Machine Learning Modeling for Real-Time Melt Pool Monitoring in Laser Powder Bed Fusion Additive Manufacturing: A Hybrid Approach
Summary
This paper presents a hybrid machine learning approach for real-time melt pool monitoring in laser powder bed fusion additive manufacturing, combining EfficientNetB0 feature extraction with Random Forest classification to achieve high accuracy and sub-millisecond inference time.
View Cached Full Text
Cached at: 06/24/26, 07:49 AM
# Machine Learning Modeling for Real-Time Melt Pool Monitoring in Laser Powder Bed Fusion Additive Manufacturing: A Hybrid Approach Source: [https://arxiv.org/abs/2606.23851](https://arxiv.org/abs/2606.23851) [View PDF](https://arxiv.org/pdf/2606.23851) > Abstract:This work investigates the implementation of artificial intelligence and machine learning \(AI/ML\) for real\-time monitoring in laser powder bed fusion \(LPBF\) additive manufacturing\. We developed a binary image classification framework for distinguishing normal and abnormal melt pool images using a balanced dataset of 1,200 images collected from Nickel superalloy 625 on the NIST AMMT platform\. The study evaluates accuracy and inference time based on control requirements and hardware limitations of open\-architecture LPBF machines\. We benchmark three transfer learning architectures \(ResNet50, EfficientNetB0, and MobileNetV2\) against two Random Forest approaches: one trained on EfficientNetB0 feature embeddings \(hybrid\) and one trained on raw pixel features \(baseline\)\. Images are stratified into 80/20 train\-test splits, with a further 90/10 validation split on the training set, and undergo standardized resizing, normalization, and label\-preserving data augmentation to emulate realistic process variability\. Each model is evaluated using accuracy, precision, recall, F1 score, and area under the receiver operating characteristic curve \(AUC\), along with training time, inference latency, and CPU & GPU usage to capture deployability constraints relevant to factory\-floor monitoring\. The hybrid EfficientNetB0\-plus\-Random Forest approach achieves the best performance on the held\-out test set, with an F1 score of 0\.9451, accuracy of 0\.9458, and AUC of 0\.9904, while maintaining sub\-millisecond per\-image inference \(1\.15 ms\)\. In contrast, purely deep learning models exhibit significantly higher inference times with lower accuracy\. These results demonstrate that combining pre\-trained convolutional features with classical ensemble methods provides a robust, computationally efficient route to real\-time melt pool anomaly detection in data\-limited additive manufacturing environments\. ## Submission history From: Xinyao Zhang \[[view email](https://arxiv.org/show-email/5d56aed3/2606.23851)\] **\[v1\]**Mon, 22 Jun 2026 18:38:29 UTC \(1,606 KB\)
Similar Articles
FusionSense: Tri-Stage Near-Sensor Learning for Runtime-Adaptive Multimodal Edge Intelligence
FusionSense introduces a tri-stage near-sensor learning framework for multimodal edge intelligence that jointly reduces compute and communication by using fusion-aware filtering, achieving up to 33× energy savings and significant data-reduction gains on RGB-Depth/LiDAR tasks.
Physics-Distilled Neural Network enabled by Large Language Models for Manufacturing Process-Property Predictive Modeling
This paper proposes a novel framework that uses LLMs to extract analytical physics priors from scientific literature and distills them into a lightweight neural network for high-accuracy, real-time manufacturing process-property prediction, even with limited data.
Multi-Head Attention-Based Feature Extractor Integration with Soft Actor-Critic for Porosity Prediction and Process Parameter Optimization in Additive Manufacturing
This paper proposes a novel architecture integrating multi-head attention with the Soft Actor-Critic algorithm for porosity prediction and process parameter optimization in additive manufacturing, achieving faster convergence and higher rewards than standard RL methods.
Geometry-free prediction of inertial lift forces in microfluidic devices using deep learning
This paper presents a novel deep learning approach to predict inertial lift forces in microfluidic devices without explicit geometric parameters, enabling better generalization to unseen channel cross-sections compared to previous models.
Physics-Informed Machine Learning for Short-Term Flood Prediction
Researchers propose a Physics-Informed Machine Learning (PIML) framework that integrates hydrological constraints into an LSTM loss function to improve short-term flood forecasting, particularly in data-scarce regimes. A 'Trend Alignment' constraint enforcing consistency between precipitation and discharge trends improves Nash-Sutcliffe Efficiency and eliminates unphysical predictions during extreme events.