Using H 2 O–TiO 2 as the working fluid, the thermal performance of a counterflow double-tube heat exchanger was numerically investigated. The investigation was performed by employing the finite element method (FEM) using COMSOL Multiphysics to analyze the flow and thermal characteristics of the fluid flow in the turbulent state with the application of realizable k–ε turbulence model. Two concentrations of nanofluid were used (5 vol% TiO 2 and 10 vol% TiO 2 ) and comparisons were made to water as a baseline across a range of Reynolds numbers for both the inner (35 000-85 000) and outer fluids (10 000-30 000). The thermophysical properties, namely density, heat capacity, thermal conductivity, and dynamic viscosity, of the nanofluid were taken into account in the single-phase model of the nanofluid which was developed based on the Maxwell model and Brinkman model, thereby simulating the behaviour of the nanofluid as accurately as possible. According to measurement results, it was evident that the presence of TiO 2 in water results in a significant increase in the heat transfer efficiency of the working fluid, where the nanofluid containing 10 vol% TiO 2 has a Nusselt number which increases by as much as 38% in this case, as well as a heat transfer coefficient which increases by 45% when compared to water at the same Reynolds number.
This research offers a 3D analysis of a solar air heater’s performance using Computational Fluid Dynamics (CFD) and COMSOL Multiphysics software. The air heater design is multi-pass and therefore, the air will remain in the heater longer, thus increasing the transfer of heat from the surface to the air. Fluid flow and heat transfer governing equations were solved via the finite element method with time-dependent boundary conditions using actual measured solar radiation and ambient temperature data. The results of the simulation showed that there is a clear indication of the diurnal thermal response of the heater; the surface temperature rose from the morning to midday, peaked during the times of maximum solar radiation, and then fell from midday to night. The air outlet temperature was always higher than the air inlet temperature over the entire operating time frame, thus demonstrating heat was transferred to the working fluid. Additionally, heat gain increased greatly during times of high solar radiation and the range of heat transfer coefficients remained tight, indicating stable convective performance. Overall, the simulations confirm that the incorporation of longitudinal baffles provides improved thermo-hydraulic behaviour and thermal effectiveness in usually outdoor conditions of a solar air heater.
In this research, the heating and cooling requirements of variances regions of Uzbekistan using basic climatic information and machine learning method based on. The relevant daily information on temperatures was collected from four cities - Termiz, Samarkand, Tashkent and Karakalpak. The heating degrees days (HDD) and the cooling degrees days (CDD) were calculated using standard bas temperature from this data. These days reflect the amount of energy received for heating and cooling. The relationship between temperature and location and energy consumption was established using Random Forest model. The input data consisted of average, max and min temperatures and geographic position information - latitude and longitude, and elevation. The data was divided into training and testing sets in order to test the adequacy of the model. There is a huge difference between the different regions through the use of this model. The temperatures in the North require heating to higher amounts and the temperature in the South requires cooling to higher rates. The values obtained were tending to be very close to the real rates that indicate the adequacy of the approached method. The main factor is temperature according to this instance.
Modeling the temporal and spatial dynamics of heat pollution within environmental systems is essential to accurately predict thermal pollution within the environment. Therefore, accurate predictions of thermal pollution dynamics require an understanding of the multiscale, nonlinear interactions that establish the mechanisms of thermal transport, dispersion, and ecological response. In this paper, we provide an integrated set of methodologies that applies both three-dimensional computational fluid dynamics (CFD) and cutting-edge machine learning (ML) and deep learning (DL) methodologies for the environmental thermal assessment. We assess the performance of six types of predictive ML models-linear regression, random forest, gradient boosting, multilayer neural networks (MNNs), support vector regression (SVRs), and long short-term memory (LSTM) networks-against benchmark datasets for predicting thermal plumes, predicting cooling tower dynamics, and forecasting river temperatures using CFD. Among these models, the LSTM networks performed by far the best for predicting thermal activities over time (R² = 0.95, RMSE = 1.5°C). Conversely, the best performing model for identifying spatial thermal patterns was gradient boosting.
The main goal of this research is to numerically analyse the thermal energy storage system using a phase change material (PCM) and extending branched fins as the method investigated in this thesis, including the design of two different configurations to the branched fin; a conventional branched radial fin and an extended multi-branched fin that would be used for heat transfer through the TES system. A two-dimensional transient model was developed using COMSOL Multiphysics to represent the thermal performance of the TES system during charging of the TES, the inlet temperature to the TES was set at 20°C and the heating fluid in the central tube at 80°C while the phase change material had a melting point at 60°C, the distribution of temperature, melting fraction and temporal behaviour of each of the PCMs were also analysed. The results illustrate the increase in heat transfer rate when using branched fins, as well as the rate of melting of PCMs, and show that the configuration with extended multi-branch fins provided better thermal results than the configuration with typical branched radial fins, thus decreasing the rate of melting of the phase change material and providing a more consistent temperature distribution than the use of typical branched radial fins. The results provide insight into the importance of fin geometry for the increased performance of thermal energy storage systems that employ phase change materials.
An analysis of how a dynamic system’s frequency responds when subjected to harmonic excitation at various input frequencies is the primary goal of this study. The study was carried out using MATLAB/Simulink as the simulation-based medium. A Simulink model (simulation) of the dynamic system (first-order) was completed using baseline functional blocks (sine wave generator, output signal display block, and transfer function block). The output signal of this dynamic system (first-order) was analysed with sinusoidal signals at varying frequencies. The results indicated that the output signal accurately followed the input signal for low input frequencies; however, at higher input frequencies, both the output signal number of amplitudes and resultant phase lag increased significantly as compared to those associated with lower frequency input signals. These results provided an indication of the system’s filtering characteristics. Consequently, the effectiveness of MATLAB/Simulink for modelling, visualizing, and analysing the frequency response of dynamic systems has been documented by this study; thereby benefiting future engineering, control, and modelling system design and analysis.
In this study, the optimal insulation thickness for external walls was analyzed in relation to the climatic conditions of the Termez region. Calculations were carried out using the degree-day (HDD) method, and heat loss, energy consumption and economic factors were evaluated together. The study determined the optimal thicknesses for foam plastic and basalt insulation materials, and observed that insulation thickness increased as the base temperature rose from 18°C to 22°C. The results showed that the optimal thickness for foam plastic is 0.029–0.038 m, and for basalt, 0.021–0.029 m. An increase in HDD values extends the heating period, requiring higher thermal resistance. At the same time, it was found that basalt insulation provides higher efficiency at lower thicknesses. This indicates the importance of material selection. The results of the analysis confirm that the optimal insulation thickness depends directly on climatic conditions, base temperature and material properties. This approach can reduce energy consumption, increase economic efficiency, and ensure thermal comfort in buildings.
The goal of this research is to evaluate energy management issues in renewable-connected smart grids, especially when the power generation produced by solar and wind fluctuates, making it difficult to maintain balanced electricity supply. There is a need for battery energy storage systems (BESS) to operate optimally in order to minimize the dependency on electrical utilities, smooth out fluctuations, and increase the amount of energy generated from renewable sources. As such, we proposed a framework using reinforcement learning (RL) to create an intelligent BESS charge-discharge schedule that can adapt to the dynamic nature of the grid. Specifically, the model uses the state of the grid (the level of renewable power produced, how much electricity is being consumed, and the battery’s state of charge) to learn the best control actions to improve BESS operation. Our simulation results indicate that using RL to optimize BESS operation will improve the efficiency of dispatching energy, increase the percentage of renewable energy used, and decrease operating costs compared to traditional ways of controlling BESS. RL shows potential for use in adapting to energy management issues in smart grids.
In this research, an analysis of the heating and cooling energy requirements throughout all regions of Uzbekistan has been conducted with the use of Heating Degree Days (HDD) and Cooling Degree Days (CDD). The research involves four locations, namely Termiz, Samarkand, Tashkent, and Karakalpak, which stand for the southern, central, and northern climate zones. The research has used the daily temperature data for the period from 2005 to 2020 in order to find HDD and CDD for the base temperatures of 18°C and 20 °C, respectively. The core finding of the study is that there is significant difference in energy consumption across regions, namely Karakalpak has the highest heating consumption level (3631 HDD), while Termiz has the lowest (1517 HDD). It has also been found that cooling consumption is the highest in Termiz (1296 CDD) and the lowest in Karakalpak (776 CDD), while the two regions of Samarkand and Tashkent have intermediate values of energy consumption. One of the techniques used for the analysis is the so-called heatmap, which illustrates the relationship between the altitude level and demand for energy (increasing demand for heating and cooling energy in the north and south respectively).
Accurate prediction of nanofluid thermal conductivity is critical for advancing heat transfer applications in energy systems, electronics cooling, and biomedical devices. This paper presents an explainable machine learning framework that integrates experimental nanofluid datasets with Koo-Kleinstreuer-Li (KKL) model-derived features to predict thermal conductivity enhancement with high fidelity. Gradient Boosting Regression and Random Forest models are trained on a consolidated public dataset of Al2O3, CuO, TiO2, and SiO2 nanofluids encompassing particle size, volume fraction, base fluid type, and temperature as primary features, achieving a coefficient of determination (R2) of 0.9871 and a root mean square error (RMSE) of 0.0041 W/mK. SHapley Additive exPlanations (SHAP) are employed to decompose model predictions and rank feature contributions globally and locally, revealing that volume fraction and temperature are the most dominant predictors of thermal conductivity enhancement. The KKL model’s Brownian motion component was engineered as an additional feature, improving predictive accuracy by 3.2% over baseline models. Comprehensive ablation studies, learning curves, and residual analyses confirm model robustness and generalizability across nanofluid families. The proposed explainable pipeline bridges the gap between black-box machine learning and physics-informed interpretability, offering actionable in-sights for nanofluid design. Results demonstrate that SHAP-driven explainability not only validates model consistency with established thermophysical theory but also identifies previously underexplored interaction effects between particle morphology and base fluid viscosity. This work establishes a reproducible open-source workflow suitable for thermal engineering researchers seeking both predictive power and scientific transparency.
This paper describes an experimental study using the software program, COMSOL Multiphysics, to evaluate how effectively an absorber tube with fins operates thermally and hydraulically in a solar parabolic trough collector (PTC). A 3D computer model of the actual absorber tube shape and characteristics (15 mm inner diameter, 60 mm outer diameter, 2000 mm length, 7 mm fin height and 5 mm fin thickness) was created to evaluate the temperature of the absorber tube surface, the velocity of the heat transfer fluid (HTF), and the temperature of both the HTF and the absorber tube wall during operation under different velocities at the point of HTF entering the tube. Increasing the HTF velocity created improved rates of convective heat transfer from the absorber tube and decreased the temperature of the absorber tube surface (the maximum surface temperature decreased from 80.6°C to 74.9°C showing the increase in cooling). In addition, all cases showed that HTF temperature increased along the length of the absorber tube; therefore, the temperature of the HTF at the outlet of the tube decreased with decreasing velocity. These findings indicate that the HTF velocity has a significant impact on improving the thermal performance of absorber tubes in solar PTCs.
The industrial sector is the largest contributor to anthropogenic (fabricated) thermal energy waste through heat energy loss, contributing an estimated 25–40% of the total energy from fuels consumed by industries to the environment in the form of low-quality heat waste. In this paper we have compiled an extensive database of the sectoral heat emission profiles for eight major industries (iron and steel, cement, chemicals, petroleum refining, glass manufacturing, pulp and paper, food processing, textiles), and evaluated the technical and economic feasibility of twelve different technologies for recovering and mitigating industrial heat waste. A new scoring matrix that incorporates thermal recovery efficiency, the capital cost of implementation, the length of payback period associated with each option, the potential for CO2 emission reductions from each option, and the complexity of implementing each option was created to evaluate the available mitigation options for industrial heat waste. The scoring matrix was used to rank the available mitigation opportunities for industrial heat waste, and develop projections of the combined effects of the mitigation options for three decarbonization scenarios (business-as-usual, moderate adoption, accelerated adoption) through the year 2040. The findings demonstrate that implementing a combination of high-efficiency heat exchangers, organic Rankine cycle (ORC) systems, and industrial heat pumps across 45% of industries could achieve a reduction in global industrial thermal emissions by 31.4% by 2035, which translates to a reduction of 2.1 Gt CO2-eq annually.
In this document, an advanced intelligent energy storage management system solution for clean energy sources from solar power will be outlined through the use predictive controls. The integrated intelligent energy management architecture includes a solar panel array, battery bank, local users, two-way converter, and utility providers. Using the proposed control scheme, battery storage charging and discharging cycles will be determined based on solar generation and user load forecasts that comply with specified operating constraints (SOC, power limits, and depth of discharge). The goal of this research project is to minimize dependence on the grid as well as reduce the amount of wear-and-tear placed on the batteries due to storing and using renewable energy. After completing this study, it is expected that the findings will greatly enhance the amount of energy available to users from renewable energy sources as well as improving the reliability of these systems even during time periods when the operated equipment would normally experience significant variations in performance. This study also demonstrates how predictive battery management could assist with integrating renewable energy systems into the existing infrastructure used for producing electricity.
In this paper, a numerical analysis of a latent heat thermal energy storage (TES) system that utilizes a phase change material (PCM) is presented. A two-dimensional geometric model of the storage unit was created within COMSOL Multiphysics, and a transient heat transfer analysis was performed using the finite element method. The model was utilized to determine the temperature distribution within the TES domain and the thermal response of the storage material during the charging cycle. The data indicated that as time progressed, the temperatures within the TES increased steadily, which is indicative of continuous heat transfer and good thermal storage capabilities. At 15 hours after the beginning of the charging cycle, while the temperatures within the storage area are low and not uniform, after 20 to 25 hours, temperatures are seen to be much higher and much more uniform across the storage domain. The temperatures were uniformly distributed around 60 °C, which is indicative of a phase change and the absorption of latent heat. At 30 hours the TES is charged, with the maximum temperature for the TES approaching 348 K. The temperature histories at 0.06 m and 0.09 m from the base of the TES confirm very stable charging characteristics with very little variation between the two locations. The results show that PCM based thermal energy systems are viable for effective thermal energy storage purposes.
Nanofluids colloidal dispersions of engineered nanoparticles in conventional heat transfer fluids have emerged as promising candidates for next-generation thermal management systems due to their superior thermophysical properties. However, the widespread deployment of nanofluids raises substantial concerns regarding nanoparticle ecotoxicology, occupational health hazards, and environmental lifecycle impacts that remain insufficiently characterized. This investigation presents a systematic comparative analysis of eight commercially relevant nanofluid systems (TiO2, Al2O3, CuO, SiO2, ZnO, Fe3O4, CNT, and graphene-based) with respect to cytotoxicity, aquatic ecotoxicity, particle stability and sedimentation, and full lifecycle environmental burden assessed through ISO 14040/14044-compliant life cycle assessment (LCA) methodology. Results demonstrate that CuO and CNT-based nanofluids exhibit the highest ecotoxicological potential, whereas SiO2 and Fe3O4 systems display comparatively benign environmental profiles. LCA results indicate that nanoparticle synthesis contributes 42–68% of total lifecycle environmental burden, suggesting that green synthesis pathways represent the highest-priority intervention for environmentally responsible nanofluid deployment.
This study performed a 3D CFD of the internal flow characteristics of a centrifugal pump using COMSOL Multiphysics. The physical geometry of the pump was developed as a complete 3D representation including the pump casing, impeller with five blades, inlet passage and outlet region. This study used a numerical simulation to characterize the velocity field, streamline configuration and pressure distribution within the pump at several operating conditions defined by the impeller speed (1000 rpm) and inlet pressure/flow parameters. The results indicate that the highest velocity area within the pump occurs within the impeller region where the transfer of mechanical energy between rotating impeller blades and fluid occurs. Additionally, the configuration of the streamlines also indicates that there is significant swirling and recirculating flow patterns present within the casing of the pump. Furthermore, when analysing the velocity and pressure profiles across the entire arc length between the pump inlet and exit, the outlet region has significantly more variability than the inlet region. Although the inlet region has stable flow characteristics, the outlet region exhibits considerable variability due to blade rotation and casing/flow interaction. Overall, CFD model can adequately replicate key features of the hydrodynamic behaviour associated with centrifugal pump operation, and therefore it can be used to support the design of another centrifugal pump.
The major factors contributing to reduced efficiency and early failure of thermal power plant (TPP) equipment are scaling and corrosion product deposits on heat transfer surfaces. A mechanistic, thermodynamic, and kinetic analysis of scaling and corrosion processes will be presented, as well as an experimental demonstration of a novel, iron scrap reactive device installed within the de-aerating tank to enhance removal of dissolved gases. Inverse solubility of CaCO3, CaSO4, MgCO3, Mg(OH)2, CaSiO3, and MgSiO3 has been characterised from 20 to 350 degrees C by means of solubility product (Ksp) analysis. Utilising a quantitative fuel consumption model, it has been found that for every 1 mm of CaCO3 scaling, there is a corresponding increase in natural gas consumption of 2.8%, an increase of 7.8% at 5 mm scaling. Electrochemical corrosion mechanisms due to O2 and CO2 will be studied in detail with a thermodynamic and kinetic perspective. The iron scrap device utilises 12 competing heterogeneous reactions to chemically absorb dissolved O2 and CO2 prior to thermal expulsion, lowering dissolved O2 by 19-20% and CO2 by 17-18% when compared to conventional thermal de-aeration, while reducing hydrazine consumption by approximately 20%. As such, the combined thermal and chemical de-aeration strategy results in an annual net savings of $185,000-$210,000 per 100 MW unit, yielding a payback period of less than one month.
The results of a numerical study of the hydrodynamics of several geometries of jet pumps (using COMSOL Multiphysics) are presented in this paper. Finite element modelling principles were used to construct a model for the analysis of the effect of the nozzle outlet diameter (d0) and the diffuser outlet diameter (d2) on the internal flow behaviour. Two cases were considered: d0=10 mm and d2=40 mm, and d0=15 mm and d2=50 mm. The simulations were completed with a specified inlet pressure of 101325 Pa and with the effect of gravity included. The velocity and pressure distributions at certain monitoring locations along with the entire jet pump domain were determined. It was determined that the geometric variant has a substantial effect on the development of the jet, behaviour of entrainment, and the recovery of pressure within the pump. The flow reached a steady state quickly after a short period of transients for both variants. The larger diameter variant provided larger velocities at several monitoring locations and a greater negative pressure region, thus providing greater suction and different momentum exchange. This study indicated that the diameter ratio is the most important factor for the performance of jet pumps and therefore will assist in the design of effective and optimal jet pump systems.
The short-term aerodynamic behaviour of a vertical axis wind turbine (VAWT) will be investigated through numerical simulation (using 3D COMSOL Multiphysics modelling). To adequately account for the unsteady nature of air around the rotating blades, a time-dependent rotational simulation needs to be employed to capture the time-varying nature of unsteady flows caused by blade rotation and the angle of attack. The incompressible Navier-Stokes equations were then solved to solve how the flow behaves throughout the entire computational domain including a cylindrical sub-domain rotating with the Arbitrary-Lagrangian-Eulerian (ALE) method. During the torque development, the angular velocity and mechanical power of the VAWT were determined to have a constant cyclical behaviour while the rotor accelerated, generating a series of rotational positions (i.e. the torque had rapidly increased and then became steady state, thus exhibiting periodic oscillations relative to the blades’ rotational positions). Velocity contours also showed significant velocity gradients at the leading edges of the blades due to wind energy extraction from the turbine, as well as development of wake structures downstream of the turbine.
