Applications Of ML & AI In Processing And Characterization Of Multi-Phase Materials
In PFL we combine the transformative potential of applied machine learning with materials engineering to drive innovation and solve complex challenges. Our work encompasses a broad spectrum of projects that harness the power of AI to enhance the field of materials engineering.
One area of our research focuses on predicting the oxidation behavior of Ultra-High Temperature Ceramics (UHTCs) using supervised machine learning. By developing models that accurately forecast oxidation, we gain valuable insights into the degradation processes of UHTCs, enabling us to mitigate their impact and advance their application in extreme environments.
We also explore the application of Generative Adversarial Networks (GANs) to filter out erroneous measurements during nonamechanical properties mapping. By refining data quality and improving the reliability of material characterization, we enhance the accuracy of our analyses and empower more precise material design.
\In the realm of ultrasonic treatment, we leverage machine learning techniques to predict the pressure sequence of cavitation. By extracting informative descriptors from segmentation models and utilizing Long Short-Term Memory (LSTM) models, we optimize the treatment process and enhance the efficiency of ultrasonic technologies.
Furthermore, we are actively involved in optimizing thermal spray coating properties through the analysis of in-flight particle parameters and machine learning. By leveraging advanced algorithms, we finetune coating quality, durability, and performance, opening up new avenues for diverse applications in various industries.
Our research lab is committed to pushing the boundaries of applied machine learning in materials engineering. We believe that by combining interdisciplinary collaboration, cutting-edge technologies, and a passion for discovery, we can revolutionize the field and shape a future where materials are engineered with unparalleled precision and efficiency.
