The increasing availability of data, which becomes a continually increasing trend in multiple fields of application, has given machine learning approaches a renewed interest in recent years. Accordingly, manufacturing processes and sheet metal forming follow such directions, having in mind the efficiency and control of the many parameters involved, in processing and material characterization. In this article, two applications are considered to explore the capability of machine learning modeling through shallow artificial neural networks (ANN). One consists of developing an ANN to identify the constitutive model parameters of a material using the force–displacement curves obtained with a standard bending test. The second one concentrates on the springback problem in sheet metal press-brake air bending, with the objective of predicting the punch displacement required to attain a desired bending angle, including additional information of the springback angle. The required data for designing the ANN solutions are collected from numerical simulation using finite element methodology (FEM), which in turn was validated by experiments.
1 March 2019 in Engineering Analysis with Boundary Elements
Press brake air bending, a process of obtaining products by sheet metal forming, can be considered at first sight a simple geometric problem. However the accuracy of the obtained geometries involves the combination of multiple parameters directly associated with the tools and the processing parameters, as well as with the sheet metal materials and dimensions. The main topic herein presented deals with the capability of predicting the punch displacement process parameter that enables the product to be accurately shaped to a desired bending angle, in press brake air bending. In our approach, it is considered separately the forming process and the elastic recovery (i.e. the springback effect). Current solutions in press brake numerical control (computer numerical control) are normally configured by analytical models developed from geometrical analysis and including correcting factors. In our approach, it is proposed to combine the use of a learning tool, artificial neural networks, with a simulation and data generation tool (finite element analysis). This combination enables modeling the complex nonlinear behavior of the forming process and springback effect, including the validation of results obtained. A developed model taking into account different process parameters and tool geometries allow extending the range of applications with practical interest in industry. The final solution is compatible with its incorporation in a computer numerical control press brake controller. It was concluded that, using this methodology, it is possible to predict efficient and accurate final geometries after bending, being also a step forward to a “first time right” solution. In addition, the developed models, methodologies and obtained results were validated by comparison with experimental tests.
In this study, direct laser deposition (DLD) of nickel-based superalloy powders (Inconel 625) on structural steel (42CrMo4) was analysed. Cladding layers were produced by varying the main processing conditions: laser power, scanning speed, feed rate, and preheating. The processing window was established based on conditions that assured deposited layers without significant structural defects and a dilution between 15 and 30%. Scanning electron microscopy, energy dispersive spectroscopy, and electron backscatter diffraction were performed for microstructural characterisation. The Vickers hardness test was used to analyse the mechanical response of the optimised cladding layers. The results highlight the influence of preheating on the microstructure and mechanical responses, particularly in the heat-affected zone. Substrate preheating to 300 °C has a strong effect on the cladding/substrate interface region, affecting the microstructure and the hardness distribution. Preheating also reduced the formation of the deleterious Laves phase in the cladding and altered the martensite microstructure in the heat-affected zone, with a substantial decrease in hardness.
The continuous evolution of metallic alloys in the automotive industry has led to the development of more advanced and flexible constitutive models that attempt to accurately describe the various fundamental properties and behavior of these materials. These models have become increasingly complex, incorporating a larger number of parameters that require an accurate calibration procedure to fit the constitutive parameters with experimental data. In this context, machine learning (ML) methodologies have the potential to advance material constitutive modeling, enhancing the efficiency of the material parameter calibration procedure. Recurrent neural networks (RNNs) stand out among various learning algorithms due to their ability to process sequential data and overcome limitations imposed by nonlinearities and multiple parameters involved in phenomenological models. This study explores the modeling capabilities of long short-term memory (LSTM) structures, a type of RNN, in predicting the hardening behavior of a sheet metal material using the results of a standardized experimental three-point bending test, with the aim of extending this methodology to other experimental tests and constitutive models. Additionally, a variable analysis is performed to select the most important variables for this experimental test and assess the influence of friction, material thickness, and elastic and plastic properties on the accuracy of predictions made by neural networks. The required data for designing and training the network solutions are collected from numerical simulations using finite element methodology (FEM), which are subsequently validated by experiments. The results demonstrate that the proposed LSTM-based approach outperforms traditional identification techniques in predicting the material hardening parameters. This suggests that the developed procedure can be effectively applied to efficiently characterize different materials, especially those extensively used in industrial applications, ranging from mild steels to advanced high-strength steels.