2016
DOI: 10.1007/s00170-016-9562-8
|View full text |Cite
|
Sign up to set email alerts
|

Prediction of bead geometry in cold metal transfer welding using back propagation neural network

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3

Citation Types

0
3
0

Year Published

2019
2019
2023
2023

Publication Types

Select...
8

Relationship

0
8

Authors

Journals

citations
Cited by 28 publications
(3 citation statements)
references
References 8 publications
0
3
0
Order By: Relevance
“…The ANN has been proven its reliability to these researchers in predicting bead geometry and other properties with a rather small dataset, which is usually less than 100 samples, in almost all of the welding methods. Additionally, the ANN also has the advantages of continuous updating with new data, handling large number of inputs and outputs neurons, and filtering noises, showing its great potential in industrial manufacturing application [19]. While all researches mentioned above focus on the bead property prediction with stable welding parameters in one sample, which only meets laboratory requirement, not the actual manufacturing requirement.…”
Section: Introductionmentioning
confidence: 99%
“…The ANN has been proven its reliability to these researchers in predicting bead geometry and other properties with a rather small dataset, which is usually less than 100 samples, in almost all of the welding methods. Additionally, the ANN also has the advantages of continuous updating with new data, handling large number of inputs and outputs neurons, and filtering noises, showing its great potential in industrial manufacturing application [19]. While all researches mentioned above focus on the bead property prediction with stable welding parameters in one sample, which only meets laboratory requirement, not the actual manufacturing requirement.…”
Section: Introductionmentioning
confidence: 99%
“…Good penetration is conductive to ensuring product quality [15][16][17]. However, the relationship between the welding process parameters and the welding depth (WD) is unknown, non-linear, and complicated, thereby rendering it impractical to determine the optimal process parameters intuitively, even for skilled operators.…”
Section: Introductionmentioning
confidence: 99%
“…The power, focal diameter, and radiation time of the thermal-based process that used Gaussian heat source were sufficient to determine the unknown heat affected zone and temperature using hybrid genetic algorithm-artificial neural network (GA-ANN) model [27]. In a gas metal arc welding process with CMT metal transfer mode, the bead characteristics such as bead width, bead height, penetration depth, and dilution area were predicted using the welding speed, peak welding current and heat input [28]. In many of the experiments, the error associated with the artificial neural network was less compared to other models, although it depends on the application and the data associated with.…”
Section: Introductionmentioning
confidence: 99%