Free Air Ball Forming Process Parameter Analysis for Wire Bonder

Manufacturing ◽  
2002 ◽  
Author(s):  
Jau-Liang Chen ◽  
Hsu-Yang Chang
Keyword(s):  
Optimal Parameter ◽  
Back Propagation ◽  
Tail Length ◽  
Back Propagation Neural Network ◽  
Parameter Analysis ◽  
Forming Process ◽  
Current Time ◽  
Ball Size ◽  
Error Back Propagation ◽  
Free Air

In this paper, we are focusing on the FAB forming technology for fine pitch wire bonding. The parameters that affect the FAB formation include: 1) tail length; 2) spark gap; 3) Electric flame-off (EFO) voltage, current, time; 4) relative position between electrode plate and tail; 5) wire material; and 6) type of capillary. Except the last two items, all the other parameters can be quantified for analysis. By using Taguchi method it was found that EFO time and EFO current are the most important parameters that affect the formation free-air ball. The error-back-propagation neural network was then used to predict suitable EFO time and current setting. The main objective in this research is to find a suitable rule for parameters setting in order to control the FAB ball size as required. The result can be used in the future for optimal parameter setting and prediction of FAB formation.

Application of Neural Network in Free Air Ball Forming for Wire Bonder

1999 ◽  
Author(s):  
Jau-Liang Chen ◽  
Yeh-Chao Lin ◽  
Chun-Hsien Liu ◽  
Wen-Chang Kuo ◽  
Tzung-Ching Lee
Keyword(s):  
Neural Network ◽  
Back Propagation ◽  
Tail Length ◽  
Back Propagation Neural Network ◽  
Ball Bond ◽  
Error Back Propagation ◽  
Fine Pitch ◽  
Free Air ◽  
Network Method

Abstract The shape and size of free-air-ball formation deeply affect the quality of wire bonding. It not only affects the bondability of first bond (ball bond), but also affects the possibility of processing low loop height bonding for thin form packages and high I/O fine pitch packages. Several parameters, such as tail length, spark gap, supplied voltage, current and time of electrical flame-off unit etc., will affect the free-air-ball formation. This paper represents a study of using error-back-propagation neural network method to analyze the effect of each parameter and to predict the final result of the ball forming. From the experiment, it is shown that neural network can not only be used to precisely predict the size of ball formation, but also saves sampling time.

Parameter analysis and optimization of paper feeding devices to minimize jamming and simultaneous feeding of multiple pages

2011 ◽  
Vol 225 (11) ◽  
pp. 2673-2684 ◽  
Author(s):  
H Kim ◽  
J Lee
Keyword(s):  
Feeding Rate ◽  
Experimental Testing ◽  
Back Propagation ◽  
Back Propagation Neural Network ◽  
Parameter Analysis ◽  
Design Parameters ◽  
Causality Analysis ◽  
Nsga Ii ◽  
Analysis And Design ◽  
Analysis Of Means

This article investigates an optimal design for a paper-feeding system that minimizes both jamming and simultaneous feeding of multiple papers, hereafter referred to as the multi-feeding rate. A total of 11 design parameters for the paper transfer device, the paper separation device, and the paper guide path are selected and analysed in this study. A test jig for feeding and transferring papers is manufactured to obtain experimental data for use in parameter analysis and design optimization. Back-propagation neural network-based causality analysis is employed to extract five dominant variables among 11 design parameters, and the results of causality analysis are compared with sensitivity results obtained from the analysis of means in the context of experimental design. Five-variable, second-order polynomial based approximate meta-models for jam rate and multi-feeding rate are then constructed, and numerical optimization is performed using NSGA-II, a non-dominated sorting genetic algorithm. Finally, two numerical Pareto optimal solutions are verified via experimental testing.

scholarly journals Bending Angle Prediction Model Based on BPNN-Spline in Air Bending Springback Process

2017 ◽  
Vol 2017 ◽  
pp. 1-11 ◽  
Author(s):  
Zhefeng Guo ◽  
Wencheng Tang
Keyword(s):  
Prediction Model ◽  
Back Propagation ◽  
Back Propagation Neural Network ◽  
Numerical Control ◽  
Experimental Sample ◽  
Bending Angle ◽  
Air Bending ◽  
Error Back Propagation ◽  
Element Simulation ◽  
Bending Process

In order to rapidly and accurately predict the springback bending angle in V-die air bending process, a springback bending angle prediction model on the combination of error back propagation neural network and spline function (BPNN-Spline) is presented in this study. An orthogonal experimental sample set for training BPNN-Spline is obtained by finite element simulation. Through the analysis of network structure, the BPNN-Spline black box function of bending angle prediction is established, and the advantage of BPNN-Spline is discussed in comparison with traditional BPNN. The results show a close agreement with simulated and experimental results by application examples, which means that the BPNN-Spline model in this study has higher prediction accuracy and better applicable ability. Therefore, it could be adopted in a numerical control bending machine system.

A Two-Stage Optimization System for the Plastic Injection Molding with Multiple Performance Characteristics

2012 ◽  
Vol 468-471 ◽  
pp. 386-390 ◽  
Author(s):  
Wen Chin Chen ◽  
G.L. Fu ◽  
Denni Kurniawan
Keyword(s):  
Injection Molding ◽  
Experimental Work ◽  
Optimal Parameter ◽  
Back Propagation ◽  
Back Propagation Neural Network ◽  
Process Parameter ◽  
Plastic Injection Molding ◽  
Two Stage ◽  
Optimization System ◽  
Plastic Injection

This study proposes a two-stage optimization system to generate the optimal process parameter settings of multi-quality characteristics in the plastic injection molding (PIM) products. In the first stage, Taguchi orthogonal array was employ to arrange the experimental work and to calculate the S/N ratio to determine the initial process parameter settings. Then, S/N ratio predictor and S/N quality predictor was constructed by employed the back-propagation neural network (BPNN). In addition, S/N ratio predictor was along with simulated annealing (SA) used to search for the first optimal parameter combination in order to reduce the PIM process variance. In the second stage, BPNN quality predictor and particle swarm optimization (PSO) was intended to find the optimal parameter settings for the best quality specification. Results from the experimental work show that the proposed two-stage optimization system can create the best process parameter settings which not only meet the quality specification, but also effectively reduce cost.

scholarly journals Chaotic Extension Neural Network-Based Fault Diagnosis Method for Solar Photovoltaic Systems

2014 ◽  
Vol 2014 ◽  
pp. 1-9 ◽  
Author(s):  
Kuo-Nan Yu ◽  
Her-Terng Yau ◽  
Jian-Yu Li
Keyword(s):  
Neural Network ◽  
Fault Diagnosis ◽  
Back Propagation ◽  
Back Propagation Neural Network ◽  
Photovoltaic System ◽  
Photovoltaic Systems ◽  
Solar Photovoltaic ◽  
Error Back Propagation ◽  
Point Tracking ◽  
Diagnosis Method

At present, the solar photovoltaic system is extensively used. However, once a fault occurs, it is inspected manually, which is not economical. In order to remedy the defect of unavailable fault diagnosis at any irradiance and temperature in the literature with chaos synchronization based intelligent fault diagnosis for photovoltaic systems proposed by Hsieh et al., this study proposed a chaotic extension fault diagnosis method combined with error back propagation neural network to overcome this problem. It used the nn toolbox of matlab 2010 for simulation and comparison, measured current irradiance and temperature, and used the maximum power point tracking (MPPT) for chaotic extraction of eigenvalue. The range of extension field was determined by neural network. Finally, the voltage eigenvalue obtained from current temperature and irradiance was used for the fault diagnosis. Comparing the diagnostic rates with the results by Hsieh et al., this scheme can obtain better diagnostic rates when the irradiances or the temperatures are changed.

scholarly journals Constructing early diagnosis model of colorectal cancer based on expression profile

2020 ◽  
Author(s):  
shaoqiang Wang ◽  
Yifan Wang ◽  
Shudong Wang
Keyword(s):  
Expression Profiles ◽  
Back Propagation ◽  
Occult Blood ◽  
Back Propagation Neural Network ◽  
Bowel Cancer ◽  
Fecal Occult Blood ◽  
Detection Model ◽  
Error Back Propagation ◽  
Model Based ◽  
Diagnosis Model

AbstractpurposeIn order to break through the restrictive factors such as the fecal occult blood test (FOBT) in the routine detection of bowel cancer, which is susceptible to diet and drugs, and the high cost and inconvenience of microscopy, Seeking a possible FOBT alternative.MethodsAn error back propagation neural network (BPNN) algorithm was used to construct a CRC diagnosis model based on expression profiles.ResultsThe accuracy of the model on the training and test sets is 0.943 and 0.935, respectively. AUC all reached above 0.95.ConclusionThe CRC molecular detection model based on expression profiles provides a possible alternative to FOBT. It provides a new approach and method for the clinical diagnosis of bowel cancer.

Dynamic Load Balancing Using Hybrid Approach

2019 ◽  
Vol 9 (3) ◽  
pp. 75-88
Author(s):  
Sunita Gond ◽  
Shailendra Singh
Keyword(s):  
Neural Network ◽  
Load Balancing ◽  
Hybrid Approach ◽  
Back Propagation ◽  
Back Propagation Neural Network ◽  
Error Back Propagation ◽  
Proposed Model ◽  
Multiple Process ◽  
Trained Neural Network ◽  
Sequence Quality

Load balancing in a cloud environment for handling multiple process of different size is an important issue. Many advanced technologies are incorporated in the processes-based resource allocation which enhances the system efficiency. The steps of allotting resources to process can be done by taking data which helps to analyze and make important decisions at runtime. This article focuses on the allocation of cloud resources where two models were developed, the first was TLBO (Teacher Learning Based Optimization), a genetic algorithm which finds the correct position for the process to execute. Here, some information used for analysis was total number of machines, memory, execution time, etc. So, the output of the TLBO process sequence was used as training input for the Error Back Propagation Neural Network for learning. This trained neural network improved the work job sequence quality. Training was done in such a way that all sets of features were utilized to pair with their process requirement and current position. For increasing the reliability of the work, an experiment was done on a real dataset. Results show that the proposed model has overcome various evaluation parameters on a different scale as compared to previous approaches adopted by researchers.

Sign in / Sign up

Export Citation Format

Share Document