8 results found
Attar HR, Foster A, Li N, 2023, Development of a deep learning platform for sheet stamping geometry optimisation under manufacturing constraints, Engineering Applications of Artificial Intelligence, Vol: 123, Pages: 1-23, ISSN: 0952-1976
Sheet stamping is a widely adopted manufacturing technique for producing complex structural components with high stiffness-to-weight ratios. However, designing such components is a non-trivial task that requires careful consideration of manufacturing constraints to avoid introducing defects in the final product. To address this challenge, this research introduces a novel deep-learning-based platform that optimises 3D component designs by considering manufacturing capabilities. This platform was realised by developing a methodology to combine two neural networks that handle non-parametric geometry representations, namely a geometry generator based on Signed Distance Fields (SDFs) and an image-based manufacturability surrogate model. This combination enables the optimisation of complex geometries that can be represented using various parameterisation schemes. The optimisation approach implemented in the platform utilises gradient-based techniques to update the inputs to the geometry generator based on manufacturability information from the surrogate model. The platform is demonstrated using two geometry classes, Corners and Bulkheads, each having three geometry subclasses, with four diverse case studies conducted to optimise these geometries under post-stamped thinning constraints. The case studies demonstrate how the platform enables free morphing of complex geometries, while also guiding manufacturability-driven geometric changes in a direction that leads to significant improvements in component quality. For instance, one of the cases shows that optimising the complex component geometry can reduce the maximum thinning from 45% to satisfy the thinning constraint of 10%. By utilising the proposed platform, designers can identify optimal component geometries that ensure manufacturing feasibility for sheet stamping, reducing design development time and design costs.
Attar HR, Foster A, Li N, 2023, Implicit neural representations of sheet stamping geometries with small-scale features, Engineering Applications of Artificial Intelligence, Vol: 123, Pages: 1-21, ISSN: 0952-1976
Geometric deep learning models, like Convolutional Neural Networks (CNNs), show promise as surrogate models for predicting sheet stamping manufacturability but lack design variables essential for inverse problems like geometric optimisation. Recent developments in deep learning have enabled geometry generation from compact latent spaces that are suitable for optimisation. However, current methods do not accurately model small-scale geometric features that are crucial for stamping performance. This study proposes a new deep learning-based method to address this limitation and generate detailed stamping geometries for optimisation. Specifically, neural networks are trained to generate Signed Distance Fields (SDFs) for stamping geometries, where the zero-level-set of each SDF implicitly represents the generated geometry. A new training approach is proposed for generating SDFs of stamping geometries, which involves supervising geometric properties of the SDFs. A novel loss function is introduced that directly acts on the zero-level-set and places high emphasis on learning small-scale features. This approach is compared with the state-of-the-art approach DeepSDF by Park et al. (2019), which explicitly supervises SDF values using ground truth data. The geometry generation performance of networks trained using both approaches is evaluated quantitatively and qualitatively. The results demonstrate significantly greater geometric accuracy with the proposed approach, which can faithfully generate small-scale features. Further analysis of the new approach reveals an organised learned latent space and varying the network input generates high-quality geometries from this space. By integrating with CNN-based manufacturability surrogate models by Attar et al. (2021), this work could enable the first-ever manufacturability-constrained optimisation of arbitrary sheet stamping geometries, potentially reducing geometry design time and cost.
Attar HR, Li N, 2022, Optimisation of deep drawn corners subject to hot stamping constraints using a novel deep-learning-based platform, IOP Conference Series: Materials Science and Engineering, Vol: 1238, Pages: 012066-012066, ISSN: 1757-8981
State-of-the-art hot stamping processes offer improved material formability and therefore have potential to successfully form challenging components. The feasibility of components to be formed through these processes is dependent on their geometric design and its complex interactions with the hot stamping environment. In industrial practice, trial-and-error approaches are currently used to update non-feasible designs where simulation runs are needed each time a design change is made. These approaches make the design process resource intensive and require considerable numerical and process expertise. To demonstrate a superior approach, this study presents a novel application of a deep-learning-based optimisation platform which adopts a non-parametric geometric modelling strategy. Here, deep drawn corner geometries from different geometry subclasses were optimised to minimise wasted volume due to radii while avoiding excessive post-stamping thinning. A neural network was trained to generate families of deep drawn corner geometries where each geometry was conditioned on an input latent vector. Another neural network was trained to predict the thinning distributions obtained from forming these geometries through a hot stamping process. Guided by these distributions, the latent vector, and therefore geometry, was iteratively updated by a new gradient-based optimisation technique. Overall, it is demonstrated that the platform is capable of optimising geometries, irrespective of complexity, subject to imposed post-stamped thinning constraints.
Attar H, Li N, Foster A, 2021, A method for determining equivalent hardening responses to approximate sheet metal viscoplasticity, MethodsX, Vol: 8, Pages: 1-20, ISSN: 2215-0161
Recently developed elevated temperature metal forming technologies improve material formability and address the springback issues of cold forming. However, the behaviour of alloys at elevated temperatures is viscoplastic and therefore considerably more complex than under cold forming conditions. This complex behaviour creates a barrier for industrial designers when designing for elevated temperature processes, and therefore leads to these processes often being overlooked. To overcome this barrier, a new method is proposed here to determine simpler strain hardening responses that are equivalent to the viscoplastic responses of alloys at elevated temperatures. The equivalent hardening responses are expressed by single material hardening curves and their hardening exponents are taken as parameters to approximate sheet metal viscoplasticity. This method therefore makes it possible to develop early-stage design guidelines that consider different materials and stamping processes. The method was successfully applied to two viscoplastic alloys under hot stamping conditions to determine equivalent hardening responses. The feasibility of the method has been assessed through comparing finite element simulations using the determined equivalent material models with ones using viscoplastic models. The result showed that the thinning distributions obtained were consistent in both cases, providing evidence that the method can be applied to a range of component designs.•The proposed equivalent material models are simpler than existing viscoplastic material models and can be derived directly from experimental stress-strain data.•Creating design guidelines from equivalent hardening exponents enables a straightforward way to compare between cold and hot stamping capabilities. This comparison makes it possible to make manufacturing process and material selection decisions quickly and effectively at the onset of a design process.•Design guidelines enabled by the proposed m
Xu Q, Nie Z, Xu H, et al., 2021, SuperMeshing: a new deep learning architecture for increasing the mesh density of physical fields in metal forming numerical simulation, Journal of Applied Mechanics, Vol: 89, Pages: 1-10, ISSN: 0021-8936
In stress field analysis, the finite element method is a crucial approach, in which the mesh-density has a significant impact on the results. High mesh density usually contributes authentic to simulation results but costs more computing resources. To eliminate this drawback, we propose a data-driven mesh-density boost model named SuperMeshingNet that uses low mesh-density as inputs, to acquire high-density stress field instantaneously, shortening computing time and cost automatically. Moreover, the Res-UNet architecture and attention mechanism are utilized, enhancing the performance of SuperMeshingNet. Compared with the baseline that applied the linear interpolation method, SuperMeshingNet achieves a prominent reduction in the mean squared error (MSE) and mean absolute error (MAE) on the test data. The well-trained model can successfully show more excellent performance than the baseline models on the multiple scaled mesh-density, including 2X, 4X, and 8X. Enhanced by SuperMeshingNet with broaden scaling of mesh density and high precision output, FEA can be accelerated with seldom computational time and cost.
Reza Attar H, Li N, Foster A, 2021, A new strategy for developing component design guidelines for aluminium alloy corners formed through cold and hot stamping processes, Materials & Design, Vol: 207, ISSN: 0264-1275
In recent sheet metal forming research, efforts have been largely focused on determining optimum processing parameters, while intuitive guidelines to efficiently develop feasible component geometries are rarely considered. Consequently, there are currently no suitable design support tools that can guide component designers using the most recent, and therefore unfamiliar, sheet metal manufacturing technologies. This paper aims to address this bottleneck by proposing a strategy for the creation of early stage manufacturing design guidelines for the common limiting design requirement of deep corners. Aluminium alloys formed under both cold, and elevated temperature working conditions are considered. A new methodology to simplify the analysis of complex viscoplastic behaviour of aluminium alloys at elevated temperatures into an equivalent strain hardening response is presented. The effects and trends of the deep corner geometry and simplified material hardening characteristics on the post-form thinning distribution are identified. New equation sets are proposed which model the identified trends and enable the development of intuitive design maps. Following the approach proposed in this paper, an awareness of the available design envelope can be created at the early stages of a design process to guide component design, material, and manufacturing process selection decisions for deep corner geometries.
Attar HR, Zhou H, Foster A, et al., 2021, Rapid feasibility assessment of components to be formed through hot stamping: A deep learning approach, Journal of Manufacturing Processes, Vol: 68, Pages: 1650-1671, ISSN: 1526-6125
The novel non-isothermal Hot Forming and cold die Quenching (HFQ) process canenable the cost-effective production of complex shaped, high strength aluminiumalloy panel components. However, the unfamiliarity of designing for the newprocess prevents its widescale adoption in industrial settings. Recent researchefforts focus on the development of advanced material models for finite elementsimulations, used to assess the feasibility of new component designs for theHFQ process. However, FE simulations take place late in design processes,require forming process expertise and are unsuitable for early-stage designexplorations. To address these limitations, this study presents a novelapplication of a Convolutional Neural Network (CNN) based surrogate as a meansof rapid manufacturing feasibility assessment for components to be formed usingthe HFQ process. A diverse dataset containing variations in component geometry,blank shapes, and processing parameters, together with corresponding physicalfields is generated and used to train the model. The results show that nearindistinguishable full field predictions are obtained in real time from themodel when compared with HFQ simulations. This technique provides an invaluabletool to aid component design and decision making at the onset of a designprocess for complex-shaped components formed under HFQ conditions.
Attar HR, Zhou H, Li N, 2021, Deformation and thinning field prediction for HFQ® formed panel components using convolutional neural networks, International Deep-Drawing Research Group Conference (IDDRG 2021), Publisher: IOP Publishing, Pages: 1-11, ISSN: 1757-8981
The novel Hot Forming and cold die Quenching (HFQ®) process can provide cost-effective and complex deep drawn solutions through high strength aluminium alloys. However, the unfamiliarity of the new process prevents its widescale adoption in industrial settings, while accurate Finite Element (FE) simulations using the most advanced material models take place late in design processes and require forming process expertise. Machine learning technologies have recently been proven successful in learning complex system behaviour from representative examples and have the potential to be used as design support tools for new forming technologies such as HFQ®. This study, for the first time, presents a novel application of a Convolutional Neural Network (CNN) based surrogate to predict the deformation and thinning fields for variable deep drawn geometries formed using HFQ® technology. A dataset based on deep drawn geometries and corresponding FE results is generated and used to train the model. The results show that near indistinguishable full field predictions in real time are obtained from the surrogate when compared with HFQ® simulations. This technique can be adopted in industrial settings to aid in both concept and detailed component design for complex-shaped panel components formed under HFQ® conditions, without underlying knowledge of the forming process.
This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.