The Zr-2.5%Nb alloy used to manufacture the pressure tubes for the CANDU 600 power plant experiences continual corrosion when the reactor’s fuel channels are functioning normally. Analysis of the hydrogen accumulation circumstances, the creation of zirconium hydride platelets, evaluation of the platelets’ reorientation in the thermo–mechanical field at volumetric flaws, and determination of the hydrogen morphology are therefore significant goals of the worldwide research being conducted in this field. In this paper, pressure tube alloy samples with artificial V flaws were hydrided by the technique of electrolytic deposition of a hydrogen film layer, followed by a specific thermal treatment. Micrographs of the Zr-2.5%Nb alloy samples in the radial cross-section were taken to examine the hydride reorientation phenomenon that occurred when the samples were heated under mechanical stress. The complex stress field around a tip flaw promoted the formation of some circles from hydride filaments. To characterize the zirconium hydride morphology in the reorientation zone at the volumetric flaw tip, image analysis and processing techniques were applied. To better characterize the reorientation zone from the point of view of embrittlement and mechanical behavior, this paper proposes a novel metric for the morphology of hydrides located in the reorientation zone. This work’s findings are beneficial for the investigation of CANDU fuel channels’ structural integrity.

Embedded voice-command systems are increasingly adopted in cyber-physical and industrial environments, yet their machine-learning pipelines remain vulnerable to data poisoning attacks that can silently degrade recognition performance or induce targeted misclassifications. This paper proposes a multi-stage poisoning detection pipeline tailored for embedded voice-command applications, combining fast integrity screening with progressively stronger detection models. First, an audio integrity layer analyzes waveform and spectrogram consistency to flag abnormal patterns indicative of injected perturbations or corrupted samples. Second, a lightweight machine-learning detector based on Random Forests operates on compact statistical descriptors extracted from spectrogram representations, enabling efficient on-device or edge-level screening. Third, a spectrogram-based Convolutional Neural Network provides high-sensitivity discrimination between clean and poisoned samples, acting as a robust validation stage in higher-assurance deployments. To increase resilience, the training procedure integrates regularization and controlled augmentation to mitigate overfitting to spurious artifacts and reduce poisoning sensitivity. The proposed approach is designed to be deployable under embedded constraints by separating low-cost screening from deeper analysis, and it offers a structured methodology for securing voice-command pipelines against poisoning threats.

In some Generation IV reactor configurations, embrittlement of the liquid metal can manifest in various forms, and this behaviour strongly depends on the specific solid–liquid couple. For the ALFRED demonstrator, which will be built at the RATEN ICN site in Romania, the study of embrittlement induced by molten lead in 316L stainless steel at temperatures of 350–450 °C is of interest. The purpose of the paper is to evaluate the effect of liquid metal embrittlement on the fracture mechanics properties of 316L tested in liquid lead. To do this, the “Normalisation Data Reduction Technique” (ASTM E1820) is used to obtain the J-R resistance curve for 316L steel in molten lead. In this way, the fracture mechanics parameters were obtained, indicating the fracture toughness of 316L steel in liquid lead with a saturated oxygen concentration at 350 °C. Optical and SEM examinations complement the analyses.