectively, when compared to the algorithms with 19 factors. Yarn twist, fiber length, and fiber length uniformity were common among the five most influential factors affecting yarn strength, elongation, and evenness, accounting for 40, 37, and 37% for the prediction of yarn strength, elongation, and evenness, respectively.

View PDFchevron_right

An investigation on yarn engineering using artificial neural networks

Prabal Kumar Majumdar

Journal of the Textile Institute, 2006

Engineering of spun yarns having specific tensile, evenness and hairiness characteristics is a long-cherished dream of spinning technologists. Selection of suitable raw materials at minimum cost and optimisation of process parameters are the two major tasks to be achieved to manufacture engineered yarn. Advent of high-speed fibre-testing machines and development of powerful modelling tools such as artificial neural network (ANN) have provided a great impetus in the yarn engineering research. This article demonstrates the feasibility of yarn engineering by developing a yarn-to-fibre 'reverse' model, using ANN. This approach is entirely different from the prevailing forward models, which predict the properties of final yarn using the fibre properties as inputs. The cost minimisation of cotton fibre mix was ensured by using the classical linear programming approach in combination with ANN. The engineered yarns demonstrated good agreement with the target yarn properties.

View PDFchevron_right

Application of artificial neural network in textile: Prediction of ring spun yarn properties

IOSR Journal of Engineering

In this paper, a neural network approach is used to understand the effects of HVI fibers characteristics and process parameters on yarn tensile properties, CVm% and hairiness. In this approach, four fibers characteristics, twist level and metric number are taken as input variables and the breaking strength and breaking elongation, CVm and hairiness as output variables. The technique of Bayesian regularization is used for improving the generalization capability of neural networks. For this purpose, five different blends of cotton were processed and spun into ring yarns having different counts. Each count was spun at five twist levels.

View PDFchevron_right

Loading Preview

Sorry, preview is currently unavailable. You can download the paper by clicking the button above.

Log In

ARTIFICIAL NEURAL NETWORKS INDUSTRIAL AND CONTROL ENGINEERING APPLICATIONS

Sign up for access to the world's latest research

Abstract

Related papers