Temperature Prediction Errors Research Articles

Accurate Oil Temperature Prediction Model and Oil Refilling Parameters Optimization for Hydraulic Closed-Circuit System

Oil temperature plays a crucial role in hydraulic closed-circuit systems (HCS), and conventional thermal equilibrium models and coupled simulation models face challenges in terms of accuracy, efficiency, and cost when calculating oil temperature. This study introduces an innovative HCS oil temperature precise prediction model and oil refilling parameter optimization method. The initial sample space was determined through a Sobol sensitivity analysis and improved Latin hypercube sampling, leading to the development of a combinatorial agent model (CAM) suitable for oil temperature prediction with superior accuracy and stability compared to other methods. Based on CAM, the optimal oil refilling flow rates under various operational conditions are computed. To validate the efficacy of the theoretical analysis, an HCS experiment platform was established. The data indicates that the temperature prediction error range of the CAM model falls between 0.30 °C and 1.05 °C, and optimizing the oil refilling flow rate can effectively enhance system efficiency while ensuring that the oil temperature remains within permissible limits. The research methodology and findings are applicable in engineering practice and can be extended to optimize the design of other hydraulic systems.

Selection of suitable geothermometers for predicting the subsurface temperatures with higher accuracy: A study based on globally distributed geothermal field data

Subsurface reservoir temperature of a geothermal province is popularly estimated using various geothermometers but it is necessary to examine the suitability of them. Present work studies the performances of five popular geothermometers Na-K, Na-K-Ca, Quartz, Chalcedony and K-Mg to forecast the reservoir temperature of 148 geothermal springs located in different parts of the world. The predicted temperature of the geothermometers have been compared with the bottom whole temperature (BHT) of the wells drilled in those geothermal fields to check the accuracy of the predictions.The results show that performance of a particular geothermometer is significantly affected by the type of rocks in a geothermal reservoir. Na-K geothermometer provides highest accuracy for the basaltic and sandstone reservoir rocks with average errors 15 % and 11 % respectively. The quartz geothermometer also provides almost equally good prediction for sandstone. The best result for a granitic reservoir rock is obtained by Na-K-Ca with an average error about 15 %. The carbonate reservoir rocks are the most difficult reservoir rock to use any geothermometer as lowest error of prediction using any geothermometer is about 26 %. The chalcedony performed the poorest among all geothermometers and is not recommended for any reservoir rock types under considerations. Thus, the study on global geothermal fields shows that the reservoir rock type must be considered to select a geothermometer else it may cause very high error in prediction of reservoir temperature. However, none of the geothermometer could produce error less than 10 %. It is worth noting that the present study though establishes the suitability of a geothermometer for a particular lithology, it does not explore the rationale for such a suitability.

An ANN Hardness Prediction Tool Based on a Finite Element Implementation of a Thermal–Metallurgical Model for Mild Steel Produced by WAAM

WAAM has emerged as a promising technique for manufacturing medium- and large-scale metal parts due to its high material deposition efficiency and automation level. However, its high heat accumulation and complex thermal evolution strongly affect the resulting microstructures and mechanical properties. The heterogeneous and unpredictable nature of these properties hinder the widespread application of WAAM in the steel construction industry. In this study, an artificial neural network (ANN) hardness model is developed, based on a thermal–metallurgical model for mild steel. The objective is to establish non-linear relationships between the input process parameters and the desired output, i.e., hardness. The thermal–metallurgical model utilizes a well-distributed heat source model, a death-and-birth algorithm, and a metallurgical model to simulate the temperature field and to calculate the microstructure phase fraction. The temperature prediction errors at four thermocouple positions are mostly below 20%. Because of the limited experimental data, twenty-five simulation experiments are performed using the L25 orthogonal array based on the Taguchi method. The analysis of variance (ANOVA) reveals that the travel speed has the greatest impact on hardness. With the dataset from the thermal–metallurgical model, an ANN model to predict hardness is developed. A comparison to experimental data shows excellent performance and accuracy, with the Mean Absolute Percentage Error (MAPE) of ANN predictions within 10% of the targeted hardness.

A Standardised Potentiometric Method for the Effective Parameterisation of Reversible Heating in a Lithium-Ion Cell

Lithium-ion batteries generate heat, degrading faster and becoming unsafe at high temperature. Yet many battery models do not consider the contribution of reversible, entropic heating when evaluating the rate of heat generation from a cell or battery pack. This leads to temperature prediction errors in battery management systems, increased safety risk, and reduced lifetime of the battery pack. Here, a standardised potentiometric method is proposed, allowing anyone with access to a typical battery lab to reliably and accurately extract the entropy coefficient for any electrochemical cell, the key parameter for the inclusion of reversible heating in a battery model. The proposed method takes 7.4 days to complete, representing a reduction of 90% compared to some methods proposed in the literature. Results highlight the importance of moving away from the multiple temperature steps, and the temperature step increases that dominate the existing literature. These arguments are justified through the observation and introduction of voltage relaxation following both kinetic and thermal excitation. These phenomena are termed post-kinetic-potentialisation and post-thermalisation-potentialisation. Post-thermalisation-potentialisation is not discussed in any published literature yet represents an important behavioural trait for any lithium-ion cell with a non-negligible length scale and thermal diffusivity.

Research on the thermal flow characteristics of viscosity oil in hydrodynamic torque converter

The increase in power density of hydrodynamic torque converters (HTCs) leads to a sharp rise in temperature within flow channels, affecting the reliability. In order to accurately predict the thermal effect and temperature distribution characteristics of the HTC internal viscosity oil, a multi-physics computational fluid dynamics (CFD) model is proposed. A specialized test bench was established, and the macro and internal flow temperature data were obtained. HTCs with different working conditions and wheel sets were studied. The results indicate that CFD model considering energy equation can accurately predict the overall hydrodynamic performance and the flow field temperature characteristics under different rotating conditions. The prediction error of the overall temperature rise is within 4.92%, and the flow field temperature prediction error of the stator is under 14.3%. The hydraulic characteristics is improved by 6.02%. The analysis of internal flow and energy exchange characteristics indicates the thermal effects and temperature distribution mechanisms caused by energy loss in the flow field within the HTC. The study provides an effective computational model for the prediction and control of the heat generation of the HTC and enhances the depth of research on the flow mechanism of inhomogeneous flow fields caused by thermal effects.

Modelling Heat and Mass Transfer in High-Capacity Natural Convection Solar Dryers

Predicting solar dryer performance under different environmental conditions or assessing their performance to dry different grains is challenging since repeatable full-scale tests are expensive and time consuming. In the present study, computational fluid dynamics approach was used to model the drying of maize in high-capacity dryers such as greenhouse and solar bubble dryer. The absorption of short-wave radiation and the greenhouse effect in the dryer with incident solar radiation was modelled using a dual-band spectrum. The distribution of airflow, temperature, and absolute humidity was analysed in this study to optimise the drying process of maize. Additionally, these results were also used to quantify the drying rate of both greenhouse and solar bubble dryer. The greenhouse dryer model overpredicted the dryer temperatures by an average of 0.12%, and overpredicted absolute humidity by 0.38 per cent. The average Root-Mean-Square Error (RMSE) of temperature prediction was 1.8 °C, and the average RMSE for absolute humidity was 0.0042 for the greenhouse model. On the other hand, the solar bubble dryer model underpredicted temperatures by 1.7%, and underpredicted humidity values by 0.3 per cent. The mean absolute percentage error for the temperature and absolute humidity prediction of the solar bubble dryer model was 1.69% and 0.28%, respectively. The predicted and observed spatial variation in the temperature was similar for both dryers.

Real-Time Temperature Prediction of Power Devices Using an Improved Thermal Equivalent Circuit Model and Application in Power Electronics.

As a core component of photovoltaic power generation systems, insulated gate bipolar transistor (IGBT) modules continually suffer from severe temperature swings due to complex operation conditions and various environmental conditions, resulting in fatigue failure. The junction temperature prediction guarantees that the IGBT module operates within the safety threshold. The thermal equivalent circuit model is a common approach to predicting junction temperature. However, the model parameters are easily affected by the solder aging. An accurate temperature prediction by the model is impossible during service. This paper proposes an improved thermal equivalent circuit model that can remove the effect of solder aging. Firstly, the solder aging process is monitored in real-time based on the case temperatures. Secondly, the model parameters are corrected by the thermal impedance from chip to baseplate based on the linear thermal characteristic. The simulation and experimental results show that the proposed model can reduce the temperature prediction error by more than 90% under the same aging condition. The proposed method only depends on the case temperatures to correct the model parameters, which is more economical. In addition, the experimental and simulation analysis in this work can help students of power electronics courses have an in-depth knowledge of power devices' mechanical structure, heat dissipation principles, temperature distribution, junction temperature monitoring, and so on.

The Impact of Wind Power and Load Power Fluctuations on Energy Storage Sizing

The study presents a method of taking into account the impact of wind power and load power fluctuations on the energy storage sizing, comprised of batteries of identical capacity. To account the impact, two methods of calculating the difference between wind power generation and load consumption were presented over some time interval: 1st and 2nd order difference methods. Each of the methods can be parameterized and non-parameterized method with and without taking into account parameters respectively, where the parameters are: discharge current, required discharge duration, depth of discharge, battery capacity, Peukert’s constant, discharge time from 100% capacity, ambient temperature and wind power prediction error. Using the parameterized method compared allows to refine the value of the number of batteries. Using the 2nd order difference method compared to the 1st order difference method can significantly reduce the required number of batteries.

A transient multi-path decentralized resistance-capacity network model for prismatic lithium-ion batteries based on genetic algorithm optimization

Battery thermal management is crucial for preventing the safety issues of lithium-ion batteries. Due to the simple modeling and fast calculation speed, the thermal resistance-capacity (RC) network model is broadly applied in the design of battery thermal management systems. However, the simplification of heat flow paths and the lumped definition of thermal capacity in traditional RC models result in large temperature prediction errors, which fail to reflect the thermal response in complex and diverse heat transfer situations. To improve the prediction accuracy, a decentralized centroid multi-path RC network model is constructed for a typical prismatic lithium-ion battery. This novel model incorporates multiple heat flow paths with additional thermal resistances and legitimately decentralizes the lumped heat capacity to other surface center points, resulting in a more realistic thermal response. A genetic algorithm is employed to determine the unknown thermal resistances and heat capacities at the attributed nodes. Results show that compared to the traditional RC network model, the multi-path decentralized RC network model can reduce the temperature prediction error from 4.24 to 0.95 ℃. This more refined modeling approach extends the application scope of thermal resistance network models to more complex scenarios while maintaining efficient simulation speed, which enables the attainment of more accurate and reliable onboard temperature predictions for lithium-ion power systems.

Multistep ahead prediction of temperature and humidity in solar greenhouse based on FAM-LSTM model

Solar greenhouses offer favorable climatic environments for the production of off-season crops in northern China. Greenhouse temperature and humidity are critical environmental factors influencing crop growth; therefore, it is crucial to predict the temperature and humidity. In this research, we constructed a feed-forward attention mechanism- long-short term memory (FAM-LSTM) model for the multistep prediction of temperature and humidity in a solar greenhouse. The FAM-LSTM model considered both internal and external environmental factors influencing crop growth, where the factors include temperature, humidity, light, soil temperature, and soil moisture. Then, we performed corresponding experiments employing this model in predicting both temperature and humidity at prediction horizons of 12, 24, 36, and 48 h. The FAM-LSTM model stands out due to its high accuracy compared with different models. The temperature prediction error of the model at four prediction horizons is primarily within [−0.5, 0.9], [−2.1, 1.7], [−0.9, 2.7], and [−1.8, 2.9] ℃, respectively. The humidity prediction errors of the model at four prediction horizons are mainly within [−1.9, 1.8], [−4.4, 3.6], [−6.9, 2.4], and [−5.8, 4.4] %, respectively. This study offers an effective temperature and humidity prediction approach for precise environmental regulation and damage warnings in solar greenhouses.

Adaptive Machine Learning-Based Proactive Thermal Management for NoC Systems

Because of the high-complex interconnection in contemporary multicore systems, the network-on-chip (NoC) technology has been proven as an efficient way to solve the communication problem in multicore systems. However, the thermal problem becomes the main design challenge in the current NoC systems due to the high-diverse workload distribution and large power density. Therefore, proactive dynamic thermal management (PDTM) is employed as an efficient way to control the system temperature. Based on the predicted temperature information, the PDTM can control the system temperature in advance to reduce the performance impact during the temperature control period. However, conventional temperature prediction models are usually built based on specific physical parameters, which are usually temperature-sensitive. Consequently, the current temperature prediction models still result in significant temperature prediction errors. To solve this problem, a novel adaptive machine learning (ML)-based PDTM is proposed in this work. The adaptive ML-based PDTM first uses an adaptive single layer perceptron (ASLP), which is composed of a single-neuron operation and a least mean square (LMS) adaptive filter technology, to precisely predict the future temperature. Afterward, the proposed adaptive reinforcement learning (RL) is used to find the proper throttling ratio to control the system temperature. In this way, the proposed adaptive ML-based PDTM can adapt to the hyperplane of the temperature behavior of the NoC system and provide a proper temperature control strategy at runtime. Compared with related works, the proposed approach reduces average temperature prediction error by 0.2%–78.0% and improves the system performance by 2.4%–43.0% with smaller hardware overhead.

A Hybrid Model for Billet Tapping Temperature Prediction and Optimization in Reheating Furnace

To predict and optimize the billet heating process in reheating furnace for rolling mills, this paper proposes a hybrid model that combines data-driven model with traditional mechanism knowledge, abbreviated as HMDM. By examining the heat conduction mechanism, a billet temperature distribution equation is established. Then, the billet temperature distribution in each heating zone is calculated and spliced with the corresponding process parameters. The Stacked-AutoEncoder is utilized to extract the features of process parameters, and the Long Short Term Memory model is employed to predict the temperature. Finally, using the previous predictions, the parameters of the subsequent heating stage are optimized and adjusted during the heating process. The experimental results on the real steel plant verify the effectiveness of HMDM. For example, the temperature prediction error has been reduced to less than <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$4^{\circ }$</tex-math></inline-formula> C, and the number of billets with abnormal tapping temperature has been decreased by <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$42.9\%$</tex-math></inline-formula> .

Structure optimization and operation characteristics of metal gas storage device based on compressed air energy storage system

Compressed air energy storage (CAES) is a key technology for promoting the replacement of fossil fuels with renewable energy. Currently, CAES systems typically require an underground salt cavern, which is limited by geological and geographical conditions. To promote the rapid popularization of CAES, there is an urgent need to develop new gas storage devices with flexible layouts. Metal pressure vessel as a gas storage device is concerned with the benefits of high storage pressure and reliable operation. Considering both strength and fatigue, this study conducted a structural optimization of Tank-D2000 and Pipe-D400 to achieve the design operating life with the minimal quality. Furthermore, the optimized structure was used to study the operating characteristics of the gas storage device by the multi-domain fluid–solid thermal coupling method and the real gas model. The results indicated that the internal convective heat transfer coefficient of the gas storage device was a linear function of time during operation, whereas the external natural convective heat transfer coefficient almost remained constant. The pressure distribution was uniform; however, the temperature difference was large, particularly for D400. Analysis and prediction based on the theory of the polytropic process showed that the polytropic index of D400 was lower than that of D2000, so the heat transfer of D400 was more adequate and the effective capacity of gas storage device under the designed operating pressure range was larger. With an increase in flow, polytropic index increased and was lower at energy storage than at energy release. The pressure prediction error was basically within the range of ±4 %, and the temperature prediction error was within the range of ±3 %.

Simulation of thermoelastic coupling in silicon single crystal growth based on alternate two-stage physics-informed neural network

This paper proposes a alternate physics-informed neural network (PINN) with a two-stage training strategy to solve the thermoelastic coupling in the growth of Czochralski silicon single crystal. Particularly, neural networks with a two-stage training strategy are established for solving partial differential equations in the thermoelastic coupling model. At present, PINN still faces challenges in solving coupling models. Especially for the imbalance between and within equations in the coupling model. To solve this problem, this paper splits the model into two independent systems with their coupling term exchanged in the alternating iteration. Then, the imbalance inside a single solution of the coupling model is solved by the proposed two-stage training strategy, and the imbalance between solutions is avoided by alternative iterations. Experimental results show that the proposed alternate two-stage PINN (ATPINN) can accurately simulate the temperature and displacement in the thermoelastic coupling model. Compared with the finite element results, the error of temperature prediction and displacement prediction is acceptable. The numerical results demonstrate the feasibility and performance of the ATPINN. Furthermore, ATPINN can be extended to any type of coupling model, especially for coupling models with large differences in the magnitudes of the loss functions. The codes developed in this manuscript are publicy available at https://github.com/callmedrcom/ATPINN.

Recurrent neural network modeling of multivariate time series and its application in temperature forecasting.

Temperature forecasting plays an important role in human production and operational activities. Traditional temperature forecasting mainly relies on numerical forecasting models to operate, which takes a long time and has higher requirements for the computing power and storage capacity of computers. In order to reduce computation time and improve forecast accuracy, deep learning-based temperature forecasting has received more and more attention. Based on the atmospheric temperature, dew point temperature, relative humidity, air pressure, and cumulative wind speed data of five cities in China from 2010 to 2015 in the UCI database, multivariate time series atmospheric temperature forecast models based on recurrent neural networks (RNN) are established. Firstly, the temperature forecast modeling of five cities in China is established by RNN for five different model configurations; secondly, the neural network training process is controlled by using the Ridge Regularizer (L2) to avoid overfitting and underfitting; and finally, the Bayesian optimization method is used to adjust the hyper-parameters such as network nodes, regularization parameters, and batch size to obtain better model performance. The experimental results show that the atmospheric temperature prediction error based on LSTM RNN obtained a minimum error compared to using the base models, and these five models obtained are the best models for atmospheric temperature prediction in the corresponding cities. In addition, the feature selection method is applied to the established models, resulting in simplified models with higher prediction accuracy.

Thermomechanical Simulation of Orthogonal Metal Cutting with PFEM and SPH Using a Temperature-Dependent Friction Coefficient: A Comparative Study.

In this work, we apply the Particle Finite Element Method (PFEM) and Smoothed Particle Hydrodynamics (SPH) to simulate the orthogonal cutting chip formation of two workpiece materials, i.e., AISI 1045 steel and Ti6Al4V titanium alloy. A modified Johnson-Cook constitutive model is used to model the plastic behavior of the two workpiece materials. No damage or strain softening is included in the model. The friction between the workpiece and the tool is modeled following Coulomb's law with a temperature-dependent coefficient. The accuracy of PFEM and SPH in predicting thermomechanical loads at various cutting speeds and depths against the experimental data are compared. The results show that both numerical methods can predict the rake face temperature of AISI 1045 with errors less than 34%. For Ti6Al4V, however, the temperature prediction errors are significantly higher than those of the steel alloy. Errors in force prediction were in the range of 10% to 76% for both methods, which compare very well with those reported in the literature. This investigation infers that the Ti6Al4V behavior under machining conditions is difficult to model on the cutting scale irrespective of the choice of numerical method.

Experimental and numerical study on the oxidation and melting behaviors of annular fuel rod under high temperature steam condition

In the safety analysis of nuclear reactor severe accident, the oxidation and melting behaviors of fuel rods under high temperature steam condition is a crucial issue that need to be concerned. As a new type of high-performance light water reactor nuclear fuel, the annular fuel is composed of outer cladding, annular fuel pellet, inner cladding and two coolant channels. At present, the oxidation and melting behaviors research of annular fuel under high temperature steam condition is insufficient. Thus, this paper builds a visual experimental platform for out-of-pile analysis, and carries out the high temperature oxidation experimental research of annular fuel rod based on real cladding material-Zircaloy. The transient temperature characteristic curve and visualization process of annular fuel rod under the conditions of high temperature melting, steam oxidation and re-flooding are obtained. In addition, this paper establishes a seven steps numerical analysis method for oxidation and melting behaviors analysis of annular fuel rod. Based on the three tests results, the accuracy of the model is verified, and the temperature prediction error is within 10%, which lays a foundation for the further calculation of the whole reactor core.

Test and Analysis on Extended Temperature Rise of 110 kV Transformer Based on Distributed Temperature Sensing

Transformer plays a vital role in grids and its internal hot-spot temperature is a crucial monitoring parameter. In this paper, three groups of temperature rise tests with the load rate of 70%, 100%, and 125% were carried out on a 110 kV transformer with built-in distributed sensing fiber. Based on the distributed temperature sensing technology, the temperature distribution of winding was acquired. Accuracy of IEC, Susa, and Swift thermal models was compared. To eliminate the non-negligible error in cooling process, a modified-Susa model based on driving factor and steady-state temperature rise was proposed. Using this, the top oil temperature prediction error during operation and cooling is reduced. In addition, the real-time tracking of hot spots and analysis of overall temperature gradient of windings was implemented. The accuracy of different hot spot calculation models was also compared. Results show that the prediction accuracy of each model is satisfactory during the operation, but the error during the cooling process cannot be ignored. The refined temperature distribution of the winding was also obtained by the high-resolution Brillouin optical time-domain analysis sensing method. Periodic temperature fluctuations and local hot spots inside windings are perceived, which is confirmed to be closely related to winding geometry.

Development and validation of lumped parameter thermal network model on rotational oil spray cooled motor for electric vehicles

The rotational oil-spray-cooling method with motor has recently attracted attention because of its compact design and cooling performance. Rotational oil-spray-cooled motors require high computational resources and manufacturing costs; therefore, a precise simulation model is required. In this study, an asymmetric lumped parameter thermal network (LPTN) model of a rotational oil-spray-cooling motor is developed. The heat loss is calculated using correlation equations and electro-magnetic analysis, and the internal temperature distribution of the motor is predicted using conjugate heat transfer and multiphase flow computational fluid dynamics (CFD) analysis. The temperature of the coil inside the motor is measured using experiments. The developed LPTN model determined that the temperature prediction errors of coil parts were 0.15% and 4.42% at the nominal and maximum speeds, respectively.

Model-Based Approach to Long Term Prediction of Battery Surface Temperature

Temperature has a substantial impact on the safety, efficiency, and reliability of a lithium-ion (Li-ion) battery. The objective of a battery thermal management system (BTMS) is to maintain the battery in preferred temperature range. State-of-the-art BTMS is designed to react to the measured temperature at the battery surface, whereas research suggests that the performance of BTMS will likely improve when used in conjunction with predicted temperature. This article focuses on surface temperature prediction of Li-ion batteries based on an equivalent circuit model-based approach. Particularly, this article presents theoretically sound approaches to model heat generation in the core of the battery and its propagation to its surface. Another novel aspect of the present article is that the heat generation and propagation model identification is tested in various practical scenarios, where the measurement noise is significant. Experimental results using data collected from high precision battery cycler and temperature sensor show an open-loop mean absolute error of 1.678 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$^{\circ }$</tex-math></inline-formula> C during a high-current discharge that lasted approximately 800 s. Using the theoretical models presented in this article, the performance battery temperature prediction error at low SNR conditions can be understood.