Search Results (20)

The rapid growth of electric vehicles (EVs) and new energy systems has put lithium-ion batteries at the center of the clean energy change. Nevertheless, to achieve the best battery performance, safety, and sustainability in many changing circumstances, major innovations are needed in Battery Management Systems (BMS). This review paper explores how artificial intelligence (AI) and digital twin (DT) technologies can be integrated to enable the intelligent BMS of the future. It investigates how powerful data approaches such as deep learning, ensembles, and models that rely on physics improve the accuracy of predicting state of charge (SOC), state of health (SOH), and remaining useful life (RUL). Additionally, the paper reviews progress in AI features for cooling, fast charging, fault detection, and intelligible AI models. Working together, cloud and edge computing technology with DTs means better diagnostics, predictive support, and improved management for any use of EVs, stored energy, and recycling. The review underlines recent successes in AI-driven material research, renewable battery production, and plans for used systems, along with new problems in cybersecurity, combining data and mass rollout. We spotlight important research themes, existing problems, and future drawbacks following careful analysis of different up-to-date approaches and systems. Uniting physical modeling with AI-based analytics on cloud-edge-DT platforms supports the development of tough, intelligent, and ecologically responsible batteries that line up with future mobility and wider use of renewable energy. Full article

The rapid development of high-power-density semiconductor devices has rendered conventional thermal management techniques inadequate for handling their extreme heat fluxes. This manuscript presents and implements an embedded microchannel cooling solution for such devices. By directly integrating micropillar arrays within the near-junction region of the substrate, efficient forced convection and flow boiling mechanisms are achieved. Finite element analysis was first employed to conduct thermo–fluid–structure simulations of micropillar arrays with different geometries. Subsequently, based on our simulation results, a complete multilayer microstructure fabrication process was developed and integrated, including critical steps such as deep reactive ion etching (DRIE), surface hydrophilic/hydrophobic functionalization, and gold–stannum (Au-Sn) eutectic bonding. Finally, an experimental test platform was established to systematically evaluate the thermal performance of the fabricated devices under heat fluxes of up to 1200 W/cm². Our experimental results demonstrate that this solution effectively maintains the device operating temperature at 46.7 °C, achieving a mere 27.9 K temperature rise and exhibiting exceptional thermal management capabilities. This manuscript provides a feasible, efficient technical pathway for addressing extreme heat dissipation challenges in next-generation electronic devices, while offering notable references in structural design, micro/nanofabrication, and experimental validation for related fields. Full article

This systematic review examines state-of-the-art artificial intelligence applications in energy systems, assessing their performance, real-world deployments and transformative potential. Guided by PRISMA 2020, we searched Web of Science, IEEE Xplore, ScienceDirect, SpringerLink, and Google Scholar for English-language studies published between January 2015 and January 2025 that reported novel AI uses in energy, empirical results, or significant theoretical advances and passed peer review. After title–abstract screening and full-text assessment, it was determined that 129 of 3000 records met the inclusion criteria. The methodological quality, reproducibility and real-world validation were appraised, and the findings were synthesised narratively around four critical themes: reinforcement learning (35 studies), multi-agent systems (28), planning under uncertainty (25), and AI for resilience (22), with a further 19 studies covering other areas. Notable outcomes include DeepMind-based reinforcement learning cutting data centre cooling energy by 40%, multi-agent control boosting virtual power plant revenue by 28%, AI-enhanced planning slashing the computation time by 87% without sacrificing solution quality, battery management AI raising efficiency by 30%, and machine learning accelerating hydrogen catalyst discovery 200,000-fold. Across domains, AI consistently outperformed traditional techniques. The review is limited by its English-only scope, potential under-representation of proprietary industrial work, and the inevitable lag between rapid AI advances and peer-reviewed publication. Overall, the evidence positions AI as a pivotal enabler of cleaner, more reliable, and efficient energy systems, though progress will depend on data quality, computational resources, legacy system integration, equity considerations, and interdisciplinary collaboration. No formal review protocol was registered because this study is a comprehensive state-of-the-art assessment rather than a clinical intervention analysis. Full article

The growing complexities, power densities, and cooling demands of modern electronic systems and batteries—such as three-dimensional integrated circuit chip packaging, printed circuit board assemblies, and electronics enclosures—have pushed the urgency for efficient and dynamic thermal management strategies. Traditional numerical methods like computational fluid dynamics (CFD) and the finite element method (FEM) are computationally impractical for large-scale or real-time thermal analysis, especially when dealing with complex geometries, temperature-dependent material properties, and rapidly changing boundary conditions. These approaches typically require extensive meshing and repeated simulations for each new scenario, making them inefficient for design exploration or optimization tasks. Physics-informed neural networks (PINNs) emerge as a powerful alternative approach that incorporates physical principles such as mass and energy conservation equations into deep learning models. This approach delivers rapid and adaptable resolutions to the partial differential equations that govern heat transfer and fluid dynamics. This review examines the basic principle of PINN and its role in thermal management for electronics and batteries, from the small unit scale to the system scale. We highlight recent advancements in PINNs, particularly their superior performance compared to traditional CFD methods. For example, studies have shown that PINNs can be up to 300,000 times faster than conventional CFD solvers, with temperature prediction differences of less than 0.1 K in chip thermal models. Beyond speed, we explore the potential of PINNs in enabling efficient design space exploration and predicting outcomes for previously unseen scenarios. However, challenges such as training convergence in fine-grained or large-scale applications remain. Notably, research combining PINNs with LSTM networks for battery thermal management at a 2.0 C charging rate has achieved impressive results—an R² of 0.9863, a mean absolute error (MAE) of 0.2875 °C, and a root mean square error (RMSE) of 0.3306 °C—demonstrating high predictive accuracy. Finally, we propose future research directions that emphasize the integration of PINNs with advanced hardware and hybrid modeling techniques to advance thermal management solutions for next-generation electronics and battery systems. Full article

Background: Primary intracardiac tumors may be diagnosed incidentally, sometimes in the case of complications. Case Report: This case report presents a 64-year-old woman who was admitted to the emergency department with cardiac complications, including heart palpitations and shortness of breath. Initial investigations revealed the presence of ground glass opacity in the left lung and significant mediastinal adenopathy. Transthoracic echocardiography (TTE) indicated severe mitral stenosis caused by a mass attached to the mitral valve, and the transesophageal echocardiography (TEE) confirmed the presence of a tumor, raising concerns about a myxoma with a high risk of embolism. The patient experienced transitory neurological dysfunction, and subsequent imaging uncovered a thrombus occluding the left internal carotid artery. An emergency surgical procedure was performed, including extracorporeal circulation and rapid deep cooling, to facilitate safe mass excision and carotid embolectomy. Histopathological analysis of the extracted tissue revealed undifferentiated pleomorphic sarcoma (FNCLCC Grade 3). Following the surgery, the patient needed extended mechanical ventilation and subsequently underwent a tracheostomy because of her ongoing respiratory support requirements. Conclusions: Despite the complexity of the surgical intervention, the prognosis remained poor due to the aggressive nature of the tumor and neurologic complications. This case underscores the rarity of primary cardiac sarcomas, the challenges in diagnosis, and the need for prompt surgical intervention to mitigate risks associated with embolic events. Full article

The combined use of laser beam and electric arc for welding thick clad steel plates in a single pass has been developed to solve the issues concerning the individual applications of the heat sources, such as the low filling efficiency of conventional electric arc methods and the drawbacks concerning laser beam defects due to rapid cooling and solidification. This work was addressed to the weldability assessment of ferritic steel plates, clad with austenitic stainless steel, under the laser-leading configuration, testing the effects of two different values of the inter-distance between the laser beam and the electric arc. Specimens of the welded zone were investigated by metallographic observations and EDS measurements; mechanical properties were characterized by the Vickers microhardness test and by the FIMEC instrumented indentation test to obtain the local values of the yield strength. Welding simulations by theoretical modelling were also carried out to outline the differences in the thermal fields generated by the two heat sources, their interaction, and their effect on the configurations of the weld pool and the thermal profiles to which the materials are subjected. The welding setup with higher inter-distance was more suitable for joining clad steel plates, since the action of the deep keyhole mode is substantially separated from that of the shallower electric arc. In this way, the addition of alloying elements, performed by melting the filler wire, concentrated in the cladding layer, helping maintain the austenitic microstructure, while the laser beam acts in depth along the thickness, autogenously welding the base steel. Full article

A simulation of an extreme precipitation event in southern Xinjiang, which is the driest area in China, seizes the whole initiation process of the intense convective cell responsible for the high hourly rainfall amount. Considering the inner connection between convection and vertical motions, the characteristics and mechanisms of the vertical accelerations during this initial development of the deep convection are studied. It is shown that three key accelerations are responsible for the development from the nascent cumuli to a precipitating deep cumulonimbus, including sub-cloud boundary-layer acceleration, in-cloud deceleration, and cloud-top acceleration. By analyzing the right-hand terms of the vertical velocity equation in the framework of the WRF model, together with a diagnosed relation of perturbation pressure to perturbation potential temperature, perturbation-specific volume (or density), and moisture, the physical processes associated with the corresponding accelerations are revealed. It is found that sub-cloud acceleration is associated with three-dimensional divergence, indicating that the amount of upward transported air must be larger than that of horizontally convergent air. This is favorable for the persistent accumulation of water vapor into the accelerated area. In-cloud deceleration is caused by the intrusion or entrainment of mid-level cold air, which cools down the developing cloud and delays the deep convection formation. Cloud-top acceleration is responsible for the rapid upward extension of the cloud top, which is highly correlated with the convergence and upward transport of moisture. Full article

The marine carbonates in the Ordovician Majiagou Formation in the Ordos Basin have significant exploration potential. Research has focused on their thermal history and hydrocarbon accumulation stages, as these are essential for guiding the exploration and development of hydrocarbons. In this paper, we study the thermal evolution history of the carbonate reservoirs of the Ordovician Majiagou Formation in the east-central Ordos Basin. Furthermore, petrographic and homogenization temperature studies of fluid inclusions were carried out to further reveal the hydrocarbon accumulation stages. The results demonstrate that the degree of thermal evolution of the Ordovician carbonate reservoirs is predominantly influenced by the deep thermal structure, exhibiting a trend of high to low values from south to north in the central region of the basin. The Fuxian area is located in the center of the Early Cretaceous thermal anomalies, with the maturity degree of the organic matter ranging from 1 to 3.2%, with a maximum value of 3.2%. The present geothermal gradient of the Ordovician Formation exhibits the characteristics of east–high and west–low, with an average of 28.6 °C/km. The average paleo-geotemperature gradient is 54.2 °C/km, the paleoheat flux is 55 mW/m², and the maximum paleo-geotemperature reaches up to 270 °C. The thermal history recovery indicates that the Ordovician in the central part of the basin underwent three thermal evolution stages: (i) a slow warming stage before the Late Permian; (ii) a rapid warming stage from the end of the Late Permian to the end of the Early Cretaceous; (iii) a cooling stage after the Early Cretaceous, with the hydrocarbon production of hydrocarbon source rocks weakening. In the central part of the basin, the carbonate rock strata of the Majiagou Formation mainly developed asphalt inclusions, natural gas inclusions, and aqueous inclusions. The fluid inclusions can be classified into two stages. The early-stage fluid inclusions are mainly present in dissolution holes. The homogenization temperature is 110–130 °C; this coincides with the hydrocarbon charging period of 210–165 Ma, which corresponds to the end of the Triassic to the end of the Middle Jurassic. The late-stage fluid inclusions are in the dolomite vein or late calcite that filled the gypsum-model pores. The homogenization temperature is 160–170 °C; this coincides with the hydrocarbon charging period of 123–97 Ma, which corresponds to the late Early Cretaceous. Both hydrocarbon charging periods are in the rapid stratigraphic warming stage. Full article

With the rapid progress in data mining, deep learning, and artificial intelligence, the demand for datacenters of various sizes increases globally. Datacenters typically require an environment with properly controlled temperature and humidity conditions for their proper operations. These needed environmental conditions are always provided by an air conditioning system. In humid and hot regions, both energy consumption and the splash of water condensate in using the fin-and-tube heat exchangers are of concern because reliability issues can occur. In this study, the effects of fin surface hydrophilic/hydrophobic coatings on the performance of the fin-and-tube heat exchangers, including the heat transfer rate, pressure drop, and water-condensate splash, were investigated experimentally. By varying the cooling air speeds and fin pitches, the results show that hydrophilic surface coating is an effective method in reducing both the pressure drop (thus saving energy) and the condensate splash, while not affecting the heat transfer rates significantly. The water splash reduction is achieved by both the increased air speed for splashing and a smaller amount of splashing. Water splash can even be completely eliminated if the airspeed was below about 3 m/s. In contrast, hydrophobic surface coating will increase both pressure drop and water splash; thus, should be applied with caution. Full article

The sustainable development of society depends on the reliable supply of electricity while keeping impacts on the environment to a minimum. A 951 MWe nuclear power plant in the semi-enclosed Nanwan Bay at the southern tip of Taiwan began operating in May 1984. Part of the bay is in Kenting National Park, which is known for its coral reefs and abundant marine life; thus, thermal pollution from the cooling water discharge is a great concern. Fortunately, the bay opens south to face the Luzon Strait, where the world’s strongest internal tides are generated. Because the bay is deep enough, internal waves bring up cold deep water and reduce the surface temperature by as much as 10 °C for a few hours every day. These internal waves and topographically generated upwelling also bring nutrients to the euphotic layer from the depths, but the upwelled waters quickly leave the bay along with the cooling water. As a result, a thermal plume with a temperature of 1 °C or higher than the ambient temperature only covers 1 km. By way of comparison, El Niño—Southern Oscillation- or Pacific Decadal Oscillation-related interannual variations in temperature are as high as 5 °C. The rapid turnover of the upwelled waters also helps to prevent heat released by the power plant from accumulating and diminishes the thermal stress, thus sustaining corals and other marine life forms. Typhoons, even hundreds of kilometers away, could also induce the upwelling of cold subsurface water. Consecutive typhoons have been observed to reduce the water surface temperature by up to 10 °C for two weeks or longer. Furthermore, the currents are such that the thermal plume flows out of the bay most of the time. All of these factors make the surface waters in the bay about 0.5 °C cooler than the waters outside of the bay, despite the operation of a nearby nuclear power plant. Full article

Micro/nanoscale fabricated devices have widely been used in modern technology and bioengineering as they offer excellent heat transfer. Removal of excess heat, coolant selection, rapid mixing, and handling proportion of colloidal metallic nanogranules in the base fluid are the main challenges in micro/nanofluidic systems. To address these problems, the primary motivation of the intended mathematical flow problem is to investigate the thermal and flow aspects of blood-based ternary nanofluid in the presence of inclined magnetic field and thermal radiations through a microfluidic pump with elastic walls. Further, the pump inner surface is smeared with fabricated cilia. The embedded cilia blow in coordination to start metachronal travelling waves along the pump wall that assist homogenous mixing and manipulation. The entire analysis is conducted in moving frame and simplified under lubrication and Rosseland approximations. Numerical solution of various flow and thermal entities are computed via the shooting method and plotted for different values of the parameters of interest. A comparative glimpse allows us to conclude that the trimetallic blood-based nanofluid exhibits elevated heat transfer rate by 12–18%, bi-metallic by about 11–12%, and mono nanofluid by about 6% compared to the conventional blood model. The study also determines that the prolonged cilia commence augmentation in flowrate and pressure-gradient around the pump deep portion. Furthermore, the radiated ternary liquid under fragile magnetic field effects may contribute to the cooling process by eliminating unnecessary heat from the system. It is also noticed that around the ciliated wall, the heat transfer irreversibility effects are appreciable over the fluid frictional irreversibility. Full article

This paper proposes a meta-modeling workflow to forecast the cooling and heating loads of buildings at individual and district levels in the early design stage. Seven input variables, with large impacts on building loads, are selected for designing meta-models to establish the MySQL database. The load profiles of office, commercial, and hotel models are simulated with EnergyPlus in batches. A sequence-to-sequence (Seq2Seq) model based on the deep-learning method of a one-dimensional convolutional neural network (1D-CNN) is introduced to achieve rapid forecasting of all-year hourly building loads. The method performs well with the load effective hour rate (LEHR) of around 90% and MAPE less than 10%. Finally, this meta-modeling workflow is applied to a district as a case study in Shanghai, China. The forecasting results well match the actual loads with R² of 0.9978 and 0.9975, respectively, for the heating and cooling load. The LEHR value of all-year hourly forecasting loads is 98.4%, as well as an MAPE of 4.4%. This meta-modeling workflow expands the applicability of building-physics-based methods and improves the time resolution of conventional data-driven methods. It shows small forecasting errors and fast computing speed while meeting the required precision and convenience of engineering in the building early design stage. Full article

Defective hard candies are usually produced due to inadequate feeding or insufficient cooling during the candy production process. The human-based inspection strategy needs to be brought up to date with the rapid developments in the confectionery industry. In this paper, a detection and classification method for defective hard candies based on convolutional neural networks (CNNs) is proposed. First, the threshold_li method is used to distinguish between hard candy and background. Second, a segmentation algorithm based on concave point detection and ellipse fitting is used to split the adhesive hard candies. Finally, a classification model based on CNNs is constructed for defective hard candies. According to the types of defective hard candies, 2552 hard candies samples were collected; 70% were used for model training, 15% were used for validation, and 15% were used for testing. Defective hard candy classification models based on CNNs (Alexnet, Googlenet, VGG16, Resnet-18, Resnet34, Resnet50, MobileNetV2, and MnasNet0_5) were constructed and tested. The results show that the classification performances of these deep learning models are similar except MnasNet0_5 with the classification accuracy of 84.28%, and the Resnet50-based classification model is the best (98.71%). This research has certain theoretical reference significance for the intelligent classification of granular products. Full article

Geothermal power is being regarded as depending on techniques derived from hydrocarbon production in worldwide current strategy. However, it has artificially been developed far less than its natural potentials due to technical restrictions. This paper introduces the Enhanced Geothermal System based on Excavation (EGS-E), which is an innovative scheme of geothermal energy extraction. Then, based on cohesion-weakening-friction-strengthening model (CWFS) and literature investigation of granite test at high temperature, the initiation, propagation of excavation damaged zones (EDZs) under unloading and the EDZs scale in EGS-E closed to hydrostatic pressure state is studied. Finally, we have a discussion about the further evolution of surrounding rock stress and EDZs during ventilation is studied by thermal-mechanical coupling. The results show that the influence of high temperature damage on the mechanical parameters of granite should be considered; Lateral pressure coefficient affects the fracture morphology and scale of tunnel surrounding rock, and EDZs area is larger when the lateral pressure coefficient is 1.0 or 1.2; Ventilation of high temperature and high in-situ stress tunnel have a significant effect on the EDZs scale; Additional tensile stress is generated in the shallow of tunnel surrounding rock, and the compressive stress concentration transfers to the deep. EDZs experiences three expansion stages of slow, rapid and deceleration with cooling time, and the thermal insulation layer prolongs the slow growth stage. Full article

During the construction of nuclear waste storage facilities, deep drilling, and geothermal energy development, high-temperature rocks are inevitably subjected to thermal shock. The physical and mechanical behaviors of granite treated with different thermal shocks were analyzed by non-destructive (P-wave velocity test) and destructive tests (uniaxial compression test and Brazil splitting test). The results show that the P-wave velocity (V_P), uniaxial compressive strength (UCS), elastic modulus (E), and tensile strength (s_t) of specimens all decrease with the treatment temperature. Compared with air cooling, water cooling causes greater damage to the mechanical properties of granite. Thermal shock induces thermal stress inside the rock due to inhomogeneous expansion of mineral particles and further causes the initiation and propagation of microcracks which alter the mechanical behaviors of granite. Rapid cooling aggravates the damage degree of specimens. The failure pattern gradually transforms from longitudinal fracture to shear failure with temperature. In addition, there is a good fitting relationship between P-wave velocity and mechanical parameters of granite after different temperature treatments, which indicates P-wave velocity can be used to evaluate rock damage and predict rock mechanical parameters. The research results can provide guidance for high-temperature rock engineering. Full article