Modelling of weld-bead geometry and hardness profile in laser welding of plain carbon steel using neural networks and genetic algorithms

Singh, Anitesh Kumar; Cooper, David E.; Blundell, N.; Pratihar, Dilip Kumar; Gibbons, Gregory John

doi:10.1080/0951192x.2013.834469

Cited by 37 publications

(10 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recent articles have reported attempts at using neural networks, and in particular, the multilayer perceptron (MLP) [1]- [4] in analysing AM data. However, the fundamental issues of small data samples, and crucially, that of characterising and accounting for uncertainties are not addressed.…”

mentioning

confidence: 99%

Variance Stabilised Optimisation of Neural Networks: A Case Study in Additive Manufacturing

Chen

Lee

Panoutsos

2021

2021 International Joint Conference on Neural Networks (IJCNN)

View full text Add to dashboard Cite

A new framework is presented for training neural networks that is based on the characterisation and stabilisation of measurement variations. The framework results in a number of useful properties that maximises the use of data as well as aiding in the interpretation of results in a principled manner. This is achieved via variance stabilisation and a subsequent standardisation step. The method is a general approach that may be used in any context where repeatability data is available. Standardisation in this manner allows goodness of fit to be quantified and measurement data to be interpreted from a statistical perspective. We demonstrate the utility of this framework in the analysis of advanced manufacturing data.

show abstract

mentioning

confidence: 99%

Variance Stabilised Optimisation of Neural Networks: A Case Study in Additive Manufacturing

Chen

Lee

Panoutsos

2021

2021 International Joint Conference on Neural Networks (IJCNN)

View full text Add to dashboard Cite

show abstract

“…According to numerous literature reports, 10,17,18 welding process parameters, such as the welding speed, welding current and protection gas flow, have a significant influence on the mechanical properties of a weld joint. To reduce the time consumption of welding experiments and simultaneously obtain the relationship between the input welding process parameters and the mechanical properties of the weld joint, the Taguchi method was applied in the welding process of GW53 magnesium alloy plates.…”

Section: Methodsmentioning

confidence: 99%

“…16 to optimize the depth of penetration, bead width and tensile strength in laser-GMAW butt welding. Singh 17 modeled weld-bead geometry and its cross-sectional micro-hardness in laser welding of plain carbon steel using neural networks and genetic algorithms. Gao 18 et al.…”

Section: Introductionmentioning

confidence: 99%

Optimization of the welding process parameters of Mg–5Gd–3Y magnesium alloy plates with a hybrid Kriging and particle swarm optimization algorithm

Sun

Cui³

et al. 2017

Proceedings of the Institution of Mechanical Engineers, Part C:

View full text Add to dashboard Cite

Selecting suitable welding process parameters to obtain optimal mechanical properties of the weld bead in AC gas tungsten arc welding is of vital importance. This paper presents a combination method of the Kriging model and particle swarm optimization for optimizing welding process parameters to achieve the optimum mechanical properties, such as the tensile strength and micro-hardness, of the weld bead in AC gas tungsten arc welding of GW53 magnesium alloy plates. The Taguchi orthogonal array is first employed to construct a database including the input process parameters (welding speed, welding current, and protection gas flow) and the responses (the tensile strength and micro-hardness of the weld joint). Then, the Kriging model is used to establish the relationships between the input process parameters and the responses. The optimal mechanical properties of the weld bead corresponding to the welding process parameters are obtained by the proposed hybrid Kriging and particle swarm optimization algorithm. Finally, the effectiveness of the proposed method is verified by contrasting the mechanical properties, such as the tensile strength and the average micro-hardness, in the welding base metal and the weld bead. Furthermore, the main reasons for the decrease in the mechanical properties of welded plates are described in this paper.

show abstract

“…As compared with the approximate equations, experimental methods have been reported to achieve better accuracy. Regression analysis [14][15][16][17][18][19][20][21][22] and artificial neural networks (ANNs) 17,20,[22][23][24][25][26][27] have been used widely to model the relationship between bead geometry and welding parameters based on experimental data. According to Xiong et al, 17 Kim et al 20 and Lee and Um, 22 ANNs show superior performance to regression analysis in predicting accuracy.…”

Section: Introductionmentioning

confidence: 99%

Optimal pass planning for robotic welding of large-dimension joints with nonuniform grooves

Yan

Fang

Ong

et al. 2017

Proceedings of the Institution of Mechanical Engineers, Part B:

View full text Add to dashboard Cite

Large-dimension welded joints are widely used in shipyards and oilfields. These joints typically have thick and nonuniform welding grooves, thus requiring multi-pass welding processes and tens of hours to complete. Although the development of robotic welding systems can shorten the welding time, it is important to have a detailed plan of the entire process before welding starts. Welding pass planning is crucial for the subsequent robot trajectory generation and provides the welding sequence in a deep groove and the welding parameters of every pass. A knowledge database is used in this research to obtain the relationship between the bead geometry and the welding parameters. A layer-by-layer welding pass planning scheme is proposed to address the nonuniform groove geometry, and an optimization method is proposed with the objective to maximize the section area of the weld bead, so that the number of passes can be minimized. The planning results show that the optimization method is able to provide feasible passes and welding parameters for every pass.

show abstract

Modelling of weld-bead geometry and hardness profile in laser welding of plain carbon steel using neural networks and genetic algorithms

Cited by 37 publications

References 23 publications

Variance Stabilised Optimisation of Neural Networks: A Case Study in Additive Manufacturing

Variance Stabilised Optimisation of Neural Networks: A Case Study in Additive Manufacturing

Optimization of the welding process parameters of Mg–5Gd–3Y magnesium alloy plates with a hybrid Kriging and particle swarm optimization algorithm

Optimal pass planning for robotic welding of large-dimension joints with nonuniform grooves

Contact Info

Product

Resources

About