Machine learning (ML) has shown to be an effective alternative to physical models for quality prediction and process optimization of metal additive manufacturing (AM). However, the inherent “black box” nature of ML techniques such as those represented by artificial neural networks has often presented a challenge to interpret ML outcomes in the framework of the complex thermodynamics that govern AM. While the practical benefits of ML provide an adequate justification, its utility as a reliable modeling tool is ultimately reliant on assured consistency with physical principles and model transparency. To facilitate the fundamental needs, physics-informed machine learning (PIML) has emerged as a hybrid machine learning paradigm that imbues ML models with physical domain knowledge such as thermomechanical laws and constraints. The distinguishing feature of PIML is the synergistic integration of data-driven methods that reflect system dynamics in real-time with the governing physics underlying AM. In this paper, the current state-of-the-art in metal AM is reviewed and opportunities for a paradigm shift to PIML are discussed, thereby identifying relevant future research directions.
more »
« less
Prediction of flotation efficiency of metal sulfides using an original hybrid machine learning model
Froth flotation process is extensively used for selective separation of base metal sulfides from uneconomic mineral resources. Reliable prediction of process outcomes (metal recovery and grade) is vital to ensure peak performance. This work employs an innovative hybrid machine learning (ML) model—constructed by combining the random forest model and the firefly algorithm—to predict froth flotation efficiency of galena and chalcopyrite in relation to various experimental process parameters. The hybrid model's prediction performance was rigorously evaluated, and compared against four different standalone ML models. The outcomes of this study illustrate that the hybrid ML model has the prediction ability to process outcomes with high‐fidelity, while consistently outperforming the standalone ML models.
more » « less- NSF-PAR ID:
- 10458144
- Publisher / Repository:
- Wiley Blackwell (John Wiley & Sons)
- Date Published:
- Journal Name:
- Engineering Reports
- Volume:
- 2
- Issue:
- 6
- ISSN:
- 2577-8196
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
more » « less
Abstract Ultrahigh temperature ceramics (UHTCs) have melting points above 3000°C and outstanding strength at high temperatures, thus making them apposite structural materials for high‐temperature applications. Di‐borides, nitride, and carbide compounds—processed via various techniques—have been extensively studied and used in the manufacture of UHTCs. Current analytical models, based on our current but incomplete understanding of the theory, are unable to produce a priori predictions of mechanical properties of UHTCs based on their mixture designs and processing parameters. As a result, researchers have to rely on experiments—which are often costly and time‐consuming—to understand composition–structure–performance links in UHTCs. This study employs machine learning (ML) models (i.e., random forest and artificial neural network models) to predict Young's modulus, flexural strength, and fracture toughness of UHTCs in relation to a wide range of mixture designs, processing parameters, and testing conditions. Outcomes demonstrate that adequately trained ML models can yield reliable predictions, a priori, of the three aforesaid mechanical properties. The prediction performance on Young's modulus is superior to flexural strength and fracture toughness. Next, the ML model with the best prediction performance is utilized to evaluate and rank the impacts of input variables on Young's modulus. Finally, on the basis of such classification of consequential and inconsequential input variables, this study develops an easy‐to‐use, closed‐form analytical model to predict Young's modulus of UHTCs. Overall, this study highlights the ability of data‐driven numerical models to complement, or even replace, time‐consuming experiments, thereby accelerating the development of UHTCs.
-
more » « less
Abstract Alkali‐activated mortar (AAM) is an emerging eco‐friendly construction material, which can complement ordinary Portland cement (OPC) mortars. Prediction of properties of AAMs—albeit much needed to complement experiments—is difficult, owing to substantive batch‐to‐batch variations in physicochemical properties of their precursors (i.e., aluminosilicate and activator solution). In this study, a machine learning (ML) model is employed; and it is shown that the model—once trained and optimized—can reliably predict compressive strength of AAMs solely from their initial physicochemical attributes. Prediction performance of the model improves when multiple compositional descriptors of the aluminosilicate are combined into a singular, composite chemostructural descriptor (i.e.,
network ratio andnumber of constraints ); thus, reducing the degrees of freedom. Through interpretation of the ML model's outcomes—specifically the variable importance for the AAMs’ compressive strength—a simple, easy‐to‐use, closed‐form analytical model is developed. Results demonstrate that the analytical model yields predictions of compressive strength of AAMs without scarifying much accuracy compared to the ML model. Overall, this study's outcomes demonstrate a roadmap—incorporates composite chemostructural descriptors in ML models—that can be employed to design AAMs to achieve targeted compressive strength. -
more » « less
Abstract This study investigates whether coupling crop modeling and machine learning (ML) improves corn yield predictions in the US Corn Belt. The main objectives are to explore whether a hybrid approach (crop modeling + ML) would result in better predictions, investigate which combinations of hybrid models provide the most accurate predictions, and determine the features from the crop modeling that are most effective to be integrated with ML for corn yield prediction. Five ML models (linear regression, LASSO, LightGBM, random forest, and XGBoost) and six ensemble models have been designed to address the research question. The results suggest that adding simulation crop model variables (APSIM) as input features to ML models can decrease yield prediction root mean squared error (RMSE) from 7 to 20%. Furthermore, we investigated partial inclusion of APSIM features in the ML prediction models and we found soil moisture related APSIM variables are most influential on the ML predictions followed by crop-related and phenology-related variables. Finally, based on feature importance measure, it has been observed that simulated APSIM average drought stress and average water table depth during the growing season are the most important APSIM inputs to ML. This result indicates that weather information alone is not sufficient and ML models need more hydrological inputs to make improved yield predictions.
-
We study online convex optimization with switching costs, a practically important but also extremely challenging problem due to the lack of complete offline information. By tapping into the power of machine learning (ML) based optimizers, ML-augmented online algorithms (also referred to as expert calibration in this paper) have been emerging as state of the art, with provable worst-case performance guarantees. Nonetheless, by using the standard practice of training an ML model as a standalone optimizer and plugging it into an ML-augmented algorithm, the average cost performance can be highly unsatisfactory. In order to address the "how to learn" challenge, we propose EC-L2O (expert-calibrated learning to optimize), which trains an ML-based optimizer by explicitly taking into account the downstream expert calibrator. To accomplish this, we propose a new differentiable expert calibrator that generalizes regularized online balanced descent and offers a provably better competitive ratio than pure ML predictions when the prediction error is large. For training, our loss function is a weighted sum of two different losses --- one minimizing the average ML prediction error for better robustness, and the other one minimizing the post-calibration average cost. We also provide theoretical analysis for EC-L2O, highlighting that expert calibration can be even beneficial for the average cost performance and that the high-percentile tail ratio of the cost achieved by EC-L2O to that of the offline optimal oracle (i.e., tail cost ratio) can be bounded. Finally, we test EC-L2O by running simulations for sustainable datacenter demand response. Our results demonstrate that EC-L2O can empirically achieve a lower average cost as well as a lower competitive ratio than the existing baseline algorithms.more » « less
