Hollow thermoelectric refrigerator with ultralow power consumption for personal thermal management

This research was carried out to reduce the energy associated with process of cooling solvents for polymerization in Polyethylene (PE) plant. Cooling of cyclohexane to 10oC in a PE plant increased power consumption to 413.3kW and the optimization reduced the power compressor power input by 32.1 percent. Optimization was carried out with the use of Aspen HYSYS version 11.0, where Peng Robinson was used as the fluid package. To optimize, double stage centrifugal compressor was used. Optimum compression ratio of 1.73 and a corresponding intermediate pressure of 800kPa were obtained. A case study was made using variable First stage discharge pressure to determine compression ratio at each stage. A plot of compression ratio against first stage discharge pressure for each stage was used to determine the optimum compressor ratio and intermediate pressure. Another case study to determine the total compressor input at variable compression ratio was done and a plot of total compression power input against the compression ratio at each stage revealed that the total power input of the chiller was 280.5kW. The optimized chiller has an actual coefficient of performance (COP) of 6.24 and a theoretical COP of 8.64 while the efficiency of the chiller was 72.2%. The optimized chiller reduced the energy cost per day from 257.9USD to 154.8USD. The heat duty of the evaporator and condenser were 8377000kJ/h and 9865000kJ/h respectively while the flow coefficient of the J-T valve was 51.57gal/min. The power consumption of the first stage was 121.8kW while the power consumption at the second stage was 158.7kW. The compressor discharge was 1300kPa and 56.89oC. R134a was used as the refrigerant.

In this article, we demonstrate a silicon-based carrier-injection optical single-sideband (OSSB) modulator featuring a double-parallel Mach-Zehnder interferometer architecture. This design simultaneously achieves ultra-low driving voltage and high sideband suppression ratio (SSR). By optimizing the p- and n-doping profiles, the device achieves a V<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">π</sub>·L of approximately 0.0375 V·cm. In OSSB operation, the unwanted sideband is suppressed by around 33 dB, highlighting the modulator’s exceptional sideband suppression. These characteristics—low driving voltage, low loss, and high suppression—make the device highly suitable for homodyne phase-locking, on-chip sensing, and other integrated photonic applications.

Vertical integration of two-dimensional materials holds tremendous potential for integrated sensing, memory, and computing applications, yet it still confronts challenges such as single device functionality, limited in-memory logic capability, and high power consumption. To address these issues, we propose an asymmetric van der Waals integration strategy based on an In₂Se₃/MoOₓ/MoS₂/graphene heterojunction, which integrates five reconfigurable logic gates (AND, OR, NOT, NOR, and NAND), dual-mode photodetection (~10 fA dark current, a high responsivity of 89.3 mA/W and a specific detectivity of 1.4 × 10¹¹Jones), and low-power neurosynaptic functions (7-bit conductance states, subfemtojoule energy consumption) into a single device. By virtue of these characteristics, the device enables high-precision image recognition, simulation of classical Pavlovian conditioning and single-pixel dual-band optical imaging. This work paves a feasible path for the development of multifunctionally integrated sensor-memory-computing devices.

Perovskite memristors hold great promise for neuromorphic computing and bioinspired sensing, yet their development is hindered by high operating voltages, environmental instability, and limited functional integration. Here, we report a robust memristor based on a 3D/2D halide perovskite heterostructure that overcomes these challenges. The device demonstrates ultralow switching voltages of ∼+0.18 V (SET) and ∼-0.4 V (RESET), a high ON/OFF ratio >104, excellent retention (>104 s), and endurance (>600 cycles) under ambient conditions. This performance stems from a synergistic "bandgap staircase and built-in electric field" mechanism at the heterointerface, which enhances field confinement for low-voltage switching while the 2D layer suppresses ion migration. Remarkably, this single platform integrates a complete neuromorphic-sensory functional chain. It emulates synaptic plasticity, achieving 90.77% accuracy in MNIST handwritten digit recognition. It also functions as a physical reservoir for temporal pattern processing, reaching perfect classification accuracy. Furthermore, it serves as an artificial nociceptor that faithfully replicates key pain perceptions such as threshold, nonadaptation, sensitization, and relaxation. With ultralow power consumption of only 36 pJ per switch, this multifunctional memristor provides a versatile hardware prototype for next-generation intelligent systems and adaptive human-machine interfaces.

Wireless sensing systems enable real-time, non-contact monitoring for next-generation intelligent platforms. Ideal wireless sensing systems feature compact, low power consumption, and long communication range. Here we report a miniaturized wireless sensing system with an integrated acoustic-resonance-driven piezoelectric microantenna (PE μ-antenna) with a 0.0196 mm2 active area. The PE μ-antenna integrated on a film bulk acoustic resonator (FBAR) achieves dual-frequency radiation at 1.85 GHz and 3.91 GHz with gains of -32.96 dBi and -20.5 dBi, respectively. The μ-antennas achieve over four orders of magnitude radiation efficiency enhancement and volume reduction compared with existing piezoelectric transmitters. We further extend this approach to high-overtone bulk acoustic resonators with high quality factors for wireless sensing. The system enables temperature and strain sensing with a transmission range up to 1 m, demonstrating state-of-the-art miniaturization and transmission performance among wireless sensing systems. This work establishes a scalable platform for ultracompact wireless sensors and communication nodes in biomedical, wearable, and aerospace applications.

Two-dimensional van der Waals heterojunctions present a promising platform for advanced optoelectronic devices. Yet, achieving self-powered operation with overall performance metrics remains challenging, especially for devices based on emerging metal phosphorus trichalcogenides (MPTs), which have historically underperformed. To tackle this issue, we constructed an optimal type-I heterojunction by integrating FePSe3 with WSe2, both featuring Se-terminated surfaces. The resulting built-in potential, confirmed to be 0.18 eV through Kelvin probe force microscope and ultraviolet photoelectron spectroscopy measurements, facilitates efficient, self-driven carrier separation. Our photodetector demonstrates an impressive combination of properties at zero bias: a high responsivity of ∼59.1 mA/W, an ultralow dark current at the 0.4 pA level, an exceptional specific detectivity of 1.2 × 1011 Jones, and a rapid response time of 49.8 μs. It shows broadband sensitivity from ultraviolet (250 nm) to near-infrared (950 nm) with distinct spectral discriminability, supported by a high photocurrent on/off ratio of 5.8 × 104. The practical viability is further validated through high-speed, single-pixel imaging application and low-noise optical communication without any power consumption. This work not only sets a performance benchmark for MPT-based photodetectors but also establishes a versatile material platform for next-generation, low-power imaging and intelligent spectrally selective sensing technologies.

We present a novel investigation into the impact of inter-drone interference on delivery efficiencies within multi-drone skyway networks . We conduct controlled experiments to analyze the behavior of drones in an indoor testbed environment. Our study compares performance between solo flights and concurrent multi-drone operations along predefined routes. This analysis captures interference occurring during both flight and at charging stations, providing a comprehensive evaluation of its effects on overall network performance. We conduct a comprehensive series of experiments across diverse scenarios to systematically understand and model the dynamics of inter-drone interference. Key metrics, such as power consumption and delivery times , are considered. This generates a comprehensive dataset for in-depth analysis of interference at both the node and segment levels. These findings are then formalized into a predictive model. The results validate the effectiveness of the developed model, demonstrating its potential to accurately forecast inter-drone interferences.

ABSTRACT A significant proportion of power consumption in data centers is ultimately converted into low‐grade waste heat (WH), which is typically discharged into the atmosphere, resulting in substantial energy loss and environmental degradation. Utilizing heat pump (HP) systems to recover this WH for district heating presents a promising approach to improving energy efficiency and reducing environmental impact. While previous research primarily focused on the feasibility and technical implementation of such recovery systems, limited attention has been given to the co‐optimization of WH utilization, particularly in systems that integrate HPs with energy storage. To address this gap, this study proposes an intelligent control framework that integrates user‐side demand response with deep reinforcement learning to optimize system performance. Specifically, the twin delayed deep deterministic policy gradient algorithm is employed to generate real‐time, adaptive control strategies. Additionally, a feasible action screening mechanism is introduced to ensure that control actions conform to the physical constraints of the system, thereby enhancing training stability and learning efficiency. Simulation results demonstrate that, compared with a benchmark model, the proposed approach improves system profit by 61.8% and increases renewable energy surplus by 25.7%.

The integration of hydrogen into natural-gas-fired gas turbines represents a promising transitional pathway for reducing greenhouse gas emissions in industrial power generation. This study presents a comparative thermodynamic and environmental assessment of a Solar Titan 130 gas turbine operating in combined heat and power (CHP) mode under two fueling conditions: conventional natural gas and a hydrogen-enriched CH4/H2 (80/20 vol.%) blend. The analysis combines validated operational data for natural gas with analytical thermodynamic modeling for the blended-fuel scenario to evaluate key performance indicators, including thermal efficiency, specific fuel consumption, power output, and carbon dioxide emissions. The results indicate that hydrogen enrichment leads to an increase in thermal efficiency from 34.1% to 36.6% and a reduction in specific CO2 emissions by approximately 13.7%, while maintaining similar thermal input within the adopted steady-state modeling framework. Compressor power consumption decreases, and net electrical output increases slightly under hydrogen-enriched operation, contributing to improved overall energy performance. Although the hydrogen-blended regime is assessed through modeling, the findings suggest that moderate hydrogen addition can enhance efficiency and environmental performance in industrial gas turbines without fundamental structural redesign of the turbine core, assuming appropriate fuel supply and control system adaptation. The study provides practical insights into the feasibility of hydrogen-assisted operation in existing CHP installations and supports its role in near-term decarbonization strategies.

The hyperloop transportation system is a promising ultra-high-speed mobility solution operating in a reduced-pressure environment, where pod performance is governed by the coupled behaviour of structural integrity, aerodynamics, and electromagnetic propulsion. This paper presents the design, numerical analysis, and performance evaluation of a lightweight hyperloop pod equipped with a linear induction motor (LIM)-based propulsion and electromagnetic stabilisation system. The pod chassis was fabricated using Carbon Fibre-Reinforced Polymer (CFRP) and Aluminium 6061-T6, achieving a significant weight reduction while maintaining structural safety. Finite Element Analysis reveals a maximum von Mises stress of 82 MPa, which is well below the material yield strength, and a maximum deformation of 0.64 mm under worst-case loading conditions. Modal analysis indicates the first natural frequency at 47.6 Hz, ensuring sufficient separation from operational excitation frequencies. Computational Fluid Dynamics analysis conducted inside a rectangular tube shows a drag coefficient reduction of approximately 18% compared to a baseline blunt design, with stable velocity distribution and no flow choking at operating speeds. The optimised nose geometry enables rapid acceleration, achieving 25 km/h within 1.1 s in prototype testing. The LIM analysis demonstrates a peak thrust of 1.85 kN at an optimal slip range of 6–8%, with operating currents between 35 and 55A and power consumption of 18–25 kW. Thermal analysis confirms a maximum stator temperature of 78 °C, remaining within safe operating limits. The integrated numerical and experimental results confirm the feasibility, efficiency, and stability of the proposed hyperloop pod design.

Biomass resources are regarded as a promising renewable energy alternative to fossil fuels in electric arc furnace (EAF) steelmaking. This research evaluates the feasibility of replacing natural gas (NG) with biomethane and coke fines with biochar in full-scrap EAF steel production through mass-energy, economic and emissions analyses. Results show that fully substituting NG with biomethane at 4.51 Nm 3 /t (136.6 MJ/t) increases production costs by about 0.4 to 1.6 USD/t. Replacing coke fines with biochar requires 25.6 kg/t of wood biochar (WB), 39.4 kg/t of rice husk biochar (RHB) and 60.0 kg/t of corn stover biochar (CSB). Power consumption changes include a decrease of 4.2 kWh/t for WB, and increases of 41.5 kWh/t for RHB and 63 kWh/t for CSB, relative to a baseline of 359.3 kWh/t crude steel (CS). These replacements increase production costs by approximately $7 to $23 USD/t. Regarding CO 2 emissions, biomethane substitution reduces emissions by 8.9 kg CO 2 eq/t CS (2.2%), while complete substitution with WB achieves the highest reduction of 83.8 kg CO 2 eq/t CS (20.9%). In contrast, RHB and CSB increase emissions by 3.6 and 42.7 kg CO 2 eq/t CS, respectively. Combining biomethane with WB can cut CO 2 emissions by 92.7 kg per ton of CS during EAF steelmaking. Overall, these findings indicate that biomass options could significantly enhance the environmental sustainability of EAF steelmaking.

Electromagnetic interference (EMI) analysis in high-speed industrial systems is increasingly challenged by multi-gigahertz sampling rates, complex transient behaviors, and stringent real-time constraints. To address these challenges, this paper proposes a pulse-aware generative and analysis framework based on a generative adversarial network (GAN) combined with pulse sparse convolution using leaky integrate-and-fire (LIF) spiking neurons. A multi-scale discriminator and gradient penalty stabilization are employed to improve waveform generation fidelity, achieving a Fréchet distance (FID) of 0.72 and a global difference metric (GDM) of 0.18 ± 0.03 on an industrial-grade Electromagnetic compatibility (EMC) dataset. The proposed framework is further applied to crosstalk prediction, where it reduces pulse-width and phase prediction errors by more than 40% compared with classical numerical solvers such as finite-difference time-domain (FDTD), finite element method (FEM), and method of moments (MoM), and consistently outperforms representative learning-based EMC models. To enable real-time deployment, the pulse sparse convolution architecture is implemented on an field-programmable gate array (FPGA) platform using fixed-point arithmetic, achieving deterministic inference at 5 GS/s with a measured power consumption of 0.71 W. Extensive experiments on traction systems, industrial robots, CNC drives, photovoltaic inverters, and UAV (Unmanned Aerial Vehicle) electronics demonstrate that the proposed approach provides accurate, stable, and energy-efficient EMI analysis suitable for practical industrial EMC applications.

System-on-chip security architecture is a critical, complex, and time-consuming activity, consuming months of effort. Furthermore, the architectural design can include subtle errors that compromise the security of the entire system. In this paper, we develop a security engine infrastructure, SEnTry , for systematically creating security architectures for protecting SoC designs against a variety of security subversions. SEnTry provides a plug-and-play, configurable subsystem composed of custom IPs that can be integrated into the platform to derive different security primitives. We develop an instance of SEnTry for supply-chain attacks. We discuss the spectrum of challenges involved in developing a unified architecture for systematic protection against the variety of attacks involved and the SEnTry approach to addressing them. We provide several case studies to demonstrate SEnTry design and perform extensive experiments to evaluate its overhead on multiple ASIC technologies. Our experiments suggest that SEnTry incurs minimal overhead in area and power consumption.

Abstract Choosing the cross-section of conductors is an essential task in planning distribution grids. The cross-sectional area of conductors significantly influences the investment cost as well as the energy loss of the electricity grid. However, previous studies have mainly used the maximum load value and the constant power load model to select the conductor cross-section. This paper presents an optimization framework that takes into consideration the influence of the load curve shape and the dependence of the load’s power consumption on voltage to determine the optimal wire cross-section. The proposed optimization framework aims at minimizing the total life-cycle expenditure of the power grid while complying with constraints such as load flow equations, node voltage limits, along with current limits on branches. The suggested optimization formulation is modeled as second-order cone programming with integer variables (MISOCP), guaranteeing a globally optimal solution using commercial optimizers such as GUROBI. The proposed MISOCP model is transformed from a nonlinearly constrained optimization model with integer variables (MINLP) by constructing a conic quadratic model of the load flow equations, an equivalent ZP formulation of voltage-dependent loads (ZIP load model), and a precise linearization process of the bilinear products involving binary and continuous variables. The suggested optimization framework is evaluated on the 16-node system and the real Vietnamese 54- and 102-node medium-voltage power grids. These systems utilize different time resolutions: the 16-node system is analyzed using a 24-time-period profile, while the Vietnamese grids are assessed using a day-night load profile with 96 time periods. The calculation results highlight the need to consider load curves and load power dependence on voltage when selecting conductor sizes for distribution grids. Ultimately, the proposed framework is positioned as a high-precision, strategic offline planning tool, prioritizing global optimality and investment accuracy over real-time computational speed.

Abstract - Electricity theft and inefficient energy monitoring are major challenges faced by power distribution companies worldwide. Conventional energy meters lack real-time monitoring capability and are unable to detect unauthorized power usage effectively. This paper presents the design and implementation of an IoT based Smart Energy Meter with an Anti-Theft Detection System using ESP32. The proposed system measures electrical parameters such as voltage, current, power, and energy consumption using appropriate sensing modules. The collected data is processed by the ESP32 microcontroller and transmitted to a cloud based platform via Wi-Fi for real-time monitoring and analysis. The anti-theft mechanism is implemented by comparing the input current from the main supply with the output current delivered to the load. If a significant mismatch beyond a predefined threshold is detected, the system identifies it as a theft condition and sends an alert notification to the utility provider. By integrating Internet of Things technology with intelligent monitoring and control, the proposed solution enhances transparency, improves billing accuracy, and reduces non-technical losses. The system is scalable, and suitable for residential as well as commercial applications. The proposed system provides an accurate, cost-effective, and reliable solution for smart energy management and electricity theft prevention. Therefore, the IoT-based smart energy meter provides. Therefore, the IoT-based smart energy meter provides a smart, efficient, and cost-effective solution for monitoring electricity usage and detecting unauthorized power consumption in modern smart grid environments. Keywords :- Internet of Things (IoT), Smart Energy Meter, Electricity Theft Detection, ESP32 Microcontroller, Real-Time Monitoring, Energy Management, Cloud-Based Monitoring

High critical switching current density (Jc) and limited charge-to-spin conversion efficiency [spin–orbit torque (SOT) efficiency, ξSOT] of heavy metal-based SOT channels are the two key challenges for the development of energy-efficient SOT magnetic random-access memory (SOT-MRAM). In this paper, we demonstrate a significant enhancement of SOT efficiency in β-tungsten (W) films through magnesium oxide (MgO) doping, achieved via co-sputtering (CS) and multilayer-heterostructure (MLH) deposition, respectively. The W/MgO nanostructures retain the β-phase and exhibit a remarkable increase in ξSOT, from 0.20 in pure W to 0.48 for CS films (4% MgO doping)—and even more impressively—to 0.51 for the MLH method. The drastic SOT efficiency improvement is attributed primarily to the significantly enhanced skew scattering induced by MgO incorporation. As direct performance gauges, Jc of W/MgO nanostructures is reduced by up to 50.6%, and the power consumption is lowered by over 42%, compared to pure W-based devices. These findings highlight W/MgO nanostructures as highly promising SOT channel materials for energy-efficient SOT-MRAM applications.

Hydroponic farming offers a sustainable alternative to traditional agriculture but is highly prone to rapid disease transmission due to its shared water systems. Timely detection of plant diseases is critical to prevent widespread crop loss. In this research, the YOLOv11n object detection model was evaluated in detail for the purpose of real-time hydroponic plant disease detection, and its accuracy, inference speed, power consumption, and resource utilization were compared through various edge devices such as Raspberry Pi 5 (Raspberry Pi Holdings, Cambridge, UK), NVIDIA Jetson Nano (NVIDIA Co., Santa Clara, CA, USA), and AMD Radeon Vega 8 (AMD Micro Devices, Inc., Santa Clara, CA, USA). The results not only confirm the applicability of light-weighted and quantized deep learning models for planting disease detection at early stages in controlled hydroponic environments but also give practical knowledge for hardware-model trade-offs in the context of sustainable edge AI-based smart farming systems.

Energy efficiency has become a primary bottleneck in hardware platforms supporting machine learning workloads, particularly as modern inference and training tasks demand sustained high-throughput computation. This challenge is further amplified in energy-harvesting and intermittently powered systems, where the available energy budget varies over time. This work introduces a run-time configurable multiply–accumulate (MAC) architecture that dynamically adjusts arithmetic precision to match instantaneous energy availability. The proposed design relies on an internally adaptive multiplier based on bit-level logic compression, enabling controlled modulation of power consumption while preserving numerical robustness. Crucially, the MAC maintains a fixed external operand interface, allowing for seamless precision adaptation without operand reformulation or datapath disruption. The architecture is implemented in System Verilog and evaluated using both ASIC synthesis in a 90 nm CMOS technology and FPGA deployment. Experimental results demonstrate approximately a fourfold improvement in power–delay product (PDP) relative to full-precision operation, with only limited degradation in inference accuracy.

An Advanced Transformer Model-Based LSTM-CNN for Power Consumption and PV Power Generation Forecasting in Modern Microgrid