Centre for High Performance Computing

ABSTRACT Polymer composites show great application potential in the field of tribology, yet existing studies mainly focus on dry friction conditions with limited exploration of microscopic mechanisms at the atomic scale. This study systematically evaluates the tribological properties and underlying micro‐mechanisms of short carbon fibers (SCF)‐reinforced polyphthalazine ether ketone (PPEK) composites under different lubrication conditions. Experimental results demonstrate that the addition of 15 wt% SCF reduces the friction coefficient and wear rate of PPEK by 42.25% and 94.79% respectively under dry friction. Tribological performance is significantly influenced by lubrication conditions, with optimal performance observed under oil lubrication. A three‐layer Fe‐polymer‐Fe friction pair model was constructed for molecular dynamics (MD) simulations, revealing that SCF decreases atomic concentration at the friction interface by 4.56%, reduces interface temperature by 45 K, and lowers interfacial interaction energy by 20.59%, thereby inhibiting adhesive wear. This work provides critical references for expanding the application of SCF/PPEK composites in marine and aerospace fields by elucidating the micro‐mechanisms of SCF‐reinforced tribological enhancement at the atomic scale.

Abstract With the development of intelligent mining, unmanned electric shovels (UES) have become a research focus as essential equipment at the mining source. Excavation trajectory planning is a core issue in UES research, directly affecting excavation efficiency and energy consumption. To address the strong structural-control coupling, high inertia, and complex working conditions of UES, this article proposes a multiobjective cooperative planning framework for excavation trajectories based on a 5-7-5 segmented variable-order polynomial (575-SVP). This framework enhances comprehensive operational performance while ensuring trajectory smoothness and motion stability. The research framework first constructs a trajectory representation model using the 575-SVP and establishes a dynamical model of the UES's working device based on the Lagrangian function to calculate excavation volume and energy consumption. On this basis, a multiobjective optimization function is formulated, considering practical operational performance and geometric constraints, with total excavation time, energy consumption per unit volume, and standard dipper fill factor as optimization targets. By applying the particle swarm optimization (PSO) algorithm to optimize trajectory control parameters, an excavation trajectory with optimal comprehensive operational performance is generated. Experimental results show that, compared to the existing sixth-degree polynomial (6-D) and 3-5-3 segmented variable-order polynomial (353-SVP) models, the proposed 575-SVP model demonstrates superior comprehensive operational performance while effectively mitigating flexible impacts during the excavation process.

This paper proposes a simple tiled approach to writing parallel algorithms. It is both easy to grasp by novice programmers, it clearly out-competes the traditional and much more complex blocking methods used in dense linear algebra. Its distribution-agnostic nature and the use of lazy and eager methods enables students to easily explore parallel program performance and understand parallel computing concepts. Indeed this approach is suitable for parallel programmers of all skill levels provided the algorithm of interest lends itself to the tiled approach.

Kearney, C, Coughlan, EK, O'Connell, A, Lacey, S, and Burns, C. Bigger but are they better? Investigating growth, maturation, and physical performance metrics in underage Gaelic games development squads. J Strength Cond Res 39(6): e806-e814, 2025-The purpose of this study was to examine (a) maturation status distribution across intercounty underage Gaelic games development squads and (b) differences in physical performance metrics relative to maturation status. The study is the first to assess physical performance metrics in intercounty development squad Gaelic Games players, with reference to maturation status. The Khamis & Roche method was used to establish maturation status in 254 male U15 and U16 Gaelic Games development squad players. Tests for lower body power (countermovement jump [CMJ] height), linear speed (10, 20 m), upper body strength (maximum chin-up test), and aerobic endurance (GAA modified Bronco) were conducted. One-way multivariate analysis of variance and analysis of variance were conducted, with an alpha level of p < 0.05 set to determine statistical significance. In both U15 and U16 groups, early developers (EDs) comprised 64.9 and 64.0%, respectively, while on-time (OT) developers comprised 31.5% (U15) and 33.7% (U16), and late developers (LDs) comprised 3.6% (U15) and 2.3% (U16). For U15, ED exhibited significantly faster 10 m speed than LD ( p = 0.045) and faster 20 m speed than both OT ( p = 0.007) and LD ( p = 0.006). After post hoc tests for U16, CMJ scores showed no significant differences ( p > 0.05), while U16 ED and OT displayed faster 20 m speed than LD ( p = 0.023, p = 0.024, respectively). Coach education around growth and maturation and strategies such as biobanding should be used in talent development settings. Practitioners should interpret speed times relative to maturation status as opposed to chronological age.

With the growth of 3D sensing technology, the deep learning system for 3D point clouds has become increasingly important, especially in applications such as autonomous vehicles where safety is a primary concern. However, there are growing concerns about the reliability of these systems when they encounter noisy point clouds, either occurring naturally or introduced with malicious intent. This paper highlights the challenges of point cloud classification posed by various forms of noise, from simple background noise to malicious adversarial/backdoor attacks that can intentionally skew model predictions. While there's an urgent need for optimized point cloud denoising, current point outlier removal approaches, an essential step for denoising, rely heavily on handcrafted strategies and are not adapted for higher-level tasks, such as classification. To address this issue, we introduce an innovative point outlier cleansing method that harnesses the power of downstream classification models. Using gradient-based attribution analysis, we define a novel concept: point risk. Drawing inspiration from tail risk minimization in finance, we recast the outlier removal process as an optimization problem, named PointCVaR. Extensive experiments show that our proposed technique not only robustly filters diverse point cloud outliers but also consistently and significantly enhances existing robust methods for point cloud classification. A notable feature of our approach is its effectiveness in defending against the latest threat of backdoor attacks in point clouds.

The ollie is the base aerial human-board maneuver, foundational to most modern skateboarding tricks.We formulate and solve an optimal control problem of a two-dimensional simplified human model and a rigid body skateboard with the objective of maximizing the height of the ollie.Our solution simultaneously discovers realistic human-applied force trajectories and optimal board geometry.We accomplish this with a direct collocation formulation using a null seed initial guess by carefully modeling the discontinuous aspects of board-ground impact and foot-board friction.This leads to efficient and robust solutions that are 10 times more computationally efficient than prior work on similar problems.The solutions show that ollie height can increase 3% by decreasing the wheelbase and that a smaller board with a back-foot dominated force strategy can give 12% higher ollies.Our model can be used to inform jump strategy and the effects of changes to the essential board geometry.

Current structural MRI‐based brain age estimates and their difference from chronological age—the brain age gap (BAG)—are limited to late‐stage pathological brain‐tissue changes. The addition of physiological MRI features may detect early‐stage pathological brain alterations and improve brain age prediction. This study investigated the optimal combination of structural and physiological arterial spin labelling (ASL) image features and algorithms. Healthy participants (n = 341, age 59.7 ± 14.8 years) were scanned at baseline and after 1.7 ± 0.5 years follow‐up (n = 248, mean age 62.4 ± 13.3 years). From 3 T MRI, structural (T1w and FLAIR) volumetric ROI and physiological (ASL) cerebral blood flow (CBF) and spatial coefficient of variation ROI features were constructed. Multiple combinations of features and machine learning algorithms were evaluated using the Mean Absolute Error (MAE). From the best model, longitudinal BAG repeatability and feature importance were assessed. The ElasticNetCV algorithm using T1w + FLAIR+ASL performed best (MAE = 5.0 ± 0.3 years), and better compared with using T1w + FLAIR (MAE = 6.0 ± 0.4 years, p < .01). The three most important features were, in descending order, GM CBF, GM/ICV, and WM CBF. Average baseline and follow‐up BAGs were similar (−1.5 ± 6.3 and − 1.1 ± 6.4 years respectively, ICC = 0.85, 95% CI: 0.8–0.9, p = .16). The addition of ASL features to structural brain age, combined with the ElasticNetCV algorithm, improved brain age prediction the most, and performed best in a cross‐sectional and repeatability comparison. These findings encourage future studies to explore the value of ASL in brain age in various pathologies.

When operating gas compressor stations during heavy snowfalls, uncontrolled formation of ice build-up on the filters of air-purifying devices (ACDs) can occur, which leads to accidents and downtime. Previously, it was proposed to use a gas-dynamic system for cleaning gas-permeable deformable surfaces (GPDS) to clean HEU filters. An experimental setup has been created. The paper discusses numerical modeling of ice destruction on the gas turbine engine. To destroy ice, the energy of a gas-dynamic jet is used, modeled as a short-term pulse. The study is carried out in a 3D setting, taking into account the suction flow pressure and the interaction of the deformable substrate with the ice crust. The results obtained were verified. The research is carried out using ANSYS Explicit Dynamics on the high-performance computing complex of PNRPU. The main results on the influence of a gas-dynamic jet on the cleaning ability of a gas turbine engine are presented

Sustainable research on computational models of neuronal networks requires published models to be understandable, reproducible, and extendable. Missing details or ambiguities about mathematical concepts and assumptions, algorithmic implementations, or parameterizations hinder progress. Such flaws are unfortunately frequent and one reason is a lack of readily applicable standards and tools for model description. Our work aims to advance complete and concise descriptions of network connectivity but also to guide the implementation of connection routines in simulation software and neuromorphic hardware systems. We first review models made available by the computational neuroscience community in the repositories ModelDB and Open Source Brain, and investigate the corresponding connectivity structures and their descriptions in both manuscript and code. The review comprises the connectivity of networks with diverse levels of neuroanatomical detail and exposes how connectivity is abstracted in existing description languages and simulator interfaces. We find that a substantial proportion of the published descriptions of connectivity is ambiguous. Based on this review, we derive a set of connectivity concepts for deterministically and probabilistically connected networks and also address networks embedded in metric space. Beside these mathematical and textual guidelines, we propose a unified graphical notation for network diagrams to facilitate an intuitive understanding of network properties. Examples of representative network models demonstrate the practical use of the ideas. We hope that the proposed standardizations will contribute to unambiguous descriptions and reproducible implementations of neuronal network connectivity in computational neuroscience.

Heart myocardia are critical to the facilitation of heart pumping and blood circulating around the body. The biaxial mechanical testing of the left ventricle (LV) has been extensively utilised to build the computational model of the whole heart with little importance given to the unique mechanical properties of the right ventricle (RV) and cardiac septum (SPW). Most of those studies focussed on the LV of the heart and then applied the obtained characteristics with a few modifications to the right side of the heart. However, the assumption that the LV characteristics applies to the RV has been contested over time with the realisation that the right side of the heart possesses its own unique mechanical properties that are widely distinct from that of the left side of the heart. This paper evaluates the passive mechanical property differences in the three main walls of the rat heart based on biaxial tensile test data. Fifteen mature Wistar rats weighing 225 ± 25 g were euthanised by inhalation of 5% halothane. The hearts were excised after which all the top chambers comprising the two atria, pulmonary and vena cava trunks, aorta, and valves were all dissected out. Then, 5 × 5 mm sections from the middle of each wall were carefully dissected with a surgical knife to avoid overly pre-straining the specimens. The specimens were subjected to tensile testing. The elastic moduli, peak stresses in the toe region and stresses at 40% strain, anisotropy indices, as well as the stored strain energy in the toe and linear region of up to 40% strain were used for statistical significance tests. The main findings of this study are: (1) LV and SPW tissues have relatively shorter toe regions of 10–15% strain as compared to RV tissue, whose toe region extends up to twice as much as that; (2) LV tissues have a higher strain energy storage in the linear region despite being lower in stiffness than the RV; and (3) the SPW has the highest strain energy storage along both directions, which might be directly related to its high level of anisotropy. These findings, though for a specific animal species at similar age and around the same body mass, emphasise the importance of the application of wall-specific material parameters to obtain accurate ventricular hyperelastic models. The findings further enhance our understanding of the desired mechanical behaviour of the different ventricle walls.