Additive manufacturing via selective laser sintering (SLS) enables the rapid production of geometrically complex polyamide 12 (PA12) components. However, conventional pointwise analysis techniques often overlook the full depth of continuous experimental datasets, thus limiting the interpretation of structure–function relationships that are essential to high-performance design. This study employs functional data analysis (FDA) to elucidate the microstructural, chemical, thermal, and mechanical behaviours of SLS-fabricated PA12, focusing on the effects of build orientation (horizontal, transverse, vertical) and wall thickness (2.0–3.0 mm). The samples were produced via a commercial SLS platform and characterised via X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR), differential scanning calorimetry (DSC), and tensile testing. The FDA was applied to raw, normalised, and first derivative datasets via Python’s Scikit-FDA package, increasing the sensitivity to latent material variations. The findings demonstrate that the build orientation has a marked influence on the crystallinity and mechanical performance: horizontal builds yield narrower gamma-phase XRD peaks, greater structural order, and enhanced tensile properties, whereas vertical builds exhibit broader peak dispersion and greater thermal sensitivity. The wall thickness effects were minor, with only isolated flux-related anomalies. The FTIR spectra confirmed the consistent chemical stability across all the conditions. The FDA successfully identified subtle transitions and anisotropies that eluded traditional methods, underscoring its methodological strength for advanced polymer characterisation. These insights offer practical guidance for refining SLS process parameters and improving predictive design strategies in polymer-based additive manufacturing.

This study investigates the feasibility of acoustic signal analysis for the assessment of weld bead quality in the shielded metal arc welding (SMAW) process. The work focuses on comparing time-domain acoustic signals and time–frequency spectrogram representations for the classification of welds as accepted or rejected according to standard welding inspection criteria. Two key acoustic descriptors, the fundamental frequency (F₀) and the harmonics-to-noise ratio (HNR), were extracted and analyzed to evaluate statistical differences between the two weld quality classes. Statistical tests, including Anderson–Darling, Levene, ANOVA, and Kruskal–Wallis (α = 0.05), revealed significant differences between accepted and rejected welds. Accepted welds exhibited a bimodal HNR distribution associated with transient arc instability at the beginning and end of the bead, whereas rejected welds showed more uniform acoustic behavior throughout the process. Subsequently, the acoustic data were represented using both audio signals and spectrograms and used as inputs for ten supervised machine learning models, including Support Vector Classifier (SVC), Logistic Regression (LR), k-Nearest Neighbors (KNN), Decision Tree (DT), Random Forest (RF), Extra Trees (ET), Gradient Boosting (GB), and Naïve Bayes (NB). The results demonstrate that spectrogram-based representations significantly outperform time-domain signals, achieving accuracies of 0.95–0.96, ROC-AUC values above 0.95, and false positive and false negative rates below 6%. These findings indicate that, while scalar acoustic descriptors provide statistically significant insight into weld quality, time–frequency representations combined with machine learning enable a more robust and reliable framework for automated non-destructive evaluation, particularly in manual SMAW processes under realistic operating conditions.

This study evaluates the feasibility of acoustic signal analysis using different spectrographic transformation methods as a tool for assessing the quality of welding beads produced through the Shielded Metal Arc Welding (SMAW) process. Acoustic emissions were recorded during manual welding operations under controlled experimental conditions, using E6013 electrodes on A36 carbon steel plates. From the acoustic recordings of 400 welding samples, previously classified as accepted or rejected, two fundamental acoustic descriptors were extracted: the fundamental frequency (F₀) and the harmonic-to-noise ratio (HNR). These were analysed using parametric and non-parametric metrics to evaluate their discriminative capability. In addition, multiple supervised classifiers were trained and validated using stratified eight-fold cross-validation. The proposed framework enables a systematic comparison of different signal transformations and classification models for the evaluation of SMAW welding quality. Among the evaluated models (SVC, Gradient Boosting, and Extra Trees), precision rates of 90–95% were observed using Spectral Contrast, MEL, and CQT transformations. The results demonstrate that the implementation of various acoustic signal-based models and transformations for welding inspection offers a scalable and cost-effective solution for industrial quality control.