Powder Bed Fusion (PBF) faces ongoing challenges in the areas of process monitoring and control. Standard methods for alleviating these issues rely on machine learning, which requires costly and time-consuming training data. Expense is compounded by the perceived necessity of using sensors with extremely high resolutions. This research avoids this cost by employing an Ensemble Kalman Filter (EnKF), which uses measured data to correct physics-based model predictions of the process, to monitor part internal temperature fields during building. This work tests EnKF performance, in simulation, for two model architectures, using simulated cameras of varying resolution as our measuring instruments. Crucially, we show that increasing camera resolution produces diminishing returns in EnKF accuracy, relative to the model predictions, with up to 81% error reduction. This result shows that current AM quality control practices with expensive sensors may be inefficient; with appropriate algorithms, cheaper setups may be used with little additional error.
more »
« less
The Ensemble Kalman Filter as a tool for estimating temperatures in the powder bed fusion process
Powder Bed Fusion (PBF) is a type of additive manufacturing process that builds parts out of metal powder in a layerwise fashion. Quality control (QC) remains an unsolved problem for PBF. Data-driven models of PBF are expensive to train and maintain, in terms of materials and machine time,
because they are sensitive to changes in processing conditions.The length and time scale discrepancies of the process make physics-based modeling impractical to implement. We propose monitoring PBF with an Ensemble Kalman Filter (EnKF). The EnKF combines the computational efficiency of datadriven
models with the flexibility of physics-based models, while mitigating the flaws of either method. We validate EnKF performance for linear process models, using finite element method data in place of measured experimental data. We show
that the EnKF can reduce the error signal 2-norm and 1-norm relative to the open loop model by as much as 75%.
more »
« less
- Award ID(s):
- 1738723
- NSF-PAR ID:
- 10331315
- Date Published:
- Journal Name:
- 2021 American Control Conference (ACC)
- Page Range / eLocation ID:
- 4369 to 4375
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract The temperature history of an additively manufactured part plays a critical role in determining process–structure–property relationships in fusion-based additive manufacturing (AM) processes. Therefore, fast thermal simulation methods are needed for a variety of AM tasks, from temperature history prediction for part design and process planning to in situ temperature monitoring and control during manufacturing. However, conventional numerical simulation methods fall short in satisfying the strict requirements of time efficiency in these applications due to the large space and time scales of the required multiscale simulation. While data-driven surrogate models are of interest for their rapid computation capabilities, the performance of these models relies on the size and quality of the training data, which is often prohibitively expensive to create. Physics-informed neural networks (PINNs) mitigate the need for large datasets by imposing physical principles during the training process. This work investigates the use of a PINN to predict the time-varying temperature distribution in a part during manufacturing with laser powder bed fusion (L-PBF). Notably, the use of the PINN in this study enables the model to be trained solely on randomly synthesized data. These training data are both inexpensive to obtain, and the presence of stochasticity in the dataset improves the generalizability of the trained model. Results show that the PINN model achieves higher accuracy than a comparable artificial neural network trained on labeled data. Further, the PINN model trained in this work maintains high accuracy in predicting temperature for laser path scanning strategies unseen in the training data.more » « less
-
Informed decision-making for sustainable manufacturing requires accurate manufacturing process environmental impact models with uncertainty quantification (UQ). For emerging manufacturing technologies, there is often insufficient process data available to derive accurate data-driven models. This paper explores an alternative mechanistic modeling approach using easy-to-access data from a given machine to perform Bayesian inference and reduce the uncertainty of model parameters. First, we derive mechanistic models of the cumulative energy demand (CED) for making aluminum (AlSi10) and nylon (PA12) parts using laser powder bed fusion (L-PBF). Initial parametric uncertainty is assigned to the model inputs informed by literature reviews and interviews with industry experts. Second, we identify the most critical sources of uncertainty using variance-based global sensitivity analyses; therefore, reducing the dimension of the problem. For metal and polymer L-PBF, critical uncertainty is related to the adiabatic efficiency of the process (a measure of the efficiency with which the laser energy is used to fuse the powder) and the recoating time per layer between laser scans. Data pertinent to both of these parameters include the part geometry (height and volume) and total build time. Between three and eight data points on part geometry and build time were collected on two different L-PBF machines and Bayesian inference was performed to reduce the uncertainty of the adiabatic efficiency and recoating time per layer on each machine. This approach was validated by subsequently taking direct parameter measurements on these machines during operation. The delivered electricity uncertainty is reduced by 40-70% after performing inference, highlighting the potential to construct accurate energy and environmental impact models of manufacturing processes using small easy-to-access datasets without interfering with the operations of the manufacturing facility.more » « less
-
null (Ed.)Abstract Despite the advantages and emerging applications, broader adoption of powder bed fusion (PBF) additive manufacturing is challenged by insufficient reliability and in-process variations. Finite element modeling and control-oriented modeling have been shown to be effective for predicting and engineering part qualities in PBF. This paper first builds a finite element model (FEM) of the thermal fields to look into the convoluted thermal interactions during the PBF process. Using the FEM data, we identify a novel surrogate system model from the laser power to the melt pool width. Linking a linear model with a memoryless nonlinear submodel, we develop a physics-based Hammerstein model that captures the complex spatiotemporal thermomechanical dynamics. We verify the accuracy of the Hammerstein model using the FEM and prove that the linearized model is only a representation of the Hammerstein model around the equilibrium point. Along the way, we conduct the stability and robustness analyses and formalize the Hammerstein model to facilitate the subsequent control designs.more » « less
-
Laser beam powder bed fusion (LB-PBF) is a widely-used metal additive manufacturing process due to its high potential for fabrication flexibility and quality. Its process and performance optimization are key to improving product quality and promote further adoption of LB-PBF. In this article, the state-of-the-art machine learning (ML) applications for process and performance optimization in LB-PBF are reviewed. In these applications, ML is used to model the process-structure–property relationships in a data-driven way and optimize process parameters for high-quality fabrication. We review these applications in terms of their modeled relationships by ML (e.g., process—structure, process—property, or structure—property) and categorize the ML algorithms into interpretable ML, conventional ML, and deep ML according to interpretability and accuracy. This way may be particularly useful for practitioners as a comprehensive reference for selecting the ML algorithms according to the particular needs. It is observed that of the three types of ML above, conventional ML has been applied in process and performance optimization the most due to its balanced performance in terms of model accuracy and interpretability. To explore the power of ML in discovering new knowledge and insights, interpretation with additional steps is often needed for complex models arising from conventional ML and deep ML, such as model-agnostic methods or sensitivity analysis. In the future, enhancing the interpretability of ML, standardizing a systemic procedure for ML, and developing a collaborative platform to share data and findings will be critical to promote the integration of ML in LB-PBF applications on a large scale.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.23919/ACC50511.2021.9482856
