Achieving adequate thermal comfort in classrooms in hot cities in southern Mexico is challenging. A heterogeneous distribution of air conditioning flow leads to thermal discomfort, affecting occupants’ academic performance and increasing energy consumption. This study evaluates the thermal comfort of occupants in an air conditioned classroom using computational fluid dynamics. We determined the effects of variations in air conditioning operating parameters (supply angle, velocity, and temperature) on PMV and modified PMV indices. An operating configuration of 60°, 3 m/s, and 22 °C ensures that thermal comfort remains within regulations while optimizing energy consumption, in contrast to the original PMV model. Using the modified PMV model, the values are 0.38 for students and 0.31 for the teacher, with percentages of dissatisfied individuals of 10% and 7.7%, respectively. This study demonstrates the importance of analyzing air conditioning operating parameters to enhance thermal comfort while reducing energy consumption.

This study presents an Adaptive PID controller designed to enhance temperature stability and energy performance in household refrigerator systems subject to non-stationary disturbances. Classical PID control is limited by fixed gains and the assumption of linear time-invariant dynamics, which is frequently violated by door opening, load variation, and compressor cycling. To address this issue, the proposed approach introduces a Laplace-distribution-based adaptive gain function L(t) that adjusts controller sensitivity according to the statistical rarity of the composite temperature error. The method preserves the conventional PID control structure while introducing a lightweight gain-scaling mechanism suitable for embedded implementation. Experimental validation using a commercial two-compartment refrigerator demonstrated substantial improvements in performance compared with a classical PID controller. The Adaptive PID achieved reduced temperature deviations in both compartments, significantly smoother compressor and fan actuation, and a 4.6% reduction in total energy consumption under an identical disturbance schedule. These results confirm that the proposed controller provides a practical, embedded-friendly solution that improves thermal regulation, actuator longevity, and energy efficiency under the tested disturbance schedule representative of typical household usage.

Advanced Caching Strategies for High-Throughput Large Language Model Serving

Harnessing renewable electricity to transform abundant environmental resources into fertilizers is central to sustainable development. Electrochemical nitrate-to-ammonia conversion provides a promising route, yet its efficiency is constrained by the elusive surface hydrogenation dynamics governing multi-step *NOx reduction. Here, a cooperative descriptor (Ψ) derived from large-language-models-assisted mining and energetic analysis successfully identifies NiCu single-atom alloys (SAAs) as optimal catalysts. Pulse electrodeposition delivers atomically dispersed alloys with tunable structures, achieving a maximum Faradaic efficiency (FE) of ∼95% and yield rate (YR) of ∼11.4mg h-1 cm-2. In situ surface-interrogation scanning electrochemical microscopy (SI-SECM) provides quantitative information on the time-resolved surface-active hydrogen (*H) generation-consumption and *NOx hydrogenation rate constants (NiCu>CoCu≫MnCu ≈ FeCu>Cu), directly aligning surface kinetics with selectivity. Theoretical investigations further confirmed that Ni doping lowers the barriers for *H formation and *NOx hydrogenation. A plasma-electrochemical-CO2 capture system demonstrated continuous "air-to-fertilizer" conversion with reduced energy consumption and potential net-negative emissions. These results establish a transferable design rule that bridges theoretical descriptors with operando hydrogenation dynamics, providing a mechanistic foundation and practical pathway toward scalable, zero-carbon fertilizer production.

Abstract - Heat exchangers play a crucial role in various industries, including power generation, chemical processing especially ventilation and air conditioning. Performance and efficiency of heat exchangers is observed to have dependent on various fluid flow parameters like Nusselt and Reynold numbers. In this research work, an attempt is made to investigate the performance of dimpled tube heat exchangers under different flow conditions, characterized by the Reynolds number and Nusselt Number. Results show that dimpled tubes outperform smooth tubes, with significant improvements in heat transfer rates and moderate increases in pressure drop. From tabular values and plots it was found that using spherical dimples leads to a significant increase in the heat transfer rate as compared to that of a normal tube without dimples. Also, it was seen that the change of dimple arrangement from inline to staggered arrangement enhances the heat transfer characteristics to a noticeable amount as compared to others but may further be studied for higher scale implementation with some corresponding moderations. This research has important implications for industries that rely on heat exchangers, offering a potential solution to improve their efficiency and reduce energy consumption which can aid in developing more efficient and sustainable heat transfer systems. Key Words: heat transfer coefficient (h), dimpled tube, nusselt Number (Nu), reynolds Number (Re), smooth tubes

Low-Rh/Mo2C/N-doped carbon nanofibers as robust bifunctional electrocatalyst for hydrazine oxidation-assisted H2 generation and Zn-hydrazine battery.

The relevance of this study is driven by the increasing requirements for the energy efficiency and indoor comfort of residential and public buildings, particularly in regions with extreme climatic conditions characterized by substantial daily and seasonal temperature fluctuations. Effective management of heat transfer through building envelopes has become a key factor in reducing energy consumption and improving indoor comfort. This paper presents the results of an experimental–numerical investigation of the thermal behavior of an adaptive exterior wall system with a controllable air cavity. Steady-state and transient simulations were performed for three envelope configurations: a baseline design, a design with vertical air channels, and an adaptive configuration equipped with adjustable openings. Quantitative analysis showed that during the winter period, the adaptive configuration increases the interior surface temperature by 1.5–2.3 °C compared to the baseline design, resulting in a 12–18% reduction in the specific heat flux through the wall. In the summer period, the temperature of the exterior cladding decreases by 3–5 °C relative to the baseline, which reduces heat gains by 8–14% and lowers the cooling load. Additional analysis of temperature fields demonstrated that the presence of vertical air channels has a limited effect during winter: temperature differences at the surfaces do not exceed 1 °C. A similar pattern is observed in warm periods; however, due to controlled air circulation, the adaptive configuration provides an improved thermal regime. The results confirm the effectiveness of the adaptive wall system under the climatic conditions of southern Kazakhstan, characterized by high solar radiation and large diurnal temperature variations. The practical significance of the study lies in the potential application of adaptive façades to enhance the energy efficiency of buildings during both winter and summer seasons.

Efficient exfoliation of montmorillonite (MMT) into single or few-layer nanosheets is essential for its high-value applications in gels, membrane separation, and composite reinforcement. In this study, Na + -exchanged bentonite underwent solvent-free, high-energy ball milling for 2 hours, with the ball-to-powder mass ratio as the sole variable. This method reduces the average layer thickness from >7 nm to 1-2 nm concomitantly enhancing surface charge, colloidal stability, gel viscosity, and 24 h water uptake. In contrast to conventional chemical intercalation and ultrasonic protocols, the proposed approach eliminates the need for organic solvents and costly intercalants, reduces energy consumption by more than 60%, and is characterized by simplicity, scalability, and user-friendliness. This work offers a green, low-cost pathway for the large-scale production of high-performance two-dimensional clay nanosheets and their multifunctional composites.

Dialysis therapy is a resource-intensive treatment for end-stage kidney disease that remains highly dependent on in-center hemodialysis in Japan. From both economic and environmental perspectives, it is necessary to reduce energy consumption and resource use, and minimize waste generation to achieve sustainable kidney healthcare. The clinic targeted in this study provides hemodialysis in a regional city and launched a resource-saving committee in 2008 to implement initiatives, appoint green champions, and monitor four environmental items (electricity, gas and water consumption, and waste generation) and financial effects. To retrospectively evaluate environmental impact, we calculated the carbon footprint. The median monthly consumption of electricity, gas, and water per hemodialysis patient was approximately 353 kWh, 17 m 3 , and 9 m 3 , respectively. These levels of resource consumption were nearly equivalent to those of an average Japanese household in 2022. Switching to a combination of city water and well water reduced both costs and environmental impact. However, the overall financial benefit and initial investment burden, such as for installation of light-emitting diode fixtures and developing the water supply system, were not fully investigated. The resource-saving committee appears to have mitigated both economic and environmental impacts to some extent; however, steady resource-saving efforts were accompanied by surging costs of electricity and medical waste disposal during the study period, indicative of recent general inflation in Japan. To achieve more sustainable dialysis therapy that balances environmental and health considerations, further proactive initiatives are needed to reduce resource use beyond the current scope, such as through individualized dialysate prescriptions.

Suppression of passivation on NiMoO4 microrod by ultrathin metal-organic-framework nanosheets in urea-assisted natural seawater splitting.

The rapid expansion of China’s photovoltaic (PV) industry has led to a significant increase in decommissioned PV modules. To address the high energy consumption and environmental impact of traditional recycling techniques, this study proposes a novel method that integrates DMPU solvent recycling with pyrolysis for recovering PV cell sheets. DMPU, an organic solvent with low volatility, non-toxicity, and excellent recyclability, was used in this study. The effects of temperature and treatment duration on the structural integrity of silicon cell sheets were systematically evaluated, establishing optimal parameters: immersion in DMPU at 200 °C for 60 min, followed by pyrolysis at 480 °C for 60 min. A case study was conducted on a small-scale recycling facility with a daily processing capacity of 200 standard PV panels, encompassing system boundaries such as transportation, pretreatment, and pyrolysis. The recycling process consumed 2.14 × 109 kJ of energy annually, reducing CO2 emissions by 9357.2 tons. Compared to conventional methods such as pyrolysis, mechanical dismantling, and chemical dissolution, the proposed approach employing a green, recyclable solvent markedly reduces energy consumption and carbon emissions, offering notable environmental benefits.

Thermal comfort and energy efficiency are two main goals of heating, ventilation, and air conditioning (HVAC) systems, which use about 40% of the total energy in buildings. This paper aims to predict optimal room temperature, enhance comfort, and reduce energy consumption while avoiding extra energy use from overheating or overcooling. Six Machine Learning (ML) models were tested to predict the optimal temperature in the classroom based on the occupancy characteristic detected by a Deep Learning (DL) model, You Only Look Once (YOLO). The decision tree achieved the highest accuracy at 97.36%, demonstrating its effectiveness in predicting the preferred temperature. To measure energy savings, the study used RETScreen software version 9.4 to compare intelligent temperature control with traditional operation of HVAC. Genetic algorithm (GA) was further employed to optimize HVAC energy consumption while keeping the thermal comfort level by adjusting set-points based on real-time occupancy. The GA showed how to balance comfort and efficiency, leading to better system performance. The results show that adjusting from default HVAC settings to preferred thermal comfort levels as well controlling the HVAC to work only if the room is occupied can reduce energy consumption and costs by approximately 76%, highlighting the substantial impact of even simple operational adjustments. Further improvements achieved through GA-optimized temperature settings provide additional savings of around 7% relative to preferred comfort levels, demonstrating the value of computational optimization techniques in fine-tuning building performance. These results show that intelligent, data-driven HVAC control can improve comfort, save energy, lower costs, and support sustainability in buildings.

The use of sustainable composite building materials is essential for developing infrastructure that benefits the environment while reducing energy consumption [...]

Conventional clay brick manufacturing is highly resource-intensive, relying on virgin clay, large volumes of process water, and high fossil-fuel-driven firing temperatures (900-1200 °C), posing significant environmental pressures. This study introduces a methodological innovation by valorizing refinery oily sludge (ROS), a hazardous petroleum byproduct, to achieve simultaneous reductions in raw material, water, and energy consumption. Clay mixtures containing 0, 5, and 10 wt % ROS were extruded and fired at 950 and 1050 °C. The optimal 5 wt % ROS substitution reduced process water demand by 25% and firing energy consumption by over 30%, while nearly doubling brick production yield from 5.6 to 9.5 units/kg of clay. Minor color changes and efflorescence were observed, but the mechanical integrity remained unaffected. These findings demonstrate a scalable pathway for the ceramic industry to transform hazardous waste into a valuable resource, achieving integrated environmental and resource savings in line with circular economy and sustainability principles.

Electrokinetic bioremediation (EK-BIO) shows significant promise for organochlorine remediation in low-permeability matrices. However, its in situ application under site-relevant scenarios remains a substantial challenge. This study establishes a comprehensive EK-BIO lab-to-field framework that includes laboratory batch experiments, array optimization, pilot-scale field validation, and life cycle assessment. Primarily, batch and column experiments optimized both the additive dosage strategy and electrode array configuration, favoring a 6-day preinoculation of the niche-preparing culture and a unidirectional one-dimensional electrode setup. Guided by these findings, the 98-day EK-BIO pilot experiment achieved over 90.0% TCE removal, with a 74.0% chloroethylene-to-ethylene conversion efficiency. Microbial community analyses further revealed a notable increase in the relative abundance of putative organohalide-respiring bacteria in the EK-BIO, approximately 25.6% and 34.3% higher than in bioaugmentation and electrokinetic treatments, respectively. Additionally, life cycle assessment results underscored the advantages of EK-BIO over conventional thermal remediation alternatives, with reductions in carbon emissions, energy consumption, and remediation costs. This study validated the feasibility and reliability of EK-BIO technology, supporting its advancement for the in situ remediation of organochlorine-contaminated sites.

To optimize resource utilization of bulk industrial solid waste and promote low-carbon production in the cement industry, the belite-ye’elimite cement (BSAC) clinker was prepared entirely from solid waste materials (carbide ash, coal gangue, steel slag, desulfurization gypsum). The mineral formation process of the clinker was systematically analyzed. Findings reveal that within the temperature range of 700-1100 ºC, the clinker primarily comprises transition minerals such as aluminosilicate, aluminate, calcium sulphosilicate. At 1100-1200 ºC, these phases gradually convert to dicalcium silicate (C2S) and ye’elimite (C4A3Š). Above 1200 ºC, changes in the minerals correlate to the structural and morphological evolution of C2S and C4A3Š crystals. However, beyond 1320 ºC, C2S exhibit an irregular morphology, and C4A3Š decomposes. BSAC prepared at 1260 °C met the mechanical requirements for 52.5 grade sulphoaluminate cement. Environmentally, BSAC clinker reduces energy consumption by 18% and carbon emissions by 65% compared to traditional Portland cement.

This paper proposes a dynamic metric-constrained optimization algorithm for edge-cloud collaborative environments with multi-dimensional situation verification under federated SOA services. The framework integrates adaptive metric encoding, federated optimization, and constraint verification to ensure robust and efficient service orchestration. Experimental evaluations on Google, Alibaba, and Azure cluster traces demonstrate the superiority of the proposed approach over HEFT, NSGA-II, and MOEA/D. Specifically, latency violations were reduced to 5%, while other constraint breaches were maintained below 7%. SLA satisfaction consistently exceeded 88% across diverse stress conditions, peaking at 91% under node churn. Furthermore, the algorithm achieved 170 ms latency, 92.7% reliability, and 75.6 Mbps throughput with reduced energy consumption of 92.3 J, outperforming static baselines. Runtime overhead was limited to 9.6 ms with 95 KB communication per round, enabling 5,200 decisions per second. These results confirm that the proposed framework achieves a practical balance between adaptability, efficiency, and robustness, making it suitable for deployment in dynamic edge-cloud systems.

This paper presents the design and analysis of an innovative solar-powered water supply system for multi-storey residential buildings under conditions of high solar insolation and limited energy resources. Instead of the conventional approach where each apartment uses an individual low-power pump, a centralized architecture is proposed based on surface multistage pumps powered directly by DC photovoltaic modules. This enables operation without grid electricity and significantly reduces operating costs. Reliability is ensured through a duty-rotation strategy and ???? + 1 redundancy: two pumps operate simultaneously while one remains on hot standby. Water is stored in rooftop tanks with a total volume of 15–20 m3, providing stable supply during short-term fluctuations in solar irradiance. The mathematical model is formulated as a Boolean programming problem that captures pump on/off logic and discrete constraints on tank filling levels; its objective function combines energy consumption, annual operating costs, and a reliability metric, enabling multi-objective optimization. A numerical experiment for a four-storey building in Jizzakh with a daily demand of 34 m3 shows that the proposed system can reduce energy consumption by more than 95%, cut annual costs from 11.6 million to less than 0.3 million UZS, and improve reliability due to optimal pump rotation. The approach has strong practical value and can be adapted to diverse operating conditions, opening prospects for broad deployment in municipal infrastructure across regions with developing energy systems.

The rapid expansion of AI models has intensified concerns over energy consumption. Analog in-memory computing with resistive memory offers a promising, energy-efficient alternative, yet its practical deployment is hindered by programming challenges and device non-idealities. Here, we propose a software-hardware co-design that trains randomly weighted resistive-memory neural networks via edge-pruning topology optimization. Software-wise, we tailor the network topology to extract high-performing sub-networks without precise weight tuning, enhancing robustness to device variations and reducing programming overhead. Hardware-wise, we harness the intrinsic stochasticity of resistive-memory electroforming to generate large-scale, low-cost random weights. Implemented on a 40 nm resistive memory chip, our co-design yields accuracy improvements of 17.3% and 19.9% on Fashion-MNIST and Spoken Digit, respectively, and a 9.8% precision-recall AUC improvement on DRIVE, while reducing energy consumption by 78.3%, 67.9%, and 99.7%. We further demonstrate broad applicability across analog memory technologies and scalability to ResNet-50 on ImageNet-100.

The paper examines the theoretical and practical aspects of rod-structured matrix filters with double-periodic ar-rangement of elements, commonly employed in high-gradient magnetic separation technologies, with particular atten-tion to comparative analysis and matrix optimization. An increasing number of studies on the development of highly efficient and lightweight high-gradient systems with reduced energy consumption is noted. The aim of the article is to substantiate and develop a method for optimizing matrix parameters according to the criterion of minimizing the spe-cific energy of the magnetic field in the extraction zone, based on the calculation of local and effective force and energy characteristics of the magnetic field. The efficiency and universality of the proposed method are confirmed by a series of computational experiments with comprehensive consideration of factors influencing the quality of the final product. Specific examples demonstrate that the formation of an array of constant-magnetic-force lines (isodynes) serves as the principal means of investigating the extraction capacity of a matrix. A simple and effective approach to visualizing potential extraction zones is proposed in order to simplify the calculation of their areas. The application of the integral equation method with respect to the magnetization vector of matrix elements is substantiated, as it ensures maximum universality, simplicity, and accuracy in the analysis of complex double-periodic structures. The necessity of determin-ing not only local but also effective magnetic field parameters in energy optimization is demonstrated. The dependence of matrix efficiency on the magnetic field intensity, rod shape and concentration, and their mutual arrangement is illus-trated. The developed method is emphasized as being of a theoretical nature and is proposed as an effective comple-ment to experimental analysis methods. It is shown that practical implementation of the method requires consideration of technological characteristics and constraints. Its value lies in the completeness and reliability of the additional in-formation obtained on the basis of operational experience and high-quality experimental studies. References 20, figures 3, table 1.