PHUNG, Van Binh ;TA, Duc Hai ;PHAM, Dinh Tung .
An Integrated ANN–GA-Based Framework for Multi-Parameter Design Optimization of a Large-Scale 3D Concrete Printer Frame in a Discrete Design Space.
Articles in Press, [S.l.], v. 0, n.0, p. , april 2026.
ISSN 0039-2480.
Available at: <https://www.sv-jme.eu/article/an-integrated-ann-ga-based-framework-for-multi-parameter-design-optimization-of-a-large-scale-3d-concrete-printer-frame-in-a-discrete-design-space/>. Date accessed: 01 jun. 2026.
doi:http://dx.doi.org/.
Phung, V., Ta, D., & Pham, D.
(0).
An Integrated ANN–GA-Based Framework for Multi-Parameter Design Optimization of a Large-Scale 3D Concrete Printer Frame in a Discrete Design Space.
Articles in Press, 0(0), .
doi:http://dx.doi.org/
@article{.,
author = {Van Binh Phung and Duc Hai Ta and Dinh Tung Pham},
title = {An Integrated ANN–GA-Based Framework for Multi-Parameter Design Optimization of a Large-Scale 3D Concrete Printer Frame in a Discrete Design Space},
journal = {Articles in Press},
volume = {0},
number = {0},
year = {0},
keywords = {3D concrete printer frame; Design optimization; ANN surrogate model; integrated ANN-GA approach; },
abstract = {This paper presents an integrated ANN–GA-based framework for multi-parameter design optimization of a large-scale concrete 3D printer frame in a discrete design space. Based on practical design requirements and operating conditions, a finite element (FE) model of the printer frame is developed using APDL® scripting, enabling automated evaluation of printhead deflection and natural frequencies. Using the FE-generated dataset, a multilayer feed-forward neural network (MLFFNN) is trained as a surrogate model to predict the structural responses of the frame. Parametric investigations demonstrate that the ANN surrogate model substantially reduces computational time while maintaining high prediction accuracy, with errors ranging from 1% to 4% compared to direct FE analysis. A mass-minimization optimization model is then formulated with ten design variables and four constraints related to printhead deflection and natural frequencies. Genetic Algorithms (GA) are employed to solve the optimization problem using two different approaches: direct optimization coupled with FE analysis and surrogate-based optimization using the ANN model. Notably, the optimization is conducted in a discrete design domain consistent with the standard dimensions of commercially available steel box sections. The optimal solutions obtained from different optimization strategies—including continuous and discrete FE-based models, the ANN surrogate model, and an experience-based design—are systematically compared. The optimization results demonstrate that the proposed framework achieves a structural weight reduction of 27%÷38% compared to the initial experience-based design. Furthermore, the ANN-based surrogate optimization reduces the total computational time from approximately 38 hours to about 200 seconds, clearly demonstrating the efficiency and practical applicability of the proposed approach for real-world large-scale machine design.},
issn = {0039-2480}, pages = {}, doi = {},
url = {https://www.sv-jme.eu/article/an-integrated-ann-ga-based-framework-for-multi-parameter-design-optimization-of-a-large-scale-3d-concrete-printer-frame-in-a-discrete-design-space/}
}
Phung, V.,Ta, D.,Pham, D.
0 April 0. An Integrated ANN–GA-Based Framework for Multi-Parameter Design Optimization of a Large-Scale 3D Concrete Printer Frame in a Discrete Design Space. Articles in Press. [Online] 0:0
%A Phung, Van Binh
%A Ta, Duc Hai
%A Pham, Dinh Tung
%D 0
%T An Integrated ANN–GA-Based Framework for Multi-Parameter Design Optimization of a Large-Scale 3D Concrete Printer Frame in a Discrete Design Space
%B 0
%9 3D concrete printer frame; Design optimization; ANN surrogate model; integrated ANN-GA approach;
%! An Integrated ANN–GA-Based Framework for Multi-Parameter Design Optimization of a Large-Scale 3D Concrete Printer Frame in a Discrete Design Space
%K 3D concrete printer frame; Design optimization; ANN surrogate model; integrated ANN-GA approach;
%X This paper presents an integrated ANN–GA-based framework for multi-parameter design optimization of a large-scale concrete 3D printer frame in a discrete design space. Based on practical design requirements and operating conditions, a finite element (FE) model of the printer frame is developed using APDL® scripting, enabling automated evaluation of printhead deflection and natural frequencies. Using the FE-generated dataset, a multilayer feed-forward neural network (MLFFNN) is trained as a surrogate model to predict the structural responses of the frame. Parametric investigations demonstrate that the ANN surrogate model substantially reduces computational time while maintaining high prediction accuracy, with errors ranging from 1% to 4% compared to direct FE analysis. A mass-minimization optimization model is then formulated with ten design variables and four constraints related to printhead deflection and natural frequencies. Genetic Algorithms (GA) are employed to solve the optimization problem using two different approaches: direct optimization coupled with FE analysis and surrogate-based optimization using the ANN model. Notably, the optimization is conducted in a discrete design domain consistent with the standard dimensions of commercially available steel box sections. The optimal solutions obtained from different optimization strategies—including continuous and discrete FE-based models, the ANN surrogate model, and an experience-based design—are systematically compared. The optimization results demonstrate that the proposed framework achieves a structural weight reduction of 27%÷38% compared to the initial experience-based design. Furthermore, the ANN-based surrogate optimization reduces the total computational time from approximately 38 hours to about 200 seconds, clearly demonstrating the efficiency and practical applicability of the proposed approach for real-world large-scale machine design.
%U https://www.sv-jme.eu/article/an-integrated-ann-ga-based-framework-for-multi-parameter-design-optimization-of-a-large-scale-3d-concrete-printer-frame-in-a-discrete-design-space/
%0 Journal Article
%R
%&
%P 1
%J Articles in Press
%V 0
%N 0
%@ 0039-2480
%8 2026-04-15
%7 2026-04-15
Phung, Van Binh, Duc Hai Ta, & Dinh Tung Pham.
"An Integrated ANN–GA-Based Framework for Multi-Parameter Design Optimization of a Large-Scale 3D Concrete Printer Frame in a Discrete Design Space." Articles in Press [Online], 0.0 (0): . Web. 01 Jun. 2026
TY - JOUR
AU - Phung, Van Binh
AU - Ta, Duc Hai
AU - Pham, Dinh Tung
PY - 0
TI - An Integrated ANN–GA-Based Framework for Multi-Parameter Design Optimization of a Large-Scale 3D Concrete Printer Frame in a Discrete Design Space
JF - Articles in Press
DO -
KW - 3D concrete printer frame; Design optimization; ANN surrogate model; integrated ANN-GA approach;
N2 - This paper presents an integrated ANN–GA-based framework for multi-parameter design optimization of a large-scale concrete 3D printer frame in a discrete design space. Based on practical design requirements and operating conditions, a finite element (FE) model of the printer frame is developed using APDL® scripting, enabling automated evaluation of printhead deflection and natural frequencies. Using the FE-generated dataset, a multilayer feed-forward neural network (MLFFNN) is trained as a surrogate model to predict the structural responses of the frame. Parametric investigations demonstrate that the ANN surrogate model substantially reduces computational time while maintaining high prediction accuracy, with errors ranging from 1% to 4% compared to direct FE analysis. A mass-minimization optimization model is then formulated with ten design variables and four constraints related to printhead deflection and natural frequencies. Genetic Algorithms (GA) are employed to solve the optimization problem using two different approaches: direct optimization coupled with FE analysis and surrogate-based optimization using the ANN model. Notably, the optimization is conducted in a discrete design domain consistent with the standard dimensions of commercially available steel box sections. The optimal solutions obtained from different optimization strategies—including continuous and discrete FE-based models, the ANN surrogate model, and an experience-based design—are systematically compared. The optimization results demonstrate that the proposed framework achieves a structural weight reduction of 27%÷38% compared to the initial experience-based design. Furthermore, the ANN-based surrogate optimization reduces the total computational time from approximately 38 hours to about 200 seconds, clearly demonstrating the efficiency and practical applicability of the proposed approach for real-world large-scale machine design.
UR - https://www.sv-jme.eu/article/an-integrated-ann-ga-based-framework-for-multi-parameter-design-optimization-of-a-large-scale-3d-concrete-printer-frame-in-a-discrete-design-space/
@article{{}{.},
author = {Phung, V., Ta, D., Pham, D.},
title = {An Integrated ANN–GA-Based Framework for Multi-Parameter Design Optimization of a Large-Scale 3D Concrete Printer Frame in a Discrete Design Space},
journal = {Articles in Press},
volume = {0},
number = {0},
year = {0},
doi = {},
url = {https://www.sv-jme.eu/article/an-integrated-ann-ga-based-framework-for-multi-parameter-design-optimization-of-a-large-scale-3d-concrete-printer-frame-in-a-discrete-design-space/}
}
TY - JOUR
AU - Phung, Van Binh
AU - Ta, Duc Hai
AU - Pham, Dinh Tung
PY - 2026/04/15
TI - An Integrated ANN–GA-Based Framework for Multi-Parameter Design Optimization of a Large-Scale 3D Concrete Printer Frame in a Discrete Design Space
JF - Articles in Press; Vol 0, No 0 (0): Articles in Press
DO -
KW - 3D concrete printer frame, Design optimization, ANN surrogate model, integrated ANN-GA approach,
N2 - This paper presents an integrated ANN–GA-based framework for multi-parameter design optimization of a large-scale concrete 3D printer frame in a discrete design space. Based on practical design requirements and operating conditions, a finite element (FE) model of the printer frame is developed using APDL® scripting, enabling automated evaluation of printhead deflection and natural frequencies. Using the FE-generated dataset, a multilayer feed-forward neural network (MLFFNN) is trained as a surrogate model to predict the structural responses of the frame. Parametric investigations demonstrate that the ANN surrogate model substantially reduces computational time while maintaining high prediction accuracy, with errors ranging from 1% to 4% compared to direct FE analysis. A mass-minimization optimization model is then formulated with ten design variables and four constraints related to printhead deflection and natural frequencies. Genetic Algorithms (GA) are employed to solve the optimization problem using two different approaches: direct optimization coupled with FE analysis and surrogate-based optimization using the ANN model. Notably, the optimization is conducted in a discrete design domain consistent with the standard dimensions of commercially available steel box sections. The optimal solutions obtained from different optimization strategies—including continuous and discrete FE-based models, the ANN surrogate model, and an experience-based design—are systematically compared. The optimization results demonstrate that the proposed framework achieves a structural weight reduction of 27%÷38% compared to the initial experience-based design. Furthermore, the ANN-based surrogate optimization reduces the total computational time from approximately 38 hours to about 200 seconds, clearly demonstrating the efficiency and practical applicability of the proposed approach for real-world large-scale machine design.
UR - https://www.sv-jme.eu/article/an-integrated-ann-ga-based-framework-for-multi-parameter-design-optimization-of-a-large-scale-3d-concrete-printer-frame-in-a-discrete-design-space/
Phung, Van Binh, Ta, Duc Hai, AND Pham, Dinh Tung.
"An Integrated ANN–GA-Based Framework for Multi-Parameter Design Optimization of a Large-Scale 3D Concrete Printer Frame in a Discrete Design Space" Articles in Press [Online], Volume 0 Number 0 (15 April 2026)
