Temperature Prediction Research Articles

Machine learning aided prediction of martensite transformation temperature of NiTi-based shape memory alloy

The martensitic transformation temperature (Ms) of shape memory alloys plays a crucial role in their performance. To achieve rapid prediction of Ms in shape memory alloys, this study employs machine learning techniques, using NiTi-based SMAs, the most widely used type, as an example. The results show that the gradient boosting machine algorithm provides the best prediction accuracy, with an R² of 0.92 and a mean absolute error (MAE) of 23.42 °C on the test set. Through correlation analysis, recursive feature elimination, and exhaustive search methods, six key features were identified. Subsequently, SHapley Additive exPlanations (SHAP) values were introduced to analyze the model interpretability, revealing the feature importance ranking and the dependence relationship between features and SHAP values. To address the issue of low Ms in NiTi-based shape memory alloys, the concept of high-entropy alloys was introduced. The study designed a TiZrHfNiCu high-entropy shape memory alloy and efficiently screened alloy compositions using the predictive model. This approach successfully increased the Ms, with the Ti19Zr19Hf19Ni37Cu6 alloy exhibiting an Ms exceeding 400 °C, significantly enhancing high-temperature application performance.

Melt flow analysis in rotational nozzle fused filament fabrication process

Fused filament fabrication (FFF) is a widely used processing method; however, heat transfer limitations within a conventional nozzle result in relatively low flow rates, leading to lengthy production times, compared to traditional processing methods, ultimately restricting its industrial application. Recently, a novel rotational nozzle FFF three-dimensional (3 D) printer has been patented and developed to enhance processing efficiency. Despite this achievement, the fundamental mechanisms behind this novel process remain unclear. In this study, both analytical analysis and numerical simulations were conducted based on a force-controlled scaled-down experimental setup. This setup, designed according to the pressure-induced melt removal theory, provided melt throughput data under varying heater temperatures, extrusion forces, and rotational speeds. Agreement between the modeling and experimental results confirms the generalizability of the models. Modeling predictions of temperature and velocity distributions indicate that viscous dissipation affects the average temperature and filament velocity. To simulate the real-world working conditions of FFF 3 D printing, a velocity-controlled simulation was introduced. It was observed that the average melt film thickness increases with nozzle rotational speed due to viscous dissipation. Additionally, the extrusion force required for the same printing speed decreases with increasing nozzle rotational speed, primarily due to the higher shear rate reducing melt viscosity.

Optimization of temperature prediction algorithms and simulation of future neighborhood-scale thermal environments in Chongqing

With climate change and urbanization, predicting the outdoor thermal environment of residential areas has become increasingly crucial, particularly during extreme weather conditions. Existing studies lack a comprehensive approach to optimizing temperature prediction algorithms for Chongqing and simulating extreme thermal environments at the neighborhood scale for future decades. This study employed various algorithms, including Levenberg-Marquardt optimization, Monte Carlo simulation, ARIMA modeling, and SARIMA prediction, to simulate and forecast temperature data for the central urban area of Chongqing from 2014 to 2023. By assessing metrics such as root mean square error (RMSE), R-squared (R2), and mean absolute percentage error (MAPE), the most effective prediction algorithm was identified. Results indicate that the Levenberg-Marquardt optimization algorithm, known for its nonlinear curve fitting capabilities, demonstrated superior performance in both the test and training sets, achieving an RMSE of 0.82 °C, MAPE of 1.81 %, and R2 of 0.75. This algorithm was subsequently used to forecast the outdoor temperature trend in Chongqing from 2024 to 2050, while also simulating the outdoor wind field, temperature distribution, building surface temperatures, and wind speeds in a residential district at neighborhood scale for 2030, 2040, and 2050 at a building density of 40 %. Findings reveal that peak pedestrian-level temperatures could reach 45.62 °C, and peak roof surface temperatures could reach 93.93 °C under extreme conditions. Furthermore, the study proposes practical recommendations for building layouts and cooling strategies. These findings are expected to improve residential planning in Chongqing, enhance residents' quality of life, and provide insights for other regions facing future temperature changes.

Edge of Transfer Learning-Based Long Short-Term Memory Neural Networks in the Application of Battery Surface Temperature Prediction for Electric Vehicles

Edge of Transfer Learning-Based Long Short-Term Memory Neural Networks in the Application of Battery Surface Temperature Prediction for Electric Vehicles

Advancing thermohydraulic performance of micro-grooved flat heat pipes through a combined numerical-experimental approach

A combined numerical-experimental model has been developed to predict the thermohydraulic characteristics of micro-grooved flat heat pipes (FHPs). This model uses a limited number of experimental data for calibration with two parameters, βexp,1 and βexp,2, which account for uncertainties in the accommodation coefficients used to determine evaporation and condensation mass fluxes. This calibration, which is a key feature of the developed model, addresses the uncertainties and challenges in determining these coefficients. Moreover, the model incorporates the effects of interfacial shear stress, axial wall conduction, heat transfer in the liquid block region, and the contact angle of the liquid–vapor interface. Furthermore, the evaporation mass flux from the curved liquid–vapor interface is calculated using a separate multiscale approach developed in two recent works for analyzing evaporation from the extended meniscus in microchannels, considering thin-film evaporation. The model demonstrates high accuracy in predicting the maximum heat transport capacity (Qmax) and wall temperature distribution across various input heat powers, with a maximum error of less than 0.5 % in wall temperature predictions, as validated against experimental data. The validated model is then used to explore a new micro-grooved wick structure featuring converging/diverging microchannels. In this structure, the channel width is constant and equal to Wf2 in the first half of the heat pipe, from the condenser end to the midpoint, and then it decreases or increases to Wf1 by the end of the evaporator section. Results indicate that a flat heat pipe with a converging micro-grooved wick structure (Wf1 = 200 µm and Wf2 = 160 µm) can enhance Qmax from 53 to 70 W, 70.8 to 91.8 W, and 91.2 to 116.5 W at working temperatures of 60 °C, 70 °C, and 80 °C, respectively. This means the new converging micro-grooved wick structure can boost flat heat pipe performance by more than 25 % compared to a flat heat pipe with a straight micro-grooved wick (Wf1 = Wf2 = 200 µm) without increasing the occupied space.

TE-LSTM: A Prediction Model for Temperature Based on Multivariate Time Series Data

In the era of big data, prediction has become a fundamental capability. Current prediction methods primarily focus on sequence elements; however, in multivariate time series forecasting, time is a critical factor that must not be overlooked. While some methods consider time, they often neglect the temporal distance between sequence elements and the predicted target time, a relationship essential for identifying patterns such as periodicity, trends, and other temporal dynamics. Moreover, the extraction of temporal features is often inadequate, and discussions on how to comprehensively leverage temporal data are limited. As a result, model performance can suffer, particularly in prediction tasks with specific time requirements. To address these challenges, we propose a new model, TE-LSTM, based on LSTM, which employs a temporal encoding method to fully extract temporal features. A temporal weighting strategy is also used to optimize the integration of temporal information, capturing the temporal relationship of each element relative to the target element, and integrating it into the LSTM. Additionally, this study examines the impact of different time granularities on the model. Using the Beijing International Airport station as the study area, we applied our method to temperature prediction. Compared to the baseline model, our model showed an improvement of 0.7552% without time granularity, 1.2047% with a time granularity of 3, and 0.0953% when addressing prediction tasks with specific time requirements. The final results demonstrate the superiority of the proposed method and highlight its effectiveness in overcoming the limitations of existing approaches.

Tracking Differentiator-Based Identification Method for Temperature Predictive Control of Uncooled Heating Processes

The temperature control of uncooled heating processes presents challenges due to a substantial lag and the absence of active cooling mechanisms, which can lead to overshoot and oscillations. To address these issues, we propose an anti-disturbance identification method based on a tracking differentiator (TD) and an input-constrained temperature predictive control (ICTPC) strategy. Our approach specifically considers the impact of unknown disturbances on model identification within a second-order heating process. By employing a TD to differentiate the input and output signals, we effectively minimize the identification error caused by low-frequency disturbances, yielding a robust anti-disturbance identification technique. Following this, we establish input constraints to limit the amplitude and variation of the control input, ensuring a more controlled and predictable system response. Using the identified model, an ICTPC algorithm is designed to achieve stable temperature control in uncooled heating processes. Experimental results from a typical uncooled heating system demonstrate that our method not only significantly reduces overshoot but also effectively mitigates temperature fluctuations, leading to enhanced control performance and system stability. This study provides a practical solution for temperature control in systems without cooling capabilities, offering substantial improvements in the efficiency and quality of industrial production processes.

A coordination attention residual U-Net model for enhanced short and mid-term sea surface temperature prediction

Sea surface temperature (SST) is crucial for studying global oceans and evaluating ecosystems. Accurately predicting short and mid-term daily SST has been a significant challenge in oceanography. Traditional deep learning methods can handle temporal data and spatial features but often struggle with long-range spatiotemporal dependencies. To address this, we propose a coordination attention residual U-Net(CResU-Net) model designed to better capture the dynamic spatiotemporal correlations of high-resolution SST. The model integrates coordinate attention mechanisms, multiple residual modules, and depthwise separable convolutions to enhance prediction capabilities. The spatiotemporal variations of SST across different areas of the South China Sea are complex, making accurate predictions challenging. Experiments across various regions of the South China Sea show the model’s effectiveness and robust generalization in predicting high-resolution daily SST. For a 10-day forecast period, the model achieves approximately 0.3 °C in RMSE, outperforming several advanced models.

Evaporation temperature prediction of the refrigerant-direct convective-radiant cooling system based on LSTM neural network

The existing radiant cooling systems are effective at handling indoor sensible heat load. However, the indoor latent load needs to be treated with an additional dehumidification system, which will increase the complexity and initial investment of the system. To settle this dilemma, a novel aluminum column-wing type refrigerant-direct convective-radiant cooling (ACT-RCC) system was proposed, which adopted refrigerant as direct medium and provided cooling and independent dehumidification simultaneously. Experiments were conducted in an enthalpy difference laboratory with the indoor air temperature of 22.0 °C − 30.0 °C and relative humidity of 60.0%. A numerical model of the ACT-RCC terminal was established and validated with the experimental data. To meet the indoor heat and latent load of residential buildings, a heuristic approach based on Long Short-Term Memory (LSTM) neural network for predicting the evaporation temperature of this system was proposed. Results showed that the predicted values were accurate enough by measured data and the maximum deviation of the evaporation temperature was 0.7 °C. The relationship between the evaporation temperature, indoor air temperature and relative humidity, and heat moisture ratio was derived. Besides, the indoor thermal comfort of the selected bedroom was also evaluated. Results showed that when the indoor air temperature and relative humidity remained 26.0 °C and 40.0% − 70.0%, the thermal comfort of this bedroom can achieve Level Ⅰ by adjusting the evaporation temperature of the ACT-RCC system of 2.0 °C − 19.8 °C.

Modelling and experimental investigation of cooling of field-operating PV panels using thermoelectric devices for enhanced power generation by industrial solar plants

The performance of commercial solar power plants degrades due to an increase in module temperatures for which standard PV-T air or water-cooling techniques are mostly used. In this study, a thermoelectric cooling system is studied for improving photovoltaic cell power efficiency and hence solar power generation. The cooling optimization requires solar cell temperature prediction of field operating PV modules, for which analysis of six models, is presented. The experimentation results show that TEC cooling maintains PV cell at 25 °C whereas PV cell without TEC operates at 55–63 °C, a higher temperature range, showing the effectiveness of the thermoelectric cooling system in precisely controlling PV cell temperature to operate at or near STC conditions in the field creating a temperature difference of 30–38 °C. The NOCT and Faiman model results are found close to the experimental values in comparison to other models. The potential for cooling and a corresponding increase in solar plant energy production is assessed using PV Syst modeling and simulation for three practical PV installation scenarios for 31 different climatic zone locations worldwide showing 6–27 % power loss due to elevated temperatures, which is not studied in previous studies adding novelty to the analysis. The results show that PV-TECS is an effective system to control the temperature of field operating PV modules, which can be used in future photovoltaic power plants. Field results and analysis of PV temperature models is crucial for the optimization and future development of PV-thermoelectric systems deployed under actual outdoor conditions as well as the expected cooling gains in different climatic locations. These aspects are collectively studied in the current work adding to the novelty of the study.

Study on the influencing factors of asphalt mixture compaction quality test method based on simulation and experimentation

The compaction quality and compaction temperature of asphalt pavement during construction are crucial to the durability of pavement. However, the current methods for testing the compaction quality and compaction temperature of pavement have the drawbacks of low test efficiency and inability to truly reflect the compaction state. Therefore, this study investigated the factors affecting the test method of asphalt mixture compaction quality and compaction temperature based on simulation and experimentation. The effect of environmental factors on the surface and internal temperature of mixture was revealed based on temperature measurement equipment. Meanwhile, the dynamic analysis and finite element simulation were used to investigate the effect of temperature, compactness and base modulus on the compaction quality test index. Subsequently, the compaction quality prediction model was constructed based on support vector machine. Based on that, the effect of temperature, compactness and base modulus on compaction quality was verified based on indoor vibration compaction experiment. The results clearly showed that the function with average relative deviation of 1.7 % including environmental factors was established as compaction temperature prediction model based on support vector machine. The vibration acceleration amplitude was chosen as the compaction quality test index based on dynamic analysis and numerical simulation. The effect of temperature on the relationship between compactness and vibration acceleration amplitude was more obvious when the compactness was larger and the temperature was lower. Compared with the flexible base modulus, the semi-rigid base modulus had more influence on compaction quality. The results of vibration compaction experiment verified the feasibility of compaction quality prediction model constructed based on numerical simulation, and the average relative deviation of prediction model was only 0.44 %.

Power Battery Temperature Prediction Based on Charging Strategy Classification and Improved Adaptive GA-BP

Power Battery Temperature Prediction Based on Charging Strategy Classification and Improved Adaptive GA-BP

CLIMATE PULSE INDIA –AN APPLICATION TO PREDICT THE CLIMATE CHANGES

Climate Pulse India is a comprehensive web platform that helps users monitor and predict weather conditions in any city, focusing on essential factors such as temperature, humidity, and wind speed. By analyzing these elements, the application classifies cities into three safety zones: red (high danger), orange (moderate risk), and green (safe). The application is powered by the ClimatePulse - API, which delivers real-time weather data that fuels its prediction and forecasting capabilities. Designed with user accessibility in mind, Climate Pulse India provides weather insights, allowing users to assess the conditions of their hometown or any travel destination. With this information, users can make informed decisions regarding safety and travel. The predictions are presented on the website in a straightforward and easy- to-understand format. This paper demonstrates about providing accurate weather predictions for any location across India. Once predictions are made, the results are displayed directly on the webpage for the user. KEYWORDS: Weather forecasting, Safety zone classifications, Wind speed analysis, Risk assessment, Temperature prediction, Humidity monitoring

Development of a liquid metal subchannel code applied to ocean conditions and study on the effect of core geometry parameters and ocean motions on flow and heat transfer characteristics

The Generation IV International Forum (GIF) proposes a type of liquid metal reactor, due to its thermal properties and efficient heat transfer in a relatively small system. This special heat transfer characteristic allows the liquid metal reactor to be modular in design and flexible in installation on a nuclear-powered ship. Accurate prediction of the coolant temperature and flow distribution between subchannels is important to ensure nuclear safety in the complex ocean environment. In this study, an ocean version of subchannel code is developed by adding the motion terms in the momentum equations for the liquid metal reactor core based on the previously land-based subchannel code. Since the research on the characteristics of flow resistance and heat transfer mechanism under ocean conditions is not mature, the original empirical model is still used in the ocean version code. The constitutive models of the liquid metal are analyzed and incorporated into the ocean code. The code has been verified and validated by comparing with Computational Fluid Dynamics (CFD) simulation results, and experimental data. The effect of geometric parameters such as rod Pitch to Diameter ratio (P/D), Rod to Wall gap (RTW) is studied in the liquid metal lead 7 rod bundles. In addition, the effect of heaving start-up and shutdown motions between liquid metal lead and water is discussed in the 5 × 5 bare rod bundles. In general, the modified subchannel analysis code can well complete the calculation of liquid metal reactor under ocean conditions, as well as provide support for the design and safety analysis of a liquid metal cooled fast reactor.

STFM: Accurate Spatio-Temporal Fusion Model for Weather Forecasting

Meteorological prediction is crucial for various sectors, including agriculture, navigation, daily life, disaster prevention, and scientific research. However, traditional numerical weather prediction (NWP) models are constrained by their high computational resource requirements, while the accuracy of deep learning models remains suboptimal. In response to these challenges, we propose a novel deep learning-based model, the Spatiotemporal Fusion Model (STFM), designed to enhance the accuracy of meteorological predictions. Our model leverages Fifth-Generation ECMWF Reanalysis (ERA5) data and introduces two key components: a spatiotemporal encoder module and a spatiotemporal fusion module. The spatiotemporal encoder integrates the strengths of convolutional neural networks (CNNs) and recurrent neural networks (RNNs), effectively capturing both spatial and temporal dependencies. Meanwhile, the spatiotemporal fusion module employs a dual attention mechanism, decomposing spatial attention into global static attention and channel dynamic attention. This approach ensures comprehensive extraction of spatial features from meteorological data. The combination of these modules significantly improves prediction performance. Experimental results demonstrate that STFM excels in extracting spatiotemporal features from reanalysis data, yielding predictions that closely align with observed values. In comparative studies, STFM outperformed other models, achieving a 7% improvement in ground and high-altitude temperature predictions, a 5% enhancement in the prediction of the u/v components of 10 m wind speed, and an increase in the accuracy of potential height and relative humidity predictions by 3% and 1%, respectively. This enhanced performance highlights STFM’s potential to advance the accuracy and reliability of meteorological forecasting.

HELI: An Ensemble Forecasting Approach for Temperature Prediction in the Context of Climate Change

HELI: An Ensemble Forecasting Approach for Temperature Prediction in the Context of Climate Change

A Prediction Model for Pressure and Temperature in Geothermal Drilling Based on Physics-Informed Neural Networks

With the global expansion of geothermal energy, accurate prediction of pressure and temperature during drilling has become essential for ensuring the safety and efficiency of geothermal wells. Traditional numerical methods, however, often struggle to handle complex wellbore environments due to their high data demands and limited computational accuracy. To address these challenges, this paper introduces an innovative predictive model based on Physics-Informed Neural Networks (PINNs). By integrating physical laws with deep learning, the model theoretically surpasses the limitations of conventional methods. Trained on pressure and temperature data from a geothermal well in the Xiong’an area, the model demonstrates exceptional accuracy and robustness. Additionally, the model was rigorously tested under extreme wellbore conditions, showcasing its strong generalization capabilities. The findings suggest that PINNs offer a highly advantageous solution for geothermal drilling, with significant potential for practical engineering applications.

Predicting the subcutaneous temperature in cryolipolysis using deep operator networks

Accurate monitoring of subcutaneous temperature is crucial for the safety and efficacy of cryolipolysis. However, existing measurement and simulation methods often require trade-offs between accuracy, depth, and computational efficiency. This study introduces a novel deep learning architecture, ConvD-DeepONet, specifically designed to predict subcutaneous temperature fields with both high accuracy and efficiency. The model effectively captures spatial information and produces multi-dimensional output, owing to the innovative integration of convolutional layers and the decoder network. An average absolute error (MAE) of 0.0038 ℃ and a root mean square error (RMSE) of 0.0083 ℃ are achieved, resulting in over a 50 % reduction compared to the baseline models. Moreover, each prediction is completed in just 5.9 ms, rendering it 120 times faster than traditional finite element method simulations. These results indicate that ConvD-DeepONet is a promising tool for real-time subcutaneous temperature prediction, with the potential to enhance the safety and efficacy of cryolipolysis.

Structural optimization and battery temperature prediction of battery thermal management system based on machine learning

Lithium-ion batteries significantly extend the driving range for electric motorcycles. The battery thermal management system (BTMS) is critical for achieving optimal battery performance. Moreover, precise battery temperature prediction is essential for efficient thermal management. Therefore, a battery thermal management system integrating air and phase change material (PCM) cooling is proposed. Initially, the impact of PCM height, PCM thickness, and air velocity on battery temperature is analyzed. Subsequently, with cost minimization as the objective and ensuring that the maximum battery temperature remains below a threshold, the Black Kite Algorithm (BKA) is employed to optimize the BTMS structure. Finally, a BKA-Convolutional Neural Network (CNN)-Self Attention (SA) model is introduced for battery temperature prediction. The results indicate that increasing the thickness of the PCM and air velocity facilitates battery heat dissipation but with diminishing marginal effects. An increase in PCM height enhances battery cooling at low air velocities but becomes detrimental at high air velocities. The optimized PCM height is 35 mm, resulting in a cost of 0.073 USD for the BTMS per battery. Additionally, the BKA-CNN-SA model achieved a maximum error of 0.45 °C on the validation set and accurately predicted battery temperature changes before and after PCM melting.

Parameter optimization of thermal network model for aerial cameras utilizing Monte-Carlo and genetic algorithm

It is crucial to precisely calculate temperature utilizing thermal models, which require the determination of thermal parameters that optimally align model outcomes with experimental data. In many instances, the refinement of these models is undertaken within space instruments. This paper introduces an optimization methodology for thermal network models, with the objective of enhancing the accuracy of temperature predictions for aerial cameras. The investigation of internal convective heat transfer coefficients for both cylindrical and planar structures provides an estimation of convective thermal parameters. Based on the identification of thermally sensitive parameters and the reliability evaluation of transient temperature data through the Monte-Carlo simulation, the genetic algorithm is employed to search for global optimal parameter values that minimize the root mean square error (RMSE) between calculated and measured node temperatures. As a result, the optimized model shows significantly improved accuracy in temperature prediction, attaining an RMSE of 1.07 ℃ and reducing the maximum relative error between predicted and experimental results from 33.8 to 3.1%. Furthermore, the flight simulation and thermal control experiments validate the robustness of the optimized model, demonstrating that discrepancies between the observed and predicted temperatures are within 2 °C after re-correcting the external convection heat transfer coefficient value.