An official website of the United States government
Abstract There are many sources of manufacturing variations in sheet metal assemblies, such as automotive bodies. These include non-isotropic material properties from cold rolling, springback in stamping, and distortion from residual stresses when components are clamped and spot welded. FE simulations have been used to predict these variations in order to better design tooling and processes. Such simulations require expertise in complex, multi-stage nonlinear analysis. We are investigating the feasibility of training machine learning algorithms in order to democratize these types of analyses. This requires the curation of large, validated, and balanced data sets. To this end, we have developed a multi-stage finite element simulation workflow encompassing component stamping and joining with a focus on examining deformations due to springback in two-part assemblies. Three connected simulations comprise the workflow: (1) component stamping with capture of springback, (2) assembly clamping, and (3) assembly joining, then release. The workflow utilizes explicit dynamic finite element analysis (FEA) and includes the transfer of intermediate solutions (geometries/stresses), as well as extraction of key geometric parameters of springback from both component- and assembly-level simulations. The NUMISHEET 1993 U-draw/bending benchmark was referenced for its tooling geometry and utilized for verification of the forming process simulation; variations of material and geometry were also simulated. In summary, this work provides a means of generating a design space of flexible two-part assemblies for applications such as dataset generation, design optimization, and machine learning.
more »
« less
Springback prediction using machine learning: an application for simplified automotive body-in-white structures
Abstract Sheet metal stamped and welded assemblies, such as the ones used in automotive body-in-white (BIW) structures, have various sources of manufacturing variations during stamping and assembly processes. One of the major contributors to these variations is the springback on clamping release due to elastic recovery. Mitigating these variations requires expert knowledge of mechanical behavior, tooling, and process design. No analytical models can be used for the variety of geometries. Nonlinear FEA is also being used to predict springback, but it is time-consuming and requires specialized expertise, which makes it difficult to use in design exploration. Machine learning holds the promise of democratizing such complex analyses. This paper presents several case studies for data curation/generation, ML training, and validation. The prediction and quantification of the effects of springback are done on two levels: (i) low granularity, which involves predicting variations in certain parameters that are critical to measuring and understand spring back, and (ii) high granularity, predicting the shape of the component while taking into account the effects of springback and the stresses in the components. The data required to train, test, and validate the ML models were generated previously using an automated, integrated multi-stage simulation approach that was necessary to produce large datasets. Stamping simulations were validated against NUMISHEET benchmarks and also compared to test results published by other researchers. Subsequently, machine learning models were trained on the curated dataset to predict 2D stamped component shapes after springback and stress distributions across these shapes. For the assembly dataset, parameters such as unconstrained planar minimum zone magnitudes, angles between component planes, and twist angles are predicted using machine learning models, including linear and polynomial regression, decision trees, gradient boosting regression, support vector regression, and fully connected neural networks, and compared for their performance using consistent metrics. Hyper-parameter tuning is performed to optimize model performance, with artificial neural networks demonstrating promising capabilities in understanding variations in forming and multi-stage assembly processes.
more »
« less
- Award ID(s):
- 2029905
- PAR ID:
- 10607971
- Publisher / Repository:
- Springer Science + Business Media
- Date Published:
- Journal Name:
- The International Journal of Advanced Manufacturing Technology
- Volume:
- 139
- Issue:
- 3-4
- ISSN:
- 0268-3768
- Format(s):
- Medium: X Size: p. 1253-1273
- Size(s):
- p. 1253-1273
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Automotive structures are primarily made of flexible sheet metal assemblies. Flexible assemblies are prone to manufacturing variations like springback which may be caused due to non-isotropic material properties from cold rolling, springback in the forming process, and distortion from residual stresses when components are clamped, and spot welded. This paper describes the curation of a large data set for machine learning. The domain is that of flexible assembly manufacturing in multi stages: component stamping, configuring components into sub-assemblies, clamping and joining. The dataset is generated by nonlinear FEA. Due to the size of the data set, the simulation workflow has been automated and designed to produce variety and balance of key parameters. Simulation results are available not just as raw FE deformed (sprung back) geometries and residual stresses at different manufacturing stages, but also in the form of variation zones and fits. The NUMISHEET 1993 U-draw/bending was used a reference for tooling geometry and verification of the forming process. Additional variation in the dataset is obtained by using multiple materials and geometrical dimensions. In summary, the proposed simulation method provides a means of generating a design space of flexible multi-part assemblies for applications such as dataset generation, design optimization, and machine learning.more » « less
-
Given the urgency of climate change, development of fast and reliable methods is essential to understand urban building energy use in the sector that accounts for 40% of total energy use in USA. Although machine learning (ML) methods may offer promise and are less difficult to develop, discrepancy in methods, results, and recommendations have emerged that requires attention. Existing research also shows inconsistencies related to integrating climate change models into energy modeling. To address these challenges, four models: random forest (RF), extreme gradient boosting (XGBoost), single regression tree, and multiple linear regression (MLR), were developed using the Commercial Building Energy Consumption Survey dataset to predict energy use intensity (EUI) under projected heating and cooling degree days by the Intergovernmental Panel on Climate Change (IPCC) across the USA during the 21st century. The RF model provided better performance and reduced the mean absolute error by 4%, 11%, and 12% compared to XGBoost, single regression tree, and MLR, respectively. Moreover, using the RF model for climate change analysis showed that office buildings’ EUI will increase between 8.9% to 63.1% compared to 2012 baseline for different geographic regions between 2030 and 2080. One region is projected to experience an EUI reduction of almost 1.5%. Finally, good data enhance the predicting ability of ML therefore, comprehensive regional building datasets are crucial to assess counteraction of building energy use in the face of climate change at finer spatial scale.more » « less
-
Cancer diagnostics is an important field of cancer recovery and survival with many expensive procedures needed to administer the correct treatment. Machine Learning (ML) approaches can help with the diagnostic prediction from circulating tumor cells in liquid biopsy or from a primary tumor in solid biopsy. After predicting the metastatic potential from a deep learning model, doctors in a clinical setting can administer a safe and correct treatment for a specific patient. This paper investigates the use of deep convolutional neural networks for predicting a specific cancer cell line as a tool for label free identification. Specifically, deep learning strategies for weight initialization and performance metrics are described, with transfer learning and the accuracy metric utilized in this work. The equipment used for prediction involves brightfield microscopy without the use of chemical labels, advanced instruments, or time-consuming biological techniques, giving an advantage over current diagnostic methods. In the procedure, three different binary datasets of well-known cancer cell lines were collected, each having a difference in metastatic potential. Two different classification models were adopted (EfficientNetV2 and ResNet-50) with the analysis given for each stage in the ML architecture. The training results for each model and dataset are provided and systematically compared. We found that the test set accuracy showed favorable performance for both ML models with EfficientNetV2 accuracy reaching up to 99%. These test results allowed EfficientNetV2 to outperform ResNet-50 at an average percent increase of 3.5% for each dataset. The high accuracy obtained from the predictions demonstrates that the system can be retrained on a large-scale clinical dataset.more » « less
-
Refractory complex concentrated alloys (RCCAs) have drawn increasing attention recently owing to their balanced mechanical properties, including excellent creep resistance, ductility, and oxidation resistance. The mechanical and thermal properties of RCCAs are directly linked with the elastic constants. However, it is time consuming and expensive to obtain the elastic constants of RCCAs with conventional trial-and-error experiments. The elastic constants of RCCAs are predicted using a combination of density functional theory simulation data and machine learning (ML) algorithms in this study. The elastic constants of several RCCAs are predicted using the random forest regressor, gradient boosting regressor (GBR), and XGBoost regression models. Based on performance metrics R-squared, mean average error and root mean square error, the GBR model was found to be most promising in predicting the elastic constant of RCCAs among the three ML models. Additionally, GBR model accuracy was verified using the other four RHEAs dataset which was never seen by the GBR model, and reasonable agreements between ML prediction and available results were found. The present findings show that the GBR model can be used to predict the elastic constant of new RHEAs more accurately without performing any expensive computational and experimental work.more » « less
