Bayesian Optimization and Data Analytics for Thermoelectric Materials
Bayesian Optimization for Additive Manufacturing and Functional Materials
The design of chemical-based products and functional materials is vital to modern technologies, yet remains expensive and slow. Artificial intelligence and machine learning offer new approaches to leverage data to overcome these challenges. This review focuses on recent applications of Bayesian optimization (BO) to chemical products and materials including molecular design, drug discovery, molecular modeling, electrolyte design, and additive manufacturing. Numerous examples show how BO often requires an order of magnitude fewer experiments than the Edisonian search. The essential equations for BO are introduced in a self- contained primer specifically written for chemical engineers and others new to the area. Finally, the review discusses four current research directions for BO and their relevance to product and materials design.
Bayesian Optimization for Flash Sintering
Engineered solid-state thermoelectric materials can significantly improve energy efficiency and reduce emissions in modern industry by converting waste heat into electricity. However, the performance of many state-of-the-art thermoelectric materials remains inadequate for adoption beyond niche applications. Current efforts to optimize flash sintering, an important step in additive manufacturing of thermoelectric devices, rely on intuition-driven Edisonian search which can be extremely time-consuming. The alternative way is using Bayesian optimization (BO) framework that leverages a probabilistic surrogate model to emulate an expensive objective function and an acquisition function to recommend future experiments that optimally balance exploitation and exploration.
This work focus on using BO to assist experimentalists to achieve record high power factor in the flash sintering manufacturing process. The project has ended up with one conference paper in PSE 2021 and one journal paper in the energy&environmental science.
Bayesian Optimization for Thermoelectric Materials Composition
With the success of utilizing BO in the manufacturing process, we extended the success experience into the material composition of thermoelectric material. The project contains two sub-direction, optimizing the composition of N-type and P-type material. Our collaborator has finished the first task to increase the power factor two times compared with the initial experiment within 10 experiments. The following work is focusing on the optimization of P-type material. The project will contribute to two journal papers in the material science community.
Data-Driven Modeling and Design of Experiments for Thickness Control
The thickness control is vital in the manufacturing process of thermoelectric material. This project focus on using physical intuition with the data-driven method to build up a precise control model. The final delivered model allows experimentalists to fabricate the film with desired thickness under certain uncertainty.
[J1] Ke Wang, Alexander W. Dowling (2022). Bayesian optimization for chemical products and functional materials, Current Opinion in Chemical Engineering, 36, p. 100728
[J2] Mortaza Saeidi-Javash, Ke Wang, Minxiang Zeng, Tengfei Luo, Alexander Dowling, Yanliang Zhang (2022), Machine Learning-Assisted Ultrafast Flash Sintering of High-Performance and Wearable Silver-Selenide Thermoelectric Devices. (Energy&Environmental Science)
[J3] Ke Wang, Minxiang Zeng, Jialu Wang, Wenjie Shang, Yanliang Zhang, Tengfei Luo, Alexander Dowling (2022), When Physics-Informed Data Analytics Outperforms Black-box Machine Learning: A Case Study in Thickness Control for Additive Manufacturing. (Digital Chemical Engineering)
[J4] Wenjie Shang, Minxiang Zeng, Ali Tanvir, Ke Wang, Mortaza Saeidi-Javash, Alexander Dowling, Tengfei Luo, Yanliang Zhang, Hybrid Data-driven Discovery of High-performance Silver Selenide-based Thermoelectric Composites. (under review)
[C1] Ke Wang, Mortaza Saeidi-Javash, Minxiang Zeng, Zeyu Liu, Yanliang Zhang, Tengfei Luo, Alexander W. Dowling. Gaussian Process Regression Machine Learning Models for Photonic Sintering (2022). 14th International Symposium on Process Systems Engineering (PSE2021+). Ed. by Y. Yamashita, M. Kano.
Department of Energy DE-EE0009103, Optimizing Additive Manufacturing of Thermoelectric Materials using Bayesian Optimization-Enhanced Transfer Learning (PI: T. Luo, Notre Dame)
Prof. Tengfei Luo, U. Notre Dame
Prof. Yanliang Zhang, U. Notre Dame
Prof. David B. Go, U. Notre Dame
Prof. Minxiang Zeng
Prof. Mortaza Saeidi-Javash