Search Results (20,617)

This paper explores the feasibility and implications of developing a privacy-preserving, data-driven cloud service for predicting the energy consumption of industrial robots. Using machine learning, we evaluated three neural network architectures—dense, LSTM, and convolutional–LSTM hybrids—to model energy usage based on robot trajectory data. Our results show that models incorporating manually engineered features (angles, velocities, and accelerations) significantly improve prediction accuracy. To ensure secure collaboration in industrial environments where data confidentiality is critical, we integrate privacy-preserving machine learning (ppML) techniques based on secure multi-party computation (SMPC). This allows energy inference to be performed without exposing proprietary model weights or confidential input trajectories. We analyze the performance impact of SMPC on different network types and evaluate two optimization strategies, using public model weights through permutation and evaluating activation functions in plaintext, to reduce inference overhead. The results highlight that network architecture plays a larger role in encrypted inference efficiency than feature dimensionality, with dense networks being the most SMPC-efficient. In addition to model development, we identify and discuss specific stages in the MLOps workflow—particularly model serving and monitoring—that require adaptation to support ppML. These insights are useful for integrating ppML into modern machine learning pipelines. Full article

Engineered landfill covers are vital for environmental sustainability. This study investigates the shear behaviors of geocomposite drainage (GCD) and geomembrane (GM) interfaces—smooth (GMS), impinged texture (GMTI), and embossed texture (GMTE)—under 10, 30, and 50 kPa of normal stress and 0%, 50%, and 100% moisture levels using large-scale direct shear tests. All interfaces showed strain-softening behavior. At 50 kPa and 0% moisture, GCD–GMTI had the highest peak strength (28 kPa), whereas GCD–GMS had the lowest (10 kPa) at 100% moisture. Moisture and normal stress showed a coupling effect, reducing strength and friction angle. At a 0% moisture level, the strength of the GCD–GMS and GCD–GMTI interfaces under 50 kPa of normal stress was 500% and 250% of that at 10 kPa, respectively; at a 100% moisture level, these proportions decreased to 310% and 230%, respectively. For GCD–GMTE, the ratio slightly increased from 3.0 to 3.2, indicating better wet performance. Texture significantly affected strength: peak strength at 50 kPa was reduced by 41% (GCD–GMS), 16% (GCD–GMTI), and 26% (GCD–GMTE) as moisture increased from 0% to 100%. Large displacement (LD)-to-peak ratios were 0.8–0.9 (GCD–GMS), 0.7–0.8 (GCD–GMTI), and up to 1.0 (GCD–GMTE). Friction angles were reduced from 18° to 9°, 23° to 18°, and 18° to 14° for GCD–GMS, GCD–GMTI, and GCD–GMTE, respectively. Vertical deformation was <0.6 mm. Shear mechanisms depended on texture and moisture. Microscopic and 3D scans revealed moisture-induced GMTI smoothing, reducing interlocking and strength. Full article

The wide application of light emitting diodes (LEDs) in lighting systems has necessitated the inclusion of spectral tunability by using multi-color LED chips. Since the lighting requirement depends on the specific application, it is very important to have flexibility in terms of the driving conditions. While many applications use single or rather white color, some recent applications require multi-spectral lighting systems especially for agricultural or human-medical treatment applications. These systems are underexplored and pose specific challenges. In this paper, a mixture of red, green, blue, white (RGB-W) LED chips was used to develop a compact light engine specifically for agricultural applications. A computational study was performed to understand the optical distribution. Later, attention was turned into development of prototype light engines followed by experimental validation for both the thermal and optical characteristics. Each LED string was driven separately at different current levels enabling an option for obtaining an infinite number of colors for numerous applications. Each LED string on the developed light engine was driven at 300 mA, 500 mA, 700 mA, and 900 mA current levels, and the optical and thermal parameters were recorded simultaneously. A set of computational models and an experimental study were performed to understand the optical and thermal characteristics simultaneously. Full article

This paper analyzes the long-term thermal effect of newly applied fiber-reinforced concrete link slabs on an existing steel bridge for a rehabilitation project of the Maryland Transportation Authority. To enhance structural resilience, thermal movement is one of the major concerns in concrete link slab design. To accommodate the global thermal expansion of a full bridge, the existing fixed bearings were modified to expansion bearings to release the longitudinal thermal movement of the super-structure. Their movements were measured by the installed LVDT devices. In this pilot study for the Maryland Transportation Authority (MDTA), engineered cementitious composite (ECC) and ultra-high-performance concrete (UHPC) were selected as candidate materials for link slabs to replace traditional steel expansion joints. To evaluate the performances of ECC and UHPC, built-in strain gauges were implemented for long-term field monitoring to ensure the durability of link slabs. For comparison, the measured data were collected over two full years to study their thermal effects in order to evaluate their sustainability. The novelty of the study is in comparing the performance of different materials side-by-side using true sensor measurements and numerical simulation. Thermal movement performance, including thermal cracking, will be one of the major selection criteria for the link slab material. Full article

High-entropy alloys (HEAs) exhibit superior properties for extreme environments, yet the effects of laser scanning speed on the microstructure and performance of laser-clad NiAlNbTiV HEA coatings remain unclear. This study systematically investigates NiAlNbTiV coatings on 316 stainless steel fabricated at scanning speeds of 800–1100 mm/min via laser cladding. Characterizations via XRD, SEM/EDS, microhardness testing, high-temperature wear testing, and electrochemical measurements reveal that increasing scanning speed enhances the cooling rate, promoting γ-(Ni, Fe) solid solution formation, intensifying TiV peaks, and reducing Fe-Nb intermetallics. Higher speeds refine grains and needle-like crystal distributions but introduce point defects and cracks at 1100 mm/min. Microhardness decreases from 606.2 HV (800 mm/min) to 522.4 HV (1100 mm/min). The 800 mm/min coating shows optimal wear resistance (wear volume: 0.0117 mm³) due to dense eutectic hard phases, while higher speeds degrade wear performance via increased defects. Corrosion resistance follows a non-linear trend, with the 900 mm/min coating achieving the lowest corrosion current density (1.656 μA·cm⁻²) due to fine grains and minimal defects. This work provides parametric optimization guidance for laser-clad HEA coatings in extreme-condition engineering applications. Full article

As internal combustion engines (ICEs) remain dominant in maritime transport, reducing their greenhouse gas (GHG) emissions is critical to meeting IMO’s decarbonization targets. Ammonia (NH₃) has gained attention as a carbon-free fuel due to its zero CO₂ emissions and high hydrogen density. However, its low flame speed and high ignition temperature pose combustion challenges. This study investigates the combustion and emission performance of NH₃-diesel dual-fuel engines, applying Response Surface Methodology (RSM) for multi-objective optimization of key operating parameters: ammonia fraction (AF: 0–30%), engine speed (1200–1600 rpm), and altitude (0–2000 m). Experimental results reveal that increasing AF led to a reduction in Brake Thermal Efficiency (BTE) from 39.2% to 37.4%, while significantly decreasing NOₓ emissions by 82%, Total hydrocarbon emissions (THC) by 61%, and CO₂ emissions by 36%. However, the ignition delay increased from 8.2 to 10.8 crank angle degrees (CAD) and unburned NH₃ exceeded 6500 ppm, indicating higher incomplete combustion risks at high AF. Analysis of variance (ANOVA) confirmed AF as the most influential factor, contributing up to 82.3% of the variability in unburned NH₃ and 53.6% in NOₓ. The optimal operating point, identified via desirability analysis, was 20% AF at 1200 rpm and sea level altitude, achieving a BTE of 37.4%, NOₓ of 457 ppm, and unburned NH₃ of 6386 ppm with a desirability index of 0.614. These findings suggest that controlled NH₃ addition, combined with proper speed tuning, can significantly reduce emissions while maintaining engine efficiency in dual-fuel configurations. Full article

In order to improve the localization accuracy and model interpretability of single-phase ground fault sections in distribution networks, a knowledge-integrated and data-driven fault localization model is proposed. The model transforms the transient zero-sequence currents into Markov Transition Field (MTF) images based on transfer entropy, and improves the two-channel feature expression with both causal and temporal structures. On this basis, a knowledge guidance mechanism based on a physical mechanism is introduced to focus on the waveform backpropagation characteristics of upstream and downstream nodes of the fault through the feature attention module, and a similarity weighting strategy is constructed by integrating the Hausdorff distance in the all-connectivity layer in order to enhance the model’s capability of discriminating between the key segments. The dataset is constructed in an improved IEEE 14-node simulation system, and the effectiveness of the proposed method is verified by t-SNE feature visualization, comparison experiments with different parameters, misclassification correction analysis, and anti-noise performance evaluation. For misclassified sample datasets, this method achieves an accuracy rate of 99.53%, indicating that it outperforms traditional convolutional neural network models in terms of fault section localization accuracy, generalization capability, and noise robustness. Research shows that the deep integration of knowledge and data can significantly enhance the model’s discriminative ability and engineering practicality, providing new insights for the construction of intelligent power systems with explainability. Full article

Fraudulent online payment operations represent a persistent challenge in digital commerce, particularly in sectors like air travel, where credit and debit card payments dominate. This study presents a novel fraud detection framework tailored to airline ticket purchases, combining a synthetic dataset generator with a modular, customizable feature engineering process. These are two machine learning models—support vector machines (SVMs) and the light gradient boosting machine (LightGBM)—for real-time fraud detection. A synthetic dataset was generated, including a rich set of engineered features reflecting realistic user, transaction, and flight-related attributes. While both models were evaluated using classification-evaluation metrics, LightGBM outperformed SVMs in terms of overall performance with an accuracy of 94.2% and a recall of 71.3% for fraudulent cases. The main contribution of this study is the design of a reusable, customizable feature engineering framework for fraud detection in the airline sector, along with the development of a lightweight, adaptable fraud detection system for merchants, especially small and medium-sized enterprises. These findings support the use of advanced machine learning methods to enhance security in digital airline transactions. Full article

The aim of this study was to analyse the relationship between sustainability and customer lifetime value (CLV) through the mediation–moderation effect of search engine optimization (SEO) metrics of websites. We obtained a data sample of 296 European sustainable firms from both industrial and technological industries. Based on the theory of source credibility, the firm’s official website, where SEO techniques are applied, is more credible regarding its sustainability activities than other sources such as social media, paid advertising, etc. As a result, we show that sustainability is a precursor of financial performance over time in sustainable firms, represented by CLV. Furthermore, we found that the value of the moderating variable, website traffic, alters the indirect effects produced by the mediating variable called website relevance (domain authority), thereby demonstrating a moderated mediation effect. The contribution of this research to the body of literature is twofold. First, it deepens the understanding of how sustainability predicts marketing outcomes based on both digital and customer metrics over time. Second, we rely on recent literature on prediction-oriented modelling (PLS-SEM) to support that it is not suitable for estimation by reflective measurement models due to the woozle effect. Full article

Despite increased focus on resilient planning and construction design in areas prone to wildfire impacts, recent research has found inconsistent approaches, a lack of evidence-based performance criteria, and limited suitable code-based verification methods for use in wildfire contexts. These limitations serve to reduce the potential effectiveness of measures intended to improve wildfire community and build resilience. The lack of suitable verification methods is particularly problematic in Australia, where complex building code requirements associated with enhanced wildfire resilience have been extended to hospitals, child care facilities, schools, and other assembly buildings. To address this issue, this paper proposes the Wildfire Expected Risk to Life and Property (WERLP) verification method. As a holistic absolute probabilistic verification method, WERLP can be applied to both building and urban design contexts within the Australian jurisdiction. The application of WERLP is demonstrated using the case study of a new hospital development. Full article

This study introduces novel semi-active vibration controllers for magnetorheological (MR) fluid-based vibration control systems, specifically a band-pass frequency-shaped semi-active control (FSSC) and a narrow-band FSSC. These algorithms are designed without requiring an accurate damper model or system identification for control current input. Unlike active controllers, the FSSC algorithms treat the MR damper as a semi-active dissipative device, and their control signal is a control current, not a control force. The performance of both FSSC algorithms is evaluated through simulation using a single-degree-of-freedom (SDOF) MR fluid-based engine mount system. A comparative analysis with the classical semi-active skyhook control demonstrates the advantages of the proposed FSSC algorithms. Full article

This paper presents a hybrid computational architecture for efficient and robust digital transmission inspired by helical genetic structures. The proposed system integrates advanced modulation schemes, such as multi-pulse-position modulation (MPPM), high-order quadrature amplitude modulation (QAM), and chirp spread spectrum (CSS), along with Reed–Solomon error correction and quantum-assisted search, to optimize performance in noisy and non-line-of-sight (NLOS) optical environments, including VLC channels modeled with log-normal fading. Through mathematical modeling and simulation, we demonstrate that the number of helical transmissions required for genome-scale data can be drastically reduced—up to 95% when using parallel strands and high-order modulation. The trade-off between redundancy, spectral efficiency, and error resilience is quantified across several configurations. Furthermore, we compare classical genetic algorithms and Grover’s quantum search algorithm, highlighting the potential of quantum computing in accelerating decision-making and data encoding. These results contribute to the field of operations research and supply chain communication by offering a scalable, energy-efficient framework for data transmission in distributed systems, such as logistics networks, smart sensing platforms, and industrial monitoring systems. The proposed architecture aligns with the goals of advanced computational modeling and optimization in engineering and operations management. Full article

The stroke control of a hydrostatic transmission (HST) is crucial for improving the energy efficiency and power variability of heavy-duty vehicles, including agricultural, construction, mining, and forestry equipment. This study introduces a new control strategy: an Artificial Neural Network (ANN) controller that imitates a Multiple Model Predictive Controller (MPC). The goal is to compare their performance in controlling the HST’s stroke. The proposed controller is designed to track complex stroke reference trajectories for both primary and secondary regulations under realistic disturbances, such as engine and load torques, which are influenced by soil and road conditions for an HST system in line with a nonlinear and time-varying mathematical model. Processor-in-the-Loop simulations suggest that the ANN controller holds several advantages over the Multiple MPC and classical control strategies. These benefits include its suitability for multi-input–multi-output systems, its insensitivity to external stochastic disturbances (like white noise), and its robustness against modeling errors and uncertainties, making it a promising option for real-time HST implementation and better than the Multiple MPC scheme in terms of simplicity and computational cost-effectiveness. Full article

Inspired by natural gradient structures observed in biological systems such as lobster exoskeletons and bamboo, this study proposes a biomimetic strategy for developing advanced automotive materials that achieve an optimal balance between strength and ductility. Against this backdrop, the present work systematically reviews the design principles underlying natural gradient structures and examines the advantages and limitations of current additive manufacturing—specifically selective laser melting (AM-SLM)—as well as conventional forming and machining processes, in fabricating nature-inspired architectures. The research systematically explores hierarchical gradient designs which endow materials with superior mechanical properties, including enhanced strength, stiffness, and energy absorption capabilities. Two primary strengthening mechanisms—hetero-deformation-induced (HDI) hardening and precipitation hardening—were employed to overcome the conventional strength–ductility trade-off. Gradient-structured materials were fabricated using selective laser melting, and microstructural analyses demonstrated that controlled interface zones and tailored precipitation distribution critically influence property improvements. Based on these findings, an integrated material design strategy combining nature-inspired gradient architectures with post-processing treatments is presented, providing a versatile methodology to meet the specific performance requirements of automotive components. Overall, this work offers new insights for developing next-generation lightweight structural materials with exceptional ductility and damage tolerance and establishes a framework for translating bioinspired concepts into practical engineering solutions. Full article