Research on the process optimization model of micro-clearance electrolysis-assisted laser machining based on BP neural network and ant colony

Sun, Aixi; Jin, Xue Qin; Chang, Yubo

doi:10.1007/s00170-016-8974-9

Cited by 13 publications

(5 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Artificial neural network (ANN) is widely used as a predictive technique for various advance machining processes. 35–40 Akbari et al 35 have proposed ANN approach for predicting the weld characteristics in laser welding of Ti6Al4V alloy. To estimate the surface roughness of steel and aluminium alloy during milling operation, Khorasani and Yazdi 36 have proposed ANN for predicting the responses.…”

Section: Introductionmentioning

confidence: 99%

“…Mondal et al 38 have used backpropagation ANN to estimate the outputs in laser machining process. The literature 40–43 suggests that ANN approach embedded with DOE method is an effective prediction technique. Since laser machining is a complex and costly process as compared to conventional machining processes, it is prudent to predict machining responses using ANN.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Drilling of micro-holes on titanium alloy using pulsed Nd:YAG laser: Parametric appraisal and prediction of performance characteristics

Chatterjee

Mahapatra

Bharadwaj

et al. 2018

Proceedings of the Institution of Mechanical Engineers, Part B:

View full text Add to dashboard Cite

Laser drilling is a preferable method in drilling micro and slender holes because of its non-physical contact between the tool and the workpiece leading to elimination of the problems like chatter and vibration. In this article, an experimental investigation of the laser drilling process is carried out on thin sheet (0.5 mm) of titanium alloy (Ti6Al4V) using pulsed Nd:YAG laser. The study investigates the effect of laser parameters such as laser energy, pulse repetition rate, pulse width and gas pressure on hole quality characteristics, namely, circularity at entry and exit, taper and spatter area. Taguchi’s L27 orthogonal array is adopted for conducting the experiments to establish significance of control parameters on the performance measures with less number of experimental run. Analysis of variance shows that laser energy and pulse width significantly influence on circularity, taper and spatter area. Optimal parameter setting for all the performance measures is suggested. Statistically valid empirical models relating hole quality characteristics with laser parameters have been developed based on least square regression analysis. Backpropagation artificial neural network is used predict the output responses considering laser process parameters as inputs. A mean square error within 5% for both the training and testing data suggests the adequacy and robustness of the proposed artificial neural network model. Variation of performance measures with respect to process parameters is predicted through the well-trained artificial neural network model. The pattern of variation exhibits good agreement with the main effect plots obtained through Taguchi method. Artificial neural network model produces mean relative error of 0.001, 0.001, 0.0437 and 0.0234 for circularity at entry, circularity at exit, taper and spatter area, respectively, when compared with experimental values. Since small values of mean relative error are obtained, the proposed artificial neural network model can be satisfactorily used for prediction of quality characteristics of laser micro-drilled holes.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Drilling of micro-holes on titanium alloy using pulsed Nd:YAG laser: Parametric appraisal and prediction of performance characteristics

Chatterjee

Mahapatra

Bharadwaj

et al. 2018

Proceedings of the Institution of Mechanical Engineers, Part B:

View full text Add to dashboard Cite

show abstract

“…The conjugate gradient method is to generate the conjugate direction of Hesse matrix of convex quadratic function f (x) in each iteration step without using the fastest descending direction at the current point. Each conjugate vector depends on the negative gradient at the iteration point, so as to establish the minimum point of not f (x) [11].…”

Section: Nonlinear Model Optimization Theory and Typical Optimization...mentioning

confidence: 99%

Research on Surveying and Mapping Data Processing Based on Nonlinear Mathematical Models and Deep Learning Optimization

Wei,

Wei

2024

SCPE

View full text Add to dashboard Cite

In order to deeply understand the mapping data processing of nonlinear mathematical model optimization, this paper uses nonlinear model optimization theory to process mapping data. When the precision of parameter approximation is high, the calculation results of each algorithm are the same, and the iteration times of the fastest descent method and simplex method are significantly increased compared with the other three algorithms. Therefore x01 = 5.42 in the parameter is kept unchanged near the truth value, and the convergence range of Newton method, the fastest descent method, conjugate gradient method and simplex method away from the truth value of parameter x02 = −0.25 is investigated. When the approximate value of undetermined parameters is high, the results of Newton algorithm and simplex algorithm are consistent, but the Newton iterative algorithm has faster convergence speed and higher computational efficiency than the simplex algorithm. Lower value when the undetermined parameter approximation precision, namely the undetermined parameter approximation and its true value is far off, may complete failure type Newton iteration algorithm, the simplex algorithm can be a supplement of the Newton method for the most part, the simplex method significantly reduces the requirements for initial value of parameters, data calculation efficiency has improved significantly.

show abstract

“…Numerous researchers have successfully applied ANN to forecasting nonlinear relationships between materials preparing process and properties, or materials compositions and properties. [12][13][14][15][16][17] However, the establishment of these neural networks basically relies on actual experimental results as training samples. It is difficult to achieve a wide range of predictions due to the limited quantity of samples.…”

Section: Introductionmentioning

confidence: 99%

Application of artificial neural network optimization for resilient ceramic parts fabricated by direct ink writing

Fan

Feng

Yang

et al. 2019

Int J Applied Ceramic Tech

View full text Add to dashboard Cite

Geometries of ceramic parts for high-temperature sealing have great influence on their compression-resilience behaviors. In this work, an accurate and large-scale artificial neural network (ANN) was established to match the relationship between structural parameters and mechanical properties of ZrO 2 parts fabricated by 3D printing. Four geometry parameters of the designed ZrO 2 parts were imported as input and apparent Young's modulus and maximum deformation simulated by finite element method (FEM) were imported as output. FEM calculation provided 400 groups of data for the training of ANN, which greatly improved the predicted accuracy of the network. The predicted results show the mechanical performance of the parts with a range of modulus from 9.24 × 10 −3 GPa to 100.35 × 10 −3 GPa and a range of maximum deformation from 2.32% to 5.80% can be forecasted with error less than 8%. Based on the optimized structural parameters, the designed ZrO 2 parts were fabricated by Direct Ink Writing (DIW) technique. The experimental compressionrebound property is comparable to that of ANN prediction. It demonstrates that the combined method of ANN and FEM is a preferable way to optimize the structure and guide the fabrication of complex ceramic parts by 3D printing method. K E Y W O R D Sadditive manufacturing, artificial neural network, finite element analysis, sealing 218 | FAN et Al. where N is the number of output, y n is the predicted value of the n th item, d n is the target value of the n th , and Q is the number of training cycle. Then the 400 groups of data were divided randomly into three parts: training groups, validation groups, and test groups with the ratio of 70%, 15%, and 15%. The network would stop training when the MSE of validation did not decrease continuously more than six times, otherwise the network was not good at generalization. Similarly, the network would also stop training when the training epochs came to 100 or the value of MSE was less than 0.00001.

show abstract

Research on the process optimization model of micro-clearance electrolysis-assisted laser machining based on BP neural network and ant colony

Cited by 13 publications

References 4 publications

Drilling of micro-holes on titanium alloy using pulsed Nd:YAG laser: Parametric appraisal and prediction of performance characteristics

Drilling of micro-holes on titanium alloy using pulsed Nd:YAG laser: Parametric appraisal and prediction of performance characteristics

Research on Surveying and Mapping Data Processing Based on Nonlinear Mathematical Models and Deep Learning Optimization

Application of artificial neural network optimization for resilient ceramic parts fabricated by direct ink writing

Contact Info

Product

Resources

About