Using Regression Analysis for Automated Material Selection in Smart Manufacturing

Pavlenko, Ivan; Piteľ, Ján; Ivanov, Vitalii; Berladir, Kristina; Mižáková, Jana; Kolos, Vitalii; Trojanowska, Justyna

doi:10.3390/math10111888

Cited by 32 publications

(6 citation statements)

References 45 publications

(41 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They are determined according to the stresses from the regulatory documents for materials or experiments. If the stress is positive, the tensile strength is selected for tensile, if negative it is selected for compression [ 13 ].…”

Section: Methodsmentioning

confidence: 99%

Composite Lamina Model Design with the Use of Heuristic Optimization

2023

View full text Add to dashboard Cite

In engineering practice, a problem is quite often faced in which the number of unknown parameters exceeds the number of conditions or requirements or, otherwise, there are too many requirements for too few parameters to design. Such under- or over-defined tasks are sometimes not possible to solve using a direct approach. The number of solutions to such problems is multiple, and it is most rational to search for the optimal one by numerical methods since the more unknown design parameters there are to be designed, the more potential solutions there are. This article discusses a way to find an optimal solution to such an underdetermined problem by heuristic optimization methods on the basis of the example of designing a composite wing skin of an aircraft. Several heuristic approaches, specifically gradient descent and Tabu search, are studied to solve the design problem and to locate an optimal solution. They are also compared to a conventional direct approach. The examined composite lamina is optimized by the target function of minimum weight with the constraints of strength and buckling failure criteria. In most of the observed cases, the heuristic method designed structures which were considerably better than the structures that were obtained by conventional direct approaches in terms of the weight to load ratio.

show abstract

Section: Methodsmentioning

confidence: 99%

Composite Lamina Model Design with the Use of Heuristic Optimization

2023

View full text Add to dashboard Cite

show abstract

“…Furthermore, the computational challenges associated with simulating and optimizing the physical model curtail its practicality [ 3 ]. Conversely, regression models may yield suboptimal outcomes when confronted with the nonlinear and intricate output landscape of extensive manufacturing systems, as they struggle to adequately capture the underlying information and intricate relationships hidden within the vast dataset [ 4 ].…”

Section: Introductionmentioning

confidence: 99%

Predictive modelling framework on the basis of artificial neural network: A case of nano-powder mixed electric discharge machining

Sana,

Farooq,

Anwar

et al. 2023

Heliyon

View full text Add to dashboard Cite

“…Modern tools of engineering analyses allow for the direct finding of the stress-strain behaviour of such structures without a replacement of the core used by some solid orthotropic material, i.e., without its "smearing" [24,25]. However, the labour intensity of this approach, most likely, is justified only in cases where verification analysis is required [26,27]. In practical terms, designers have always used and continue to use analytical models of honeycomb structures, allowing them to theoretically calculate the consolidated physico-mechanical characteristics of cores, expressed via their geometric parameters and the properties of the material that they are made of [28].…”

Section: Introductionmentioning

confidence: 99%

Effect of Ply Orientation on the Mechanical Performance of Carbon Fibre Honeycomb Cores

Kondratiev,

Píštěk,

Gajdachuk

et al. 2023

Polymers

View full text Add to dashboard Cite

Carbon fibres used as a honeycomb core material (subject to a proper in-depth analysis of their reinforcement patterns) allows solving the thermo-dimensional stability problem of the units for space systems. Based on the results of numerical simulations with the support of finite element analysis, the paper provides an evaluation of the accuracy of analytical dependencies for the determination of the moduli of elasticity of a carbon fibre honeycomb core in tension/compression and shear. It is shown that a carbon fibre honeycomb reinforcement pattern has a significant impact on the mechanical performance of the carbon fibre honeycomb core. For example, for honeycombs measuring 10 mm in height, the maximum shear modulus values corresponding to the reinforcement pattern of ±45° exceed the minimum values for a reinforcement pattern of 0° and 90° by more than 5 times in the XOZ plane and 4 times for the shear modulus in the YOZ plane. The maximum modulus of the elasticity of the honeycomb core in the transverse tension, corresponding to a reinforcement pattern of ±75°, exceeds the minimum modulus for the reinforcement pattern of ±15° more than 3 times. We observe a decrease in the values of the mechanical performance of the carbon fibre honeycomb core depending on its height. With a honeycomb reinforcement pattern of ±45°, the decrease in the shear modulus is 10% in the XOZ plane and 15% in the YOZ plane. The reduction in the modulus of elasticity in the transverse tension for the reinforcement pattern does not exceed 5%. It is shown that in order to ensure high-level moduli of elasticity with respect to tension/compression and shear at the same time, it is necessary to focus on a reinforcement pattern of ±64°. The paper covers the development of the experimental prototype technology that produces carbon fibre honeycomb cores and structures for aerospace applications. It is shown by experiments that the use of a larger number of thin layers of unidirectional carbon fibres provides more than a 2-time reduction in honeycomb density while maintaining high values of strength and stiffness. Our findings can permit a significant expansion of the area of application relative to this class of honeycomb cores in aerospace engineering.

show abstract

Using Regression Analysis for Automated Material Selection in Smart Manufacturing

Cited by 32 publications

References 45 publications

Composite Lamina Model Design with the Use of Heuristic Optimization

Composite Lamina Model Design with the Use of Heuristic Optimization

Predictive modelling framework on the basis of artificial neural network: A case of nano-powder mixed electric discharge machining

Effect of Ply Orientation on the Mechanical Performance of Carbon Fibre Honeycomb Cores

Contact Info

Product

Resources

About