Search Results (1,730)

Human population growth and increasing transportation demands have led to rising global tire consumption and associated waste. In response, various material and energy recovery strategies, such as pyrolysis, have been developed to produce high-value-added products such as pyrolysis oils, which can be reused as materials or fuels. However, these oils often contain heteroatom-containing compounds (e.g., nitrogen, oxygen, sulfur) that can hinder their valorization and must therefore be identified and removed. To characterize heteroatomic compounds present in distillation fractions of pyrolysis oils, GC × GC/TOFMS and GC/HRMS were employed. For non-target analysis, data processing parameters were optimized using a Central Composite Design (CCD). The most influential parameters for GC × GC/TOFMS were the minimum number of mass-to-charge ratio (m/z) signals kept in the deconvoluted spectra (minimum stick count) and peak signal-to-noise ratio (S/N), while for GC/HRMS, optimization focused on the m/z S/N threshold, peak S/N, and total ion current (TIC). Under optimal conditions, 129 and 92 heteroatomic compounds were identified via GC × GC/TOFMS and GC/HRMS, respectively, within a single distillation fraction, with 57 compounds identified using both techniques. Notably, GC × GC/TOFMS exclusively identified 72 compounds, while there were only 5 unique to GC/HRMS. These results highlight the effectiveness of GC × GC/TOFMS in characterizing heteroatomic compounds in complex mixtures, while also underlining the complementary value of GC/HRMS. Full article

Against the backdrop of increasing demands for the safety and longevity of the bridge infrastructure, this review synthesizes the recent advances in structural health monitoring (SHM) sensing systems. Carbon nanotube (CNT), piezoelectric, RFID, wireless, fiber optic, and computer-vision-based sensing are thoroughly explored and elucidated in the existing literature survey that distills their working principles, documented deployments, and anticipated research directions. CNT sensors detect minute resistance variations for strain and crack surveillance; piezoelectric devices transduce mechanical stimuli into high-resolution electrical signals; RFID tags combine location tracking with modular sensing and wireless data relay; and wireless sensing technology integrates sensor nodes with microprocessors and communication modules, which can facilitate efficient data processing and autonomous management. Fiber optic sensing technology, known for precision and interference resistance, is ideal for high-precision monitoring under strong electromagnetic interference conditions, and vision-based systems emulate human perception to extract geometric descriptors via image analytics. The comparative analysis reveals complementary strengths that guide practitioners in selecting optimal sensor suites for specific bridge conditions. The findings underscore the transformative role of these technologies in enhancing SHM reliability and suggest that synergistic integration with robotics and emerging materials will further advance future resilient monitoring frameworks. Full article

To develop a CO₂-free process for recovering Fe and Zn metals from electric arc furnace (EAF) dust, this study investigated the molten oxide electrolysis of various Fe₂O₃–ZnO mixtures in a B₂O₃–Na₂O electrolyte. Electrolysis was conducted using an Fe cathode and Pt anode at 1173 K by applying cell voltages that were determined based on thermodynamic calculations and cyclic voltammetry measurements. When electrolysis was conducted at a cell voltage of 1.1 V, the selective reduction of Fe oxide to Fe metal was observed without ZnO reduction. However, when 1.6 V was applied, the co-reduction of Fe oxide and ZnO to the Fe–Zn alloy was observed. In the vacuum distillation of the Fe–Zn alloy at 1000–1200 K, Zn metal with a purity of ≥99.996% was obtained with a recovery efficiency of ≥99.9%, under certain conditions. This study demonstrates the feasibility of recovering Fe and Zn from EAF dust using low-temperature molten oxide electrolysis and subsequent vacuum distillation. Full article

This study investigates the optimal duration for forming a uniform oxide layer and evaluates its influence on water-splitting performance. We selected a Ti₅₀Cu₃₂Ni₁₅Sn₃ amorphous ribbon, which is known to simultaneously form anatase TiO₂ and Sn oxide via a single hydrothermal process. Hydrothermal treatments were conducted at 220 °C in 150 mL of distilled water for durations of 3 and 6 h. The process successfully formed nanoscale metal oxides on the alloy surface, with the uniformity of the oxide layer increasing over time. The amorphous phase of the alloy was retained under all conditions. X-ray photoelectron spectroscopy (XPS) analysis confirmed the formation of TiO₂ and SnO_x, while Cu and Ni remained in their metallic state. Furthermore, we verified the coexistence of these oxides with metallic Ti and Sn. Photoelectrochemical analysis showed that the sample treated for 6 h exhibited the best water-splitting performance, which correlated directly with the most uniform oxide coverage. This time-controlled hydrothermal oxidation method, using only water, presents a promising and efficient approach for developing functional surfaces for electronic and photoelectrochemical applications of metallic glasses (MGs). Full article

The study undertook a comprehensive review and comparative analysis of natural language processing techniques for news article classification, with a particular focus on Java language libraries. The dataset comprised an excess of 200,000 items of news metadata sourced from The Huffington Post. The traditional algorithms based on mathematical statistics and deep machine learning were evaluated. The libraries chosen for tests were Apache OpenNLP, Stanford CoreNLP, Waikato Weka, and the Huggingface ecosystem with the Pytorch backend. The efficacy of the trained models in forecasting specific topics was evaluated, and diverse methodologies for the feature extraction and analysis of word-vector representations were explored. The study considered aspects such as hardware resource management, implementation simplicity, learning time, and the quality of the resulting model in terms of detection, and it examined a range of techniques for attribute selection, feature filtering, vector representation, and the handling of imbalanced datasets. Advanced techniques for word selection and named entity recognition were employed. The study compared different models and configurations in terms of their performance and the resources they consumed. Furthermore, it addressed the difficulties encountered when processing lengthy texts with transformer neural networks, and it presented potential solutions such as sequence truncation and segment analysis. The elevated computational cost inherent to Java-based languages may present challenges in machine learning tasks. OpenNLP model achieved 84% accuracy, Weka and CoreNLP attained 86% and 88%, respectively, and DistilBERT emerged as the top performer, with an accuracy rate of 92%. Deep learning models demonstrated superior performance, training time, and ease of implementation compared to conventional statistical algorithms. Full article

The separation of butadiene 1,3 from pyrolysis products is a critical step in the petrochemical industry, as butadiene is a key raw material for producing synthetic rubber and other polymers. This study presents a detailed model-based analysis of the separation process, focusing on optimizing operational parameters to maximize butadiene recovery, enhance product purity, and reduce energy consumption. The simulation was conducted using Aspen Plus, evaluating critical variables such as the solvent-to-feed ratio, reflux ratio, number of column stages, and energy integration between distillation units. The simulation results indicated that an optimal solvent-to-feed ratio of 1.5:1 and a reflux ratio of 4.2:1 in the extractive distillation column provided the highest separation efficiency. Under these conditions, the recovery rate of butadiene 1,3 reached 98%, with a final product purity of 99.5%. Furthermore, this study revealed that increasing the number of theoretical stages in the distillation column improved the separation process without significantly increasing energy demand. Energy integration, specifically through heat recovery between the primary distillation and extractive distillation columns, led to a 12% reduction in total energy consumption. These findings demonstrate the importance of fine-tuning operational parameters to achieve high separation efficiency and product quality while minimizing energy use. This model-based analysis provides valuable insights into the design and optimization of industrial-scale butadiene separation processes, offering strategies to reduce operational costs and improve sustainability in production. The methodology and results can serve as a basis for further improvements in similar separation processes across the petrochemical industry. Full article

The European Green Deal (EGD) aims at driving the green transition in the EU and positions cities as pivotal actors in achieving climate neutrality and environment protection. Despite ambitious policy commitments, significant implementation gaps persist at the local level impeding urban green transitions. This study assesses barriers to the EGD urban implementation by integrating several methods (scoping literature review, expert consultations, and computational network analysis) to identify structural conditions hampering change. Barriers are clustered into five domains and reviewed by experts to distill eight structural conditions perpetuating the status quo of urban development, hindering transformative change. The findings illustrate how the emerged structural conditions, ranked by their in-degree centrality, regard insufficient policy implementation; upgrade of consolidated built environments’ layout; short-term mindset; lack of knowledge and data sharing among stakeholders; silos in policymaking and development processes; competition among stakeholders over space use; limited social acceptance; and limited financial resources. Conversely, high-out-degree barriers—such as limited technical expertise in urban departments and GDP-oriented paradigms—emerge as system triggers where targeted interventions could catalyze change. This research provides actionable insights for policymakers by identifying leverage points which could promote urban green transitions and enhance the EGD local implementation for accelerating urban green transitions. Full article

Solving math word problems automatically is a critical task in the field of natural language processing. Due to the insufficient size of existing MWP datasets, recent models have reached a performance bottleneck. Large-scale and high-quality training examples are crucial for training a robust math solver, but existing high-quality datasets have limited scale, and annotating or synthesizing vast MWPs explicitly is highly expensive. To address these issues, we propose a novel hidden space-based retrieval augmentation self-distillation method, named RASD, to improve the mathematical reasoning performance of MWP solvers with semantic representation augmentation and self-distillation learning. RASD enhances problem representations by retrieving and merging similar ones. It then inputs both the original and augmented representations into the decoder for solution reasoning. A self-distillation objective is used to maintain reasoning consistency between them. Extensive experiments on five popular math word problem-solving benchmarks, including MAWPS, Math23K, ASDiv-A, SVAMP, and GeoQA, show the effectiveness and universality of our RASD on improving the math reasoning ability of multiple popular baseline solvers. Full article

Municipal sewage sludge (S.S.) and abundant olive-tree pruning on Lesvos Island present both a disposal challenge and an untapped energy resource. This study proposes and evaluates on a preliminary level an integrated system that utilizes both sewage sludge and pruning. The integrated system converts sewage sludge into Hydrochar (HC) via Hydrothermal Carbonization (HTC), removes the aqueous phase using passive solar distillation, and co-gasifies the dried HC with olive pruning in an autothermal downdraft gasifier. HTC experiments on anaerobically digested sludge produced HC with higher heating values exceeding 20 MJ kg⁻¹ while reducing the chemical oxygen demand of the process liquor. Gasification modelling, using the MAGSY equilibrium model, demonstrated that replacing up to 50% of lignocellulosic biomass with HC increased hydrogen content and the Lower Heating Value (LHV) of syngas. Mass and energy balances suggest that the system could provide approximately 590 kW of continuous power, contributing around 4720 MWh to the island’s annual electricity generation. These results indicate that combining HTC, solar distillation, and co-gasification offers a viable pathway to close waste loops, reduce landfill needs, and deliver renewable energy. Future work will focus on Aspen Plus design and optimization, along with a life-cycle assessment in order to assess the environmental benefits. Full article

Waste animal fat (WAF) can be converted to distillate fractions similar to petroleum solvents and used as solvents via pyrolysis and fractional distillation. Pyrolysis oil from triglyceride materials presents adequate viscosity and volatility, compared to petroleum fuels, but shows acid values between 60–140 mg KOH/g, impeding its direct use as biofuels without considerable purification of its distillates. Fractional distillation can be applied for the purification of bio-oil, but only a few studies accurately describe the process. The purpose of this study was to evaluate the effect of temperature in the conversion of waste animal fat into fuel-like fractions by pyrolysis and fractional distillation in a semi-batch stirred bed reactor (2 L) according to reaction time. Waste animal fat was extracted (rendering) from disposed meat cuts obtained from butcher shops and pyrolyzed in a stainless-steel stirred bed reactor operating in semi-batch mode at 400–500 °C. The obtained liquid fraction was separated according to reaction time. The pyrolysis bio-oil at 400 °C was separated into four distinct fractions (gasoline, kerosene, diesel, and heavy phase) by fractional distillation with reflux. The bio-oil and distillate fractions were analyzed by density, kinematic viscosity, acid value, and chemical composition by gas chromatography coupled to mass spectra (GC-MS). The results show that, for semi-batch reactors with no inert gas flow, higher temperature is associated with low residence time, reducing the conversion of fatty acids to hydrocarbons. The distillate fractions were tested in a common application not sensible to the fatty acid concentration as a diluent in the preparation of diluted asphalt cutback for the priming of base pavements in road construction. Kerosene and diesel fractions can be successfully applied in the preparation of asphalt cutbacks, even with a high acid value. Full article

The Centralized Teacher with Decentralized Student (CTDS) framework is a multi-agent reinforcement learning (MARL) approach that utilizes knowledge distillation within the Centralized Training with Decentralized Execution (CTDE) paradigm. In this framework, a teacher module learns optimal Q-values using global observations and distills this knowledge to a student module that operates with only local information. However, CTDS has limitations including inefficient knowledge distillation processes and performance gaps between teacher and student modules. This paper proposes the evolutionary sampling method that employs genetic algorithms to optimize selective knowledge distillation in CTDS frameworks. Our approach utilizes a selective sampling strategy that focuses on samples with large Q-value differences between teacher and student models. The genetic algorithm optimizes adaptive sampling ratios through evolutionary processes, where the chromosome represent sampling ratio sequences. This evolutionary optimization discovers optimal adaptive sampling sequences that minimize teacher–student performance gaps. Experimental validation in the $StarCraft$ Multi-Agent Challenge (SMAC) environment confirms that our method achieved superior performance compared to the existing CTDS methods. This approach addresses the inefficiency in knowledge distillation and performance gap issues while improving overall performance through the genetic algorithm. Full article

Actions related to reducing CO₂ emissions have led to the development of technologies using raw materials in the form of broadly understood biomass as CO₂-neutral fuels. There has been a rapid development of pyrolysis processes (carbonization, dry distillation) of various types of biomass toward the production of biochar for industrial applications. Particularly high hopes are associated with the use of biochar as a substitute for fossil fuel in energy-intensive sectors of the economy, especially the metallurgical and steel industries. This paper characterizes the current state and potential for biochar application, using the iron and steel industry as a case study. The analysis focuses primarily on the characteristics of biochar production and its industrial application potential. The characterization includes the diversity of biomass feedstocks, processing methods, and reactor types, the influence of operational parameters on biochar yield, as well as the properties and applications of biochar. As part of the analysis of biomass use potential in the iron and steel industry, the study reviews the current levels of coal substitution achieved at the laboratory scale and presents examples of biochar implementation in existing industrial facilities. In addition, key factors limiting the feasibility of coal substitution in the iron and steel industry are identified. The summary includes the main directions for further research aimed at increasing the use of biochar in industry. Full article

This review provides a comprehensive overview of the evolution and application of artificial intelligence (AI) and large language models (LLMs) in engineering, with a specific focus on chemical engineering. The review traces the historical development of LLMs, from early rule-based systems and statistical models like N-grams to the transformative introduction of neural networks and transformer architecture. It examines the pivotal role of models like BERT and the GPT series in advancing natural language processing and enabling sophisticated applications across various engineering disciplines. For example, GPT-3 (175B parameters) demonstrates up to 87.7% accuracy in structured information extraction, while GPT-4 introduces multimodal reasoning with estimated token limits exceeding 32k. The review synthesizes recent research on the use of LLMs in software, mechanical, civil, and electrical engineering, highlighting their impact on automation, design, and decision-making. A significant portion is dedicated to the burgeoning applications of LLMs in chemical engineering, including their use as educational tools, process simulation and modelling, reaction optimization, and molecular design. The review delves into specific case studies on distillation column and reactor design, showcasing how LLMs can assist in generating initial parameters and optimizing processes while also underscoring the necessity of validating their outputs against traditional methods. Finally, the review addresses the challenges and future considerations of integrating LLMs into engineering workflows, emphasizing the need for domain-specific adaptations, ethical guidelines, and robust validation frameworks. Full article

This paper introduces a general AI text review framework for the automated analysis of textual reviews using advanced natural language processing techniques. The framework uniquely integrates sentiment analysis, topic modeling, and abstractive summarization within a modular architecture. It leverages transformer-based models (e.g., DistilBERT and FASTopic), vector databases, and caching mechanisms to ensure scalability and real-time performance. To validate the general approach, we developed a domain-specific implementation, VisitorLens AI, which performs advanced textual analysis for Google Maps reviews of the UNESCO World Heritage Site, Kotor Fortress. We demonstrated that the designed system generates structured and actionable insights for both tourists and local authorities, and increases institutional capacity to evaluate UNESCO criteria compliance. Finally, we performed both quantitative and expert evaluations, demonstrating the high performance of our framework across NLP tasks. The outputs confirm the framework’s generalizability, robustness, and practical value across domains. Full article

The escalating scale and psychological harm of cyberbullying across digital platforms present a critical social challenge, demanding the urgent development of highly accurate and reliable automated detection systems. Standard fine-tuned transformer models, while powerful, often fall short in capturing the nuanced, context-dependent nature of online harassment. This paper introduces a novel hybrid deep learning model called Robustly Optimized Bidirectional Encoder Representations from the Transformers with the Bidirectional Long Short-Term Memory-based Attention model (RoBERTa-BiLSTM), specifically designed to address this challenge. To maximize its effectiveness, the model was systematically optimized using the Optuna framework and rigorously benchmarked against eight state-of-the-art transformer baseline models on a large cyberbullying dataset. Our proposed model achieves state-of-the-art performance, outperforming BERT-base, RoBERTa-base, RoBERTa-large, DistilBERT, ALBERT-xxlarge, XLNet-large, ELECTRA-base, DeBERTa-v3-small with an accuracy of 94.8%, precision of 96.4%, recall of 95.3%, F1-score of 95.8%, and an AUC of 98.5%. Significantly, it demonstrates a substantial improvement in F1-score over the strongest baseline and reduces critical false negative errors by 43%, all while maintaining moderate computational efficiency. Furthermore, our efficiency analysis indicates that this superior performance is achieved with a moderate computational complexity. The results validate our hypothesis that a specialized hybrid architecture, which synergizes contextual embedding with sequential processing and attention mechanism, offers a more robust and practical solution for real-world social media applications. Full article