Skip to menu

Research

Sensors and AI in Manufacturing

 

Cyber Physical Manufacturing Systems for Powder Bed Fusion Additive Manufacturing

LPBF_sensor2.jpgOne of the most promising additive manufacturing techniques is laser powder bed fusion (LPBF). An outstanding problem in LPBF is part quality issues caused by the complex and multi-dimensional relationship between its process parameters and part quality. Because of its layer-based fabrication nature, part quality is inconsistent even with the same material and process parameters, which requires discrete real-time process control with real-time metrology. This project addresses the knowledge gap of the real-time defect identification and the real-time process control to mitigate the defects. To bridge this gap, the project will address the following objectives: (1) Develop real-time measurement science for LPBF using multi-sensor, data fusion, and deep learning, (2) Develop a self-adaptive multi-dimensional correlation model between various defects and process parameters using deep learning, and (3) Develop a data-driven cyber manufacturing platform of LPBF for its real-time process control.

Development of Inverse Sintering Solution for Additive Manufacturing using Machine Learning

The database of actual and virtual sintering experiments assembled with the microstructure analysis data and the generative adversarial design model will be used to tackle the inverse design problem to optimize the process parameters of the sintering assistive additive manufacturing process. To mitigate the risk of overfitting and loss of generalization, we will consider two approaches for supervised machine learning. The first is a direct mapping between the macro/micro-structure characteristics and the processing parameters in the form of an artificial neural network. The second approach is a physics-informed deep-network model that also incorporates the microstructures as a separately trained representation. By comparing the two models and closing the feedback loop to experiments, we will be able to validate the reliability of the models which paves the way to deployment of machine learning models in sintering assistive additive manufacturing.

 

High-Resolution and Fast Fringe Projection Profilometry (FPP)

dualfpp.pngEmergence and advancement of high precision additive manufacturing (AM) techniques such as powder bed fusion require highly accurate full-field and real-time 3D measurements for monitoring process-induced defects and for ensuring the quality of printed products. However, most of the off-the-shelf 3D sensing techniques are not optimized for in-situ measurements for AM and the existing studies on topography measurement for AM failed to provide necessary resolution or field of view. To bridge this gap, the project will address the following objectives: (1) Develop a novel spatiotempotal FPP that applies simultaneous interferences in both spatial and temporal domains in FPP, (2) Apply multiplex color projection on the spatiotemporal FPP for real-time measurement, and (3) Study phase retrival from fringe patterns using deep learning.

 

 

Image-based Automatic Defect Identification and Classification using Machine Learing

mldefect.pngSmart factory is a current trend that causes increasing demand of automatic defect inspection using machine vision and sensory hardware. For the production of well-defined patterned products such as liquid crystal display glass and printed circuit board, rule-based algorithms can be used to find defects. However, if a product contains irregular or atypical patterns, it is difficult to automatically estimate its defects using conventional image processing algorithms. For example, defects on additive manufactured layer, solar cell, fiber, or leather might not be obvious to define and detect. To solve this problem, our interests lie in using machine learning techniques for defect detection and classification without human involvements. The machine learning techniques enable automatic learning of defect features regardless of the complexity and irregularity of sample images. The research results are expected to provide a smart solution for real-time defect inspection and classification in various manufacturing processes.

 

Sensors and AI in Biotechnology

 

Silent Speech Recognition using Wearable EMG/EEG Sensors and Machine Learning

 

 

Real-time Monitoring of Human Emotions using Wearable Sensors and Machine Learning

 

 

Soft Tissue Characterization (STC) using Sub-microscale Tensile Testing

tensiletesting.png

Characterizing the mechanical properties of biological tissues is a critical step in many medical and bioengineering fields such as regenerative medicine, robot-aided surgery, and mathematical modeling of biological systems. Although various tissue-characterization methods have been developed such as AFM, indentation, pipette aspiration, pressure myography, and elastography, there remains some limitations on accuracy and validity for measuring small and flexible soft tissues. To bridge this gap, the project will address the following objectives: (1) Develop a sub-microscale tensile testing technique to measure the mechanical propertoes of soft tissues and (2) Design a method for attaching the tissue samples to the clamps of the tensile tester and to avoid slipping of samples during testing.

 

3D-Printed Wearable Sensor
biosensor.jpg

Polyvinylidene difluoride (PVDF) is a polymer that has been used in various research endeavors for the piezoelectric properties on display in the beta phase. There are numerous studies on how to optimize sensitivity and elastic dexterity of this material for tactile sensing purposes. One of the techniques for fabricating PVDF for this use is the fused deposition modeling (FDM), one of the widely used 3D printing techniques. We study the application of FDM in the fabrication of a tactile sensor by printing PVDF material in textile. The focus of this research will be on the optimization of the shape of the sensing element for tactile sensitivity, structural dexterity and integrity.