Industry 4.0, in its search for improvements to processes and efficient products, has increasingly invested in the use and development of high-performance materials for its production lines. This is exemplified by the introduction of CFRP in the aeronautical industry, since this composite material has reduced the weight of aircraft and improved their performance. For the construction of large structures, drilling processes are also necessary to fix parts. However, this machining process can cause failures in the structure as a whole. These structural failures occur due to fragmentation, tearing, or detachment of the matrix fiber, significantly reducing the quality and reliability of the final equipment. In this scenario, it is important to preventively detect these intrinsic production failures that lead to the condemnation of the final parts. One indirect detection method is acoustic emission. This work presents a feasibility study focused on the application of data-driven methods for delamination detection and tool wear monitoring in composite machining. A setup for a helical interpolation end-milling drilling process was performed under varying machining conditions, from mild to severe, on CFRP composite plates. Acoustic emission (AE) signals were acquired at each machining pass. The methodology involved selecting an optimal frequency band to obtain information about the wear of the drilling tool through RMS and power spectral density (PSD) analysis, followed by using correlation indices to characterize tool wear progression. The results demonstrate the potential of spectral and statistical techniques to support real-time monitoring and decision-making in advanced composite manufacturing.

Tool Condition Monitoring (TCM) systems have become increasingly important in industrial automation due to the need to improve efficiency and reduce manufacturing costs. These systems use advanced sensors to capture signals during machining processes, allowing for the early detection of faults and prediction of tool life. This study explores the potential of using the cosine similarity (CS) method as a practical technique for analyzing acoustic emission (AE) signals and monitoring tool wear during milling operations. Acoustic signals were applied to the CS method under reference conditions and after potential damage. We used 9000 samples of the milling cutter passing over the workpiece, collected from experiments with milling machines using the AE sensor WD925 at a frequency of 100 kHz. The CS method tracked wear proportionally in each case. As the tool wore down, its similarity to the intact tool decreased, proving to be an effective indicator for condition monitoring. However, the change in CS calculation was not as pronounced as the tool wear observed, suggesting that having enough data is crucial for this methodology in condition monitoring. A longer sampling period is necessary to capture significant signal variations and effectively detect losses in similarity. This provides a significant amount of data and, as a result, leads to more conclusive findings for the process in question.

Falls pose a significant threat to the elderly population, often resulting in severe health complications such as fractures and other adverse outcomes, which can drastically lower their quality of life. The early detection of fall risks is crucial in mitigating the impact of such events. Various technologies have been developed to address this issue, including alert systems that notify users of imminent risks due to environmental factors or physiological changes. However, accurately detecting and distinguishing between normal activities, imminent fall risks, and actual falls remains a challenge. This study proposes a machine learning approach using the XGBoost algorithm to improve the fall detection accuracy among the elderly. A dataset comprising 2039 samples of data on the proximity to objects, spatial location changes, heart rate, blood oxygen saturation (SpO₂), blood sugar levels, and pressure applied by the user, categorized into normal, imminent fall risk, and fall classes, was utilized to train and test the model. The model was trained on 70% of the data, with 30% allocated for testing. Hyperparameter optimization was performed using a randomized search with cross-validation. Previous studies have reported an accuracy of 0.9667 for the same dataset. In contrast, this study achieved an accuracy of 1.0, demonstrating a significant improvement in the overall performance compared to earlier work. The confusion matrix demonstrates the model’s ability to distinguish between all three classes with no false positives. Additionally, sensitivity tests were conducted by varying the training sample sizes and randomizing the data splits, confirming the model’s robustness in different conditions. These results show that the proposed method was able to correctly sort all the samples in the training and tests, outperforming previous studies in detecting fall-related events, reducing the likelihood of false alarms, and enhancing resource allocation for elderly care.