KXKe Xu

Ke Xu

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Ke Xu has a total of 5 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Integrating Machine Learning with Multimodal Monitoring System Utilizing Acoustic and Vision Sensing to Evaluate Geometric Variations in Laser Directed Energy Deposition

    by Ke Xu, Chaitanya Krishna Vallabh, Souran Manoochehri

    Laser directed energy deposition (DED) additive manufacturing faces significant challenges in maintaining consistent part quality due to complex melt pool dynamics and process variations. While extensive research has focused on part defect detection, limited work has systematically validated the capability of utilizing process monitoring systems to evaluate melt pool dynamics, and process quality. This study presents a novel multimodal monitoring framework that synergistically integrates contact-based acoustic emission (AE) sensing with vision-based coaxial camera monitoring to enable layer-wise identification and evaluation of geometric variations in DED fabricated parts. The experimental study employs three distinct part configurations: a baseline part without any holes, the part with 3mm diameter through-hole, and the part with 5mm diameter through-hole. These specimens are served as test cases to evaluate the monitoring system's discerning capabilities. Raw sensor data undergoes preprocessing, where acoustic signals are subjected to filtering and feature extraction using time-domain and frequency-domain analysis, while coaxial camera data undergoes melt pool segmentation followed by morphological feature extraction. Multiple machine learning algorithms including support vector machines (SVM), neural networks, random forest, gradient boosting, XGBoost, and logistic regression are evaluated to identify optimal models for layer-wise geometric variations classification. The integrated multimodal strategy demonstrates superior classification performance of 94.4% as compared to individual sensing modalities of 87.8% (AE only) and 86.7% (Camera only). Experimental validation reveals that the integrated multimodal monitoring system effectively captures both structural vibration signatures and surface morphological changes associated with studied geometric variations. While this study focuses on specified part geometric variations, the demonstrated capability to discriminate between different geometric features establishes the technical foundation for future applications in characterizing any part variations such as geometric accuracies and manufacturing-induced defects.

  • TDIP: Tunable Deep Image Processing, a Real Time Melt Pool Monitoring Solution

    by Javid Akhavan, Youmna Mahmoud, Ke Xu, Jiaqi Lyu, Souran Manoochehri

    In the era of Industry 4.0, Additive Manufacturing (AM), particularly metal AM, has emerged as a significant contributor due to its innovative and cost-effective approach to fabricate highly intricate geometries. Despite its potential, this industry still lacks real-time capable process monitoring algorithms. Recent advancements in this field suggest that Melt Pool (MP) signatures during the fabrication process contain crucial information about process dynamics and quality. To obtain this information, various sensory approaches, such as high-speed cameras-based vision modules are employed for online fabrication monitoring. However, many conventional in-depth analyses still cannot process all the recorded data simultaneously. Although conventional Image Processing (ImP) solutions provide a targeted tunable approach, they pose a trade-off between convergence certainty and convergence speed. As a result, conventional methods are not suitable for a dynamically changing application like MP monitoring. Therefore, this article proposes the implementation of a Tunable Deep Image Processing (TDIP) method to address the data-rich monitoring needs in real-time. The proposed model is first trained to replicate an ImP algorithm with tunable features and methodology. The TDIP model is then further improved to account for MP geometries and fabrication quality based on the vision input and process parameters. The TDIP model achieved over 94% estimation accuracy with more than 96% R2 score for quality, geometry, and MP signature estimation and isolation. The TDIP model can process 500 images per second, while conventional methods taking a few minutes per image. This significant processing time reduction enables the integration of vision-based monitoring in real-time for processes and quality estimation.