Part distortion optimization of aluminum-based aircraft structures using finite element modeling and artificial neural networks

A numerical study of the effect of the thickness parameter on machining distortion for aluminum alloy plates

“…This work is a direct continuation of previous studies reported on the analysis of part distortion in aeronautical structures [18,19]; since these studies deal with a complex component, this work presents a new study for a relatively simple structure, a flat plate, intending to investigate the influence of the thickness geometrical parameter on the part distortion of it, and from this, draw conclusions of the phenomenon at vertical machining positioning conditions. As shown in figures 5, 6 and 7, it is possible to make a correlation between the optimum machining position t y that presents the lowest distortion and the location where the least residual stress exists; it is observed that regardless of the thickness e of the plates, the minimum distortion always presents in the location where the residual stresses are close to zero.…”

“…Several recent studies have reported that machining position is one of the parameters through which the impact of part distortion in aerostructures can be reduced; since the residual stresses existing in the raw material, and the material blanks, are self-balanced, the position from where a specific part is extracted directly affects the deformation that can be obtained at the end of a machined part. Thus, as reported by Chantzis et al [17], and more recently in the works by Rodríguez-Sánchez et al, and Barcenas et al [18,19], it is possible through finite element numerical methodologies to proactively mitigate even from the engineering analysis and design the final deformation of a structure after machining. For this reason, the analysis tools presented in such studies provide a proactive and predictive methods capable not only of supporting the reduction of such phenomenon, but it is also possible to investigate the effect of geometrical parameters of a structure on the final deformation after machining.…”

“…The present work presents the application of a numerical tool to study the part distortion phenomenon on plate structures; due to its relevance in engineering design, the effect of wall thickness in plates extracted from a target of aluminum alloy 7050-T7451 widely used in aeronautical structures and which presents residual stress profiles at two levels, high and low, will be analyzed. Since it has been demonstrated that this tool is effective in the prediction of structural part distortion, the algorithm reported in [19] will be implemented in ANSYS APDL software to run numerical simulations on flat plates with different thicknesses and different geometries to objectively answer the following question from a numerical analysis perspective: why thinner structures present higher distortion after machining?…”

Neural networks modeling of strain energy density and Tsai-Wu index in laminated composites

“…For example, to predict the strength value of a compound, 17 and for structural buckling analysis, 18 and also for machining optimization in structural applications. 19 In developing metamodels based on ML, the input and output parameters are diverse and focused on the property to be analyzed. For instance, in a study by Liu et al, 18 the laminate configuration (fiber steering and thickness) is introduced as inputs for an artificial neural network, whose function is to predict the performance of a part manufactured with laminated composite materials.…”

Research on ERS point registration of large-size measurement field based on neural network non-uniform temperature field

“…For the problem of thermal deformation caused by the temperature of machine tool processing, scholars use neural networks to build prediction models to predict accurately and compensate for the thermal deformation error of machine tools (Yan and Yang, 2009; Li et al , 2019; Zhang et al , 2019). Neural networks in aviation have advantages in aircraft parts machining deformation and hole parts stress prediction problems, using neural network models to optimize machining deformation in parts structure and predict hole parts stress distribution (Younis et al , 2018; Rodríguez-Sánchez et al , 2020). In the area of part deformation and defect prediction, neural networks are used to effectively predict structural deformation and defects in parts to reduce safety hazards (Chang and Wang, 2021; Wu et al , 2021; Wu et al , 2022).…”