Higher Institute on Territorial Systems for Innovation
PORTO@iris (Publications Open Repository TOrino - Politecnico di Torino)Not a member yet
146173 research outputs found
Sort by
Multiple input tangential interpolation-driven damage detection of a jet trainer aircraft
Full text linking systems. However, relevant literature mostly focuses on subsystems and parts, rather than full airframes. In
structural dynamics, modal parameters, such as natural frequencies and mode shapes, from any structure are
the main building blocks of vibration-based damage detection. However, traditional comparisons of these parameters are often ambiguous in large systems, complicating damage detection and assessment. The modified
total modal assurance criterion (MTMAC), an index well-known in the field of finite element model updating,
is extended to address this challenge and is proposed as an index for damage identification and severity assessment. To support the requirement for precise and robust modal identification of Structural Health Monitoring
(SHM), the improved Loewner Framework (iLF), known for its reliability and computational performance, is pioneeringly employed within SHM. Since the MTMAC is proposed solely as a damage identification and severity
assessment index, the coordinate modal assurance criterion (COMAC), also a well-established tool, but for damage localisation using mode shapes, is used for completeness. The iLF SHM capabilities are validated through
comparisons with traditional methods, including least-squares complex exponential (LSCE) and stochastic subspace identification with canonical variate analysis (SSI-CVA) on a numerical case study of a cantilever beam.
Furthermore, the MTMAC is validated against the traditional vibration-based approach, which involves directly
comparing natural frequencies and mode shapes. Finally, an experimental dataset from a BAE Systems Hawk T1A
jet trainer ground vibration test is used to demonstrate the iLF and MTMAC capabilities on a real-life, real-size
SHM problem, showing their effectiveness in detecting and assessing damage
Tribology of dual Pickering double emulsions: Machine learning-aided inner droplet analysis
Full text linkThis study investigated the tribological performance of Pickering water-in-oil emulsions and dual Pickering water-in-oil-in-water double emulsions (DEs) stabilized with particles at both the interfaces. W/O emulsions were stabilized by cocoa butter-based oleogel (CBolg) crystals, while DEs incorporating these emulsions were stabilized by whey protein microgels (WPM). The influence of temperature (21 and 37 °C) and surface texture (smooth vs biomimetic tongue-like surface) were investigated in tribology of W/O emulsions (30–60 % v/v water) and DEs (with 20 and 60 wt% W/O phase). In smooth surfaces, CBolg played a critical role in reducing the friction coefficient (μ) primarily via a fat-driven lubrication mechanism that was temperature dependent. While in DEs, smaller oil droplets encapsulating water provided similar lubrication to oil-based systems until starvation occurred. Strikingly, the water content in W/O emulsions exhibited distinct differences between emulsion systems within the biomimetic tongue-like surfaces, demonstrating lower lubricity at higher water concentration. Confocal microscopy images analyzed using Machine Learning (ML)-supported droplet segmentation enabled a more precise evaluation of structural changes within DEs when subjected to tribological stress. We demonstrated that although changes in inner droplet size altered in DEs, their contribution to the overall lubrication performance was minimal, due to their limited entrainment. Of more importance, the tribological performance was governed by the WPM with minimal influence from the droplet-entrained phenomena. These fundamental insights highlight the relevance of structured water in understanding frictional performance in emulsified systems, with structural integrity, composition, and topography of the tribological surface emerging as key factors
Light and shadows of smart contract development with LLMs
Full text linkSmart contract development remains almost inaccessible to non-experts developers despite the growing adoption of blockchain technology across industries. This paper evaluates the potential of Large Language Models (LLMs) for automated smart contract generation from legal agreements. The work systematically assesses the capabilities of four leading commercial LLMs – gpt-4-turbo (OpenAI), claude-3.5-sonnet (Anthropic), mistral-large (MistralAI), and gemini-1.5-pro (Google) – across a diverse range of legal agreements with varying complexity. The evaluation framework consists of a in-depth evaluation of structured code patterns – typical to smart contracts – to provide nuanced insights into model performances. The results reveal a performance hierarchy with claude-3.5-sonnet and gpt-4-turbo consistently outperforming mistral-large and gemini-1.5-pro, particularly when handling complex agreements such as mortgage note agreement and property sales agreement. A nonlinear relationship has been observed between contract complexity and model performance, with even top-performing models showing significant degradation when processing intricate legal structures. Although achieving syntactic correctness has become increasingly feasible, ensuring functional completeness and security remains challenging, as evidenced by high-impact vulnerabilities detected across all generated smart contracts. This work contributes to the growing discourse on LLM applications in blockchain technology by providing empirical evidence of current capabilities and limitations, establishing a robust foundation for future research in AI-assisted smart contract development
Fault tolerant multi-object tracking via temporal consistency filtering
No full textThis paper explores the fault tolerance of deep learning-based multi-object tracking under transient hardware faults. Such faults, often caused by radiation-induced bit flips or temporary electrical disturbances, can corrupt model parameters or intermediate activations during inference, leading to potential performance degradation. Unlike prior work focused on image classification, we examine the more complex task of tracking using ByteTrack with YOLOX and YOLO-Nano backbones. Our GPU memory fault injection framework reveals that common mitigation methods like activation clipping are ineffective in this context. We propose a novel Temporal Consistency Filter (TCF) leveraging frame-to-frame similarity to detect and correct faulty feature extractions. TCF significantly improves tracking stability, boosting MOTA by over 9\% and reducing identity switches by 17%, emphasizing the value of temporal consistency in robust tracking systems
Fluid dynamic and thermal characterization of a turbulent shear layer
Full text linkPlanar particle image velocimetry (PIV) and hot-wire anemometry are used to study the turbulent shear layer that forms naturally from an initial laminar state due to velocity differences between a jet and its surrounding quiescent region. Spectral Proper Orthogonal Decomposition (SPOD) applied to PIV data reveals the presence of dominant large-scale coherent structures in the post-transition turbulent shear layer. These structures span the full thickness of the shear layer and account for approximately 30% of the total turbulent kinetic energy. They are associated with a Strouhal number of approximately 0.2, similar to that of Kelvin–Helmholtz instabilities typically observed in pre-transition regimes. This observation raises questions about the underlying growth mechanism: while the elongated spatial patterns are consistent with continuous structure growth, the Strouhal number suggests a behaviour more typical of pairing-dominated dynamics. The estimated structure spacing supports the hypothesis of a nearly constant spacing-to-thickness ratio. The thermal shear layer, characterized with CCA and CVA cold wires, was shown to exhibit self-similarity. The growth rate of the thermal shear layer differs from that of the velocity shear layer. New analytical formulations for universal non-dimensional profiles of mean temperature and standard deviation are presented
Teaching systemic design to foster sustainability learning in non-design curricula
Full text linkCircular economy and sustainable development in rural areas are phenomena that call for new skills and knowledge encompassing methods and tools to foster systemic thinking. Over the past 20 years, the design discipline has significantly turned towards complexity, advocating a “designerly” approach to systems thinking that brings the focus closer to humanity. This paper aims to present a successful teaching module on systemic design as a case research study for designers and non-designers in a cross-disciplinary educational context
Benchmark Suite for Resilience Assessment of Deep Learning Models
Full text linkThe reliability assessment of systems powered by artificial intelligence (AI) is becoming a crucial step prior to their deployment in safety and mission-critical systems. Recently, many efforts have been made to develop sophisticated techniques to evaluate and improve the resilience of AI models against the occurrence of random hardware faults. However, due to the intrinsic nature of such models, the comparison of the results obtained in state-of-the-art works is crucial, as reference models are missing. Moreover, their resilience is strongly influenced by the training process, the adopted framework and data representation, and so on. To enable a common ground for future research targeting Convolutional Neural Networks (CNNs) resilience analysis/hardening, this work proposes a first benchmark suite of Deep Learning (DL) models commonly adopted in this context, providing the models, the training/test data, and the resilience-related information (fault list, coverage, etc.) that can be used as a baseline for fair comparison. To this end, this research identifies a set of axes that have an impact on the resilience and classifies some popular CNN models, in both PyTorch and TensorFlow. Some final considerations are drawn, showing the relevance of a benchmark suite tailored for the resilience context
Electrochemical measurements to support energy transition: water electrolysers characterization
Full text linkL'abstract è presente nell'allegato / the abstract is in the attachmen
The promise of transparent wood as a multifunctional energy material
No full textTransparent wood has potential not only as a sustainable substitute for glass, but also as a multifunctional energy material whose value lies in the integration of diffuse light management, thermal insulation, mechanical load bearing and sustainability. Its widespread adoption will require application-driven design, realistic durability assessments and alignment with standards
Tailoring NiO defectivity to boost the electrocatalytic activity toward nitrate reduction into ammonia
No full textAmmonia is vital for global agriculture, yet its conventional synthesis via the Haber–Bosch process is energy-intensive and environmentally burdensome, contributing ∼2% of global CO2 emissions. Simultaneously, excessive use of ammonia-based fertilizers has led to nitrate pollution in water systems. Electrochemical nitrate reduction (E-NO3RR) offers a dual solution: mitigating nitrate contamination while enabling decentralized, sustainable ammonia production. Here, we explore nickel oxide (NiO) nanoparticles as efficient, low-cost electrocatalysts for E-NO3RR, capitalizing on their earth abundance and inherent ability to suppress competing hydrogen evolution. NiO is synthesized via a scalable precipitation method using different ethanol/water solvent ratios to modulate defect density, porosity, and crystallinity. Materials-related differences are probed by thermal, structural, and spectroscopy methods. Electrochemical tests reveal that increasing ethanol content during synthesis enhances defectiveness, correlating with improved Faradaic efficiency and ammonia production rates. This work underscores the critical role of synthetic parameters in tailoring catalytic performance and positions defect-engineered NiO as a promising platform for green ammonia generation via nitrate reduction