Gaussian Process Regression Model Research Articles

Low-frequency chatter suppression for robotic milling using a novel MRF absorber

Robotic milling has a unique advantage for large and complex parts. However, it is extremely prone to low-frequency chatter due to the robot structure mode. In robotic milling, low-frequency chatter has a huge impact on machining quality and efficiency. In this paper, a novel MRF (Magnetorheological fluid) absorber is used to suppress low-frequency chatter during robotic milling based on the proposed low-frequency chatter suppression strategy. First, the GPR (Gaussian process regression) model is used to predict the impulse response function based on a few measured impulse response functions. Next, a cutting force model is developed considering the tool runout and the robot structure mode. The parameters and coefficients of the cutting force model are optimized by the measured cutting forces. Then, the low-frequency chatter frequency is predicted considering the modal coupling effect based on the predicted impulse response function and the simulated cutting force. Finally, an MRF absorber is designed based on the chatter frequency range. The controlled storage modulus of the MRF in the pre-yield region enables a wider frequency shift characteristic of the designed MRF absorber. The MRF absorber controls the input current based on the predicted low-frequency chatter frequency to achieve chatter suppression during robotic milling. Robotic milling experiments verified that the MRF absorber can effectively suppress low-frequency chatter. The experimental results show that the Energy Ratio Chatter Indicator of the industrial robot with MRF absorber is less than 0.1 under different milling conditions.

Monitoring cropland LAI using Gaussian Process Regression and sentinel – 2 surface reflectance data in Google Earth Engine

ABSTRACT Assessing Leaf Area Index (LAI) is one of the key indicators to study and understand cropland’s productivity. To this end, Sentinel-2 (S2) ideally serves to monitor LAI given its high spatial and temporal resolution. Thanks to the cloud computing prowess of the revolutionary Google Earth Engine (GEE), assessing the temporal changes in LAI over a large area can be achieved quickly, making crop health monitoring in near real-time plausible. To achieve this, this work has integrated the Gaussian Process Regression (GPR) algorithm and the PROSAIL model into GEE to map LAI values for the crops being cultivated accurately. The Automated Radiative Transfer Mode Operator (ARTMO) software generated PROSAIL simulation spectra. The GPR model was trained using the PROSAIL simulation spectra and optimized through the Euclidean distance-Based Diversity (EBD) Active Learning (AL) method. The final GPR model was validated against in-situ data collected on 23rd February, 2023 from the research farms of the Indian Agricultural Research Institute (IARI), New Delhi, India. Through ARTMO, the validated GPR model produced an estimated LAI map over croplands managed by IARI, with an accuracy of 0.688 R2 value, and these LAI values and a precision of 0.96 R2 were observed between the LAI values observed in the ARTMO LAI map and the GEE-generated map. This model was then run into GEE to study the temporal changes in LAI values for Mustard, Chickpea, and Wheat, three prominent crops for the Rabi season. These temporal LAI patterns can assist in understanding the cropland’s phenological changes. Thanks to GEE, crop phenology monitoring can be efficiently realized using temporal LAI estimates.

A novel methodology for gamma-ray spectra dataset procurement over varying standoff distances and source activities

The adoption of machine learning approaches for gamma-ray spectroscopy has received considerable attention in the literature. Many studies have investigated the deployment of various algorithm architectures to a specific task. However, little attention has been afforded to the development of the datasets leveraged to train the models. Such training datasets typically span a set of environmental or detector parameters to encompass a problem space of interest to a user. Variations in these measurement parameters will also induce fluctuations in the detector response, including expected pile-up and ground scatter effects. Fundamental to this work is the understanding that 1) the underlying spectral shape varies as the measurement parameters change and 2) the statistical uncertainties associated with two spectra impact their level of similarity. While previous studies attribute some arbitrary discretization to the measurement parameters for the generation of their synthetic training data, this work introduces a principled methodology for efficient spectral-based discretization of a problem space. A signal-to-noise ratio (SNR) respective spectral comparison measure and a Gaussian Process Regression (GPR) model are used to predict the spectral similarity across a range of measurement parameters. This innovative approach effectively showcased its capability by dividing a problem space, ranging from 5 cm to 100 cm standoff distances and 5 μCi–100 μCi of 137Cs, into three unique combinations of measurement parameters. The findings from this work will aid in creating more robust datasets, which incorporate many possible measurement scenarios, reduce the number of required experimental test set measurements, and possibly enable experimental training data collection for gamma-ray spectroscopy.

Efficient predictive modeling of resilient modulus in stabilized clayey soil using automated machine learning

Pavement subgrade design relies on the resilient modulus (Mr) to analyze structural response to vehicle-like loading. Adding stabilizers to the soil subgrade makes estimating Mr difficult and resource-intensive. This study uses an automated machine learning (ML) strategy to predict the Mr of stabilized clayey soil using recycled plastic waste. The proposed method automates model selection and hyperparameter tuning, making it a feasible alternative to tedious ML modeling and costly laboratory testing. From extensive laboratory investigation involving 3285 experimental data points, the automated ML model using Bayesian optimization evaluates ensembles, support vector machine (SVM), neural network (NET), decision trees, (TREE), and Gaussian process (GP) regression models, identifying the best model based on cross-validation mean squared error (MSE). Bayesian optimization explores hyperparameter spaces to find optimal configurations, enhancing the accuracy, scalability, and reliability of the prediction model. The optimization process yielded the best results for the ensemble least square boost (LSBoost) model with a cross-validation mean squared error (MSE) value of 6.723×10−29. The optimized ML model’s performance is measured using R2 and adjusted R2. The LSboost model’s R2 and adjusted R2 values of 0.9999 suggested overfitting, prompting further investigations using performance metrics like root mean squared error (RMSE), mean absolute error (MSE), and probability density function (PDF) for normalized absolute error (NAE) for training and testing datasets for the predictive ML model. The small RMSE and MAE (0.0049 and 0.0005) values and symmetrical NAE distribution of the proposed ML model demonstrate its high accuracy and generalization capabilities. The proposed model was subsequently tested on the new, unseen data and achieved predictions with an error rate of 0.24 %. This confirms the proposed ML model’s superiority over conventional deformation models, making it ideal for reliable geotechnical engineering applications.

Evaluating enhanced predictive modeling of foam concrete compressive strength using artificial intelligence algorithms

Artificial intelligence algorithms have recently demonstrated their efficacy in accurately predicting concrete properties by optimizing mixing proportions and overcoming design limitations. In this regard, foam concrete (FC) production presents a unique challenge, necessitating extensive experimental trials to attain specific properties such as compressive strength (CS). In this context, linear regression (LR), support vector regression (SVR), a multilayer-perceptron artificial neural network (MLP-ANN), and Gaussian process regression (GPR) algorithms, were used to predict the CS of FC. 261 experimental results were utilized, incorporating input variables such as density, water-to-cement ratio, and fine aggregate-to-cement ratio. During the training phase, 75% of the experimental dataset was utilized. The experimental data is then validated using metrics such as coefficient of determination (R2), root mean square error, and root mean error. In comparison, the GPR algorithm reveals high-accuracy towards the estimation of CS, as proved by its high R2-value, which equals 0.98, while the R2 for ANN, SVR, and LR are 0.97, 0.90, and 0.89, respectively. Additionally, parametric and sensitivity analyses were used to assess the performance of the GPR and LR algorithms. Results revealed that density exerted the most significant influence on CS, with the GPR model showing a pronounced negative impact of fine aggregate-to-cement ratio on CS, particularly in low-density FC, contrasting with the LR model. This study confirmed that the GPR algorithm provided reliable accuracy in predicting the CS of FC. Therefore, it is recommended to utilize the prediction algorithms within the range of input variables employed in this investigation for optimal results.

Microwave-based and convective drying of cabbage (Brassica oleracea L. var capitata L.): Computational intelligence modeling, thermophysical properties, quality and mid-infrared spectrometry

This study assessed and compared the impact of hot air oven drying (HAD) at 50, 60 and 70 °C and microwave drying (MWD) at 195, 307 and 521 W on the kinetics of thermophysical properties and quality of dried cabbage. The thermophysical properties were computed using established model equations. The proximate composition, bioactive compounds, antioxidant capacity (AOC), color and functional groups were analyzed using Association of official analytical chemists (AOAC) test protocol, High performance liquid chromatography (HPLC), Ultraviolet-visible (UV–Vis) spectrometry and Attenuated total reflectance coupled to Fourier transformed infrared spectroscopy (ATR-FTIR), respectively. The moisture ratio was modeled using probabilistic computational intelligence modeling approaches. The results revealed that only density, thermal diffusivity and thermal effusivity appeared to be kinetically controlled by moisture content. Stepwise linear regression (SLR) model and Gaussian process regression (GPR) model were more accurate in simulating the moisture ratio. Increasing HAD temperature preserved all of the proximate compositions except carbohydrates. Similarly, increasing MWD power preserved all the proximate compositions except crude fiber. Increasing HAD temperature from 50 to 60 °C and 60–70 °C decreased the total flavonoid content (TFC) and total phenolic content (TPC) by 21.31 % and 13.60 %, respectively but increased the AOC from 27.2 to 39.6 %. Increasing MWD power levels from 195 to 307 W and 307–521 W decreased TFC by 13.2 % and 0.7 %, respectively but increased the TPC by 9.72 % and 13.14 %, thus AOC increased from 35.2 to 45.7 %. MWD presented higher AOC relative to HAD. Drying at 50 °C and 307 W gave the highest whitening index in HAD and MWD. ATR-FTIR analysis revealed the formation of new compounds by thermal effect. SLR and GPR were more accurate in modeling the drying kinetics. HAD and MWD affected the quality parameters of cabbage differently, thus the choice of the dryer type will depend on the parameters of interest.

Exploring robust microwave absorption properties of functional polyvinyl butyral (PVB) via facile cross-linking and data-driven material discovery

This work introduces a chemically facile approach for synthesizing cross-linked polyvinyl butyral (XPVB) at room temperature, utilizing divinyl sulfone (DVS) as the cross-linking agent. Leveraging data-driven discovery methods, we optimize the DVS concentration to enhance the material's reflection loss (RL) performance, establishing XPVB as a formidable contender in microwave absorption without the need for conventional filler materials. Gaussian Process Regression (GPR) was used in this analysis, the material's development process was streamlined to achieve a high-fidelity predictive model that closely aligns with the observed experimental data. The synergy between our GPR model and the empirical evidence is particularly noteworthy, with the GPR predictions exhibiting exceptional agreement across the investigated frequency range, especially in the Ku-band, where high RL values are crucial. These findings indicate that XPVB exhibits exceptional dielectric properties that translate into an unprecedented absorption bandwidth encompassing the entire Ku-band, paired with strong RL, thereby achieving a significant attenuation of electromagnetic waves even at minimal material thicknesses. The optimized XPVB sample, without any filler addition, delivers an exceptional RL performance. Notably, a 2.4 mm-thick XPVB sample demonstrated an extraordinary RL value of -44 dB with RL≤-10 dB absorption Ku-bandwidth (12.4–18 GHz), showcasing the material's potential as a standalone absorber. This synthesis methodology, combined with the analytical approach, lays the groundwork for further research and development in microwave absorbers and has the potential to significantly impact the fields of defense, telecommunications and self-powered devices by providing a new standard for material performance.

Adaptive machine learning surrogate based multiobjective optimization for scavenging residual saltwater behind subsurface dams

High-fidelity physically based groundwater flow and solute transport models have been limited for seawater intrusion remediation design because of computationally intensive evolutionary algorithms. Data-driven machine learning approaches are promising to substitute expensive-cost groundwater numerical models within optimization due to computing efficiency. However, machine learning surrogates may accumulate error of forecasting and thereby result in infeasible optimal solutions. To achieve desired fidelity level, this study proposes a novel adaptive machine learning surrogate based multiobjective optimization method for coastal aquifer desalination. An adaptive modeling algorithm is newly introduced to iteratively retrain poorly-performed machine learning models and enhance predicting accuracy. The method is demonstrated to seek optimal extraction and injection strategies for scavenging residual saltwater trapped in an upstream aquifer behind a subsurface dam. Two conflicting objectives of minimizing the total extraction-injection rate and maximizing saltwater removal effectiveness are considered. Three machine learning models including artificial neural network (ANN), Gaussian process (GP) and response surface regression model (RSR) are developed to replace a high-fidelity seawater intrusion model for predicting chloride concentration and salinity mass. Non-dominated Sorting Genetic Algorithm II (NSGA-II) is employed to derive Pareto fronts. Pareto optimal solutions obtained from machine learning models are compared against those from the seawater intrusion model. Results indicate that the developed machine learning models do not only have strong predicting capability, but also maintain good quality of Pareto optimal solutions, while achieving substantial computational saving up to 95%. Especially, the adaptively retrained RSR model inhibits error accumulation of forecasting and accelerates correct convergence to the true Pareto front. The proposed method is found superior performance in accurate convergence, widely-spread diversity as well as high computing efficiency than conventional methods.

Information consistency of dependent convolved Gaussian processes regression

This paper presents a non-parametric Bayesian approach for modeling multiple response variables using a Gaussian process regression (GPR) model. The response functions are modeled using a dependent Gaussian process (GP) prior, and the estimation, prediction, and inference issues are discussed within this framework. To establish the information consistency of the dependent GPs prediction strategy, a stretching-restriction method is proposed. The covariance structure is constructed using convolved Gaussian processes (CGPs) to illustrate the results. Simulations and real data analyses show that the proposed dependent GPR yields reasonably good prediction accuracy.

Gaussian Process Regression Based Silver Price Forecasts

A significant number of market participants have placed a high level of importance on price estimates for the primary metal commodities for a considerable amount of time. To tackle the problem, we investigate the daily reported price of silver in our study. The sample that is being analyzed spans a period of 13 years, starting on April 20, 2011, and ending on April 19, 2024. The price series which is being investigated has important implications for the commercial world. Specifically, when it comes to this unique circumstance, Gaussian process regression models are developed utilizing cross-validation strategies and Bayesian optimization procedures. The forecasting of prices is therefore accomplished via the use of the methods that are developed as a result of the situation. For the out-of-sample evaluation period that extends from October 5, 2021 to April 19, 2024, our empirical forecasting approach yields price estimates deemed reasonably accurate. The relative root mean square error reached for the silver price is 0.2257%, with the corresponding root mean square error of 0.0515, mean absolute error of 0.0389, and correlation coefficient of 99.967%. Due to the availability of models that forecast prices, investors and governments are supplied with the information they need to make educated judgments on the silver market by providing them with the knowledge they require. The framework of the Gaussian process regression with Bayesian optimizations demonstrates its good potential for modeling and forecasting sophisticated commodity price series for market participants.

Semiconductors for enhanced solar photovoltaic-thermoelectric 4E performance optimization: Multi-objective genetic algorithm and machine learning approach

In this study, a groundbreaking exploration of a concentrated photovoltaic-thermoelectric (CPV-TE) module employing an unprecedented selection of six thermoelectric materials across diverse temperature ranges is presented. This work innovatively employs a multi-objective optimization framework that combines genetic algorithms and goal attainment methods, aiming to optimize energy, exergy, environmental, and economic (4 E) performance. Notably, this study is the first to construct and evaluate five regression-based machine learning models with an emphasis on minimal root mean squared error and mean absolute error for rapid CPV-TE performance predictions, essential for real-world applications. Our analysis unveils that among the tested materials, the CPV-TE module incorporating lead telluride (PbTe) achieves the highest 4 E performance, demonstrating a peak exergy efficiency of 14.2 %, and annual energy savings and carbon reduction of 13.5 kWh and 6.4 kg, respectively. Furthermore, the Gaussian Process Regression model is identified as the most effective among the machine learning models for forecasting performance across the materials. This study significantly advances the field by providing novel insights into thermoelectric material selection and optimization for CPV-TE modules, and establishing pioneering forecasting tools that catalyze the efficient deployment of these systems.

Monitoring cropland LAI using Gaussian Process Regression and sentinel – 2 surface reflectance data in Google Earth Engine

ABSTRACT Assessing Leaf Area Index (LAI) is one of the key indicators to study and understand cropland’s productivity. To this end, Sentinel-2 (S2) ideally serves to monitor LAI given its high spatial and temporal resolution. Thanks to the cloud computing prowess of the revolutionary Google Earth Engine (GEE), assessing the temporal changes in LAI over a large area can be achieved quickly, making crop health monitoring in near real-time plausible. To achieve this, this work has integrated the Gaussian Process Regression (GPR) algorithm and the PROSAIL model into GEE to map LAI values for the crops being cultivated accurately. The Automated Radiative Transfer Mode Operator (ARTMO) software generated PROSAIL simulation spectra. The GPR model was trained using the PROSAIL simulation spectra and optimized through the Euclidean distance-Based Diversity (EBD) Active Learning (AL) method. The final GPR model was validated against in-situ data collected on 23rd February, 2023 from the research farms of the Indian Agricultural Research Institute (IARI), New Delhi, India. Through ARTMO, the validated GPR model produced an estimated LAI map over croplands managed by IARI, with an accuracy of 0.688 R2 value, and these LAI values and a precision of 0.96 R2 were observed between the LAI values observed in the ARTMO LAI map and the GEE-generated map. This model was then run into GEE to study the temporal changes in LAI values for Mustard, Chickpea, and Wheat, three prominent crops for the Rabi season. These temporal LAI patterns can assist in understanding the cropland’s phenological changes. Thanks to GEE, crop phenology monitoring can be efficiently realized using temporal LAI estimates.

Improving the energy efficiency and riding comfort of high-speed trains across slopes by the optimized suspension control

As the link between cities, the high-speed train (HST) not only effectively enhances national and regional accessibility, but also induces much energy consumption. On the other hand, passengers have a higher requirement for riding comfort of HST than before. Considering the impacts of slope gradient on energy consumption and vertical carbody acceleration, this paper optimizes the scale factor of the damper controller depending on the running condition to juggle the energy efficiency and riding comfort when HST is across slopes. Firstly, the multibody dynamics simulation model of HST is established to evaluate the energy consumed by motion resistances and carbody vibrations. Then the Skyhook control strategy is applied to the secondary vertical damper of the vehicle model. The effects of vehicle speed, slope gradient, and scale factor of the Skyhook controller on energy saving and vertical carbody acceleration of HST are investigated. Based on the simulation results, the Gaussian process regression model is established to predict the vehicle performance and describe the bi-objective optimization problem. It is demonstrated that the optimized Skyhook control system can at most save energy consumed by motion resistances by 8.52% or reduce vertical carbody acceleration by 37.14% without sacrificing the other target. Finally, a comprehensive feasibility and economic analysis of the proposed control system is carried out to promote its practical implementation.

Comprehensive evaluation and systematic comparison of Gaussian process (GP) modelling applications in peptide quantitative structure-activity relationship

Peptide quantitative structure-activity relationship (pQSAR) is a specific extension of traditional QSARs from small-molecule drugs to bioactive peptides. Since peptides are linear biopolymers that are essentially different to small-molecule compounds in terms of their structural features such as ordering sequence, large size and intrinsic flexibility, the pQSAR methodology (including structural characterization and regression modelling) should be further exploited relative to traditional QSARs. Gaussian process (GP) serves as a pioneering Bayesian-based machine learning (ML) solution for tackling linear/nonlinear-hybrid regression issues in intricate domains. However, the applications of GP regression in QSAR and, particularly, the pQSAR still remain largely unexplored to date. In this work, we launched a comprehensive pQSAR study with GP regression modelling, aiming to the deep evaluation of GP performance based on different characterizations and also the systematic comparison of GP with other routine MLs. Here, we culled two distinct classes of peptide datasets, which separately comprise 12 panels of sophisticated benchmarks and 46 panels of extended samples, totally containing 8804 peptide samples and systematically resulting in 522 regression models. Our study indicated that the GP can generally provide an effective solution for many pQSAR problems with the potential to promote ML regression modelling in this area, which is comparable with or even better than those widely used methods on both the sophisticated benchmarks and extended samples. In addition, GP also has many advantages as compared to traditional MLs, such as hyperparameter self-consistency, overfitting resistance, interpretable output and estimable uncertainty.

A local space transfer learning-based parallel Bayesian optimization with its application

The optimization of process parameters in polyolefin production can bring significant economic benefits to the factory. However, due to small data sets, high costs associated with parameter verification cycles, and difficulty in establishing an optimization model, the optimization process is often restricted. To address this issue, we propose using a transfer learning Bayesian optimization strategy to improve the efficiency of parameter optimization while minimizing resource consumption. Specifically, we leverage Gaussian process (GP) regression models to establish an integrated model that incorporates both source and target grade production task data. We then measure the similarity weights of each model by comparing their predicted trends, and utilize these weights to accelerate the solution of optimal process parameters for producing target polyolefin grades. In order to enhance the accuracy of our approach, we acknowledge that measuring similarity in a global search space may not effectively capture local similarity characteristics. Therefore, we propose a novel method for transfer learning optimization that operates within a local space (LSTL-PBO). This method employs partial data acquired through random sampling from the target task data and utilizes Bayesian optimization techniques for model establishment. By focusing on a local search space, we aim to better discern and leverage the inherent similarities between source tasks and the target task. Additionally, we incorporate a parallel concept into our method to address multiple local search spaces simultaneously. By doing so, we can explore different regions of the parameter space in parallel, thereby increasing the chances of finding optimal process parameters. This localized approach allows us to improve the precision and effectiveness of our optimization process. The performance of our method is validated through experiments on benchmark problems, and we discuss the sensitivity of its hyperparameters. The results show that our proposed method can significantly improve the efficiency of process parameter optimization, reduce the dependence on source tasks, and enhance the method's robustness. This has great potential for optimizing processes in industrial environments.

A data-driven computational optimization framework for designing thin-walled lenticular deployable composite boom with optimal load-bearing and folding capabilities

The thin-walled lenticular deployable composite boom (LDCB) is promising for aerospace engineering applications due to its lightweight and compact nature, but its mechanical behaviors are the main challenges limiting its practical application. Here, the axial compression and folding behaviors of LDCB were numerically simulated and the simulation results were verified according to the experimental data in the literature. Then, a new double chains quantum genetic algorithm (DCQGA)-gaussian process regression (GPR) model and data-driven computational optimization framework were proposed and the new model was trained using a database with 400 simulation results, the superiority of which was demonstrated by prediction accuracy evaluation. Additionally, benchmarking five state-of-the-art algorithms found that the coupling technology of DCQGA-GPR-non-dominated sorting genetic algorithm III (NSGA-III) is an excellent optimization strategy to obtain a well-designed structure of LDCB that maximizes the effectiveness of the material while satisfying the lightweight. The optimized LDCB design exhibits a significant performance improvement, with a 26.3 % reduction in the folding moment, a 34.2 % reduction in the maximum Tsai-Hill failure index, a 43.6 % increase in the critical load, and an 11 % reduction in linear density of mass compared to the initial design.

Prediction and reliability analysis of shear strength of RC deep beams

This study explores machine learning (ML) capabilities for predicting the shear strength of reinforced concrete deep beams (RCDBs). For this purpose, eight typical machine-learning models, i.e., symbolic regression (SR), XGBoost (XGB), CatBoost (CATB), random forest (RF), LightGBM, support vector regression (SVR), artificial neural networks (ANN), and Gaussian process regression (GPR) models, are selected and compared based on a database of 840 samples with 14 input features. The hyperparameter tuning of the introduced ML models is performed using the Bayesian optimization (BO) technique. The comparison results show that the CatBoost model is the most reliable and accurate ML model (R2 = 0.997 and 0.947 in the training and testing sets, respectively). In addition, simple and practical design expressions for RCDBs have been proposed based on the SR model with a physical meaning and acceptable accuracy (an average prediction-to-test ratio of 0.935 and a standard deviation of 0.198). Meanwhile, the shear strength predicted by ML models was then compared with classical mechanics-driven shear models, including two prominent practice codes (i.e., ACI318, EC2) and two previous mechanical models, which indicated that the ML approach is highly reliable and accurate over conventional methods. In addition, a reliability-based design was conducted on two ML models, and their reliability results were compared with those of two code standards. The findings revealed that the ML models demonstrate higher reliability compared to code standards.

Optimization of ECAP parameters of ZX30 alloy using feature engineering assisted machine learning and response surface approaches

The precise control of equal channel angular pressing (ECAP) deformation parameters is crucial for producing ultrafine grains in structural ZX30-Mg alloy biodegradable components. To address this challenge, this study aims to conduct a comprehensive analysis of the effects of various ECAP parameters on the microstructure, corrosion, and mechanical properties of ZX30 alloy. ZX30 was utilized through a maximum of 4 passes along routes A, Bc, and C, using die angles of 90ᵒ and 120ᵒ, by utilizing machine learning (ML), artificial neural networks (ANNW), response surface methodology (RSM), and simulated annealing (SA) techniques. Experimental investigations revealed the significance of the number of passes in reducing grain size by up to 1.9 µm, consequently enhancing both mechanical and corrosion resistance properties, notably in route Bc. The corrosion rate was enhanced by 97.6 % while the corrosion resistance was increased by 254 % compared to the as-annealed condition. RSM and simulated annealing optimization results closely aligned with experimental findings. The evaluation of each extrusion parameter revealed that the correlation coefficient, as determined by the optimized Gaussian Process Regression model, ranged from 94 % to 97 % for all evaluated responses, and near to unity via ANNW. The overall distribution of mechanical and corrosion measures showed a significant correlation with the number of passes, with route A, Bc, and die angle following closely behind. Interestingly, route Bc had the main impact on the yield strength, followed by the number of passes and die angle. Thus, the combination of statistically and mathematically selected techniques effectively promotes the attainment of optimal properties. Furthermore, the Simulated Annealing (SA) algorithm demonstrated its capability to provide optimal solutions for ECAP deformation parameters. This study presents a promising avenue for improving the accuracy of ECAP-induced structural optimization in ZX30 alloy.

Multi-cell sensorless internal temperature estimation based on electrochemical impedance spectroscopy with Gaussian process regression for lithium-ion batteries safety

Recently, fire incidents due to electrical vehicle (EV) lithium-ion battery (LIB)-based energy storages have been increasingly occurring. Preemptive diagnosis of the thermal runaway is essential for preventing such fires. However, precise and accurate temperature monitoring at the single cell level is required for all the batteries present in the EV pack. Electrochemical impedance spectroscopy (EIS) closely correlates with the battery's internal temperature (Tint) for sensorless temperature estimation, but is limited for multi-cell use because of the effect of the cell-to-cell differences on the EIS impedance values, causing inaccuracies in the use case of predefined models. This study developed a framework combining the EIS method with Gaussian process regression (GPR) machine learning for multiple LIBs cell temperature estimation. The intrinsic relationship linking the temperature, state of charge, state of health and temperature sensitivity with the impedance properties of the batteries at different frequencies was investigated. The results showed that from 51 extracted frequencies only single-point frequency measurements at 10 Hz were necessary to develop the temperature estimation model. The data at this frequency was subsequently utilized to train the GPR model. Five kernel functions were studied, trained, and validated experimentally. The root means square error (RMSE) for the most accurate kernel was 0.2 °C at the validation step, and the test RMSE for the estimation of a new cell was 0.79 °C, confirming the high accuracy of the model.

Impact of Solder Voids on IGBT Thermal Behavior: A Multi-Methodological Approach

This study investigates the thermal behavior of Insulated Gate Bipolar Transistors (IGBTs) with a focus on the influence of solder voids within the device. Utilizing a combination of Finite Element Method (FEM) simulations, X-ray imaging, and SEM-EDX analysis, we accurately modeled the internal structure of IGBTs to assess the impact of void characteristics on thermal resistance. The findings reveal that the presence and characteristics of solder voids—particularly their size, number, and distribution—significantly affect the thermal resistance of IGBT devices. Experimental measurements validate the FEM model’s accuracy, confirming that voids disrupt the heat flow path, which can lead to increased thermal resistance and potential device failure. Five regression models, including Gaussian process regression (GPR) and neural networks, were employed to predict the thermal resistance based on void characteristics, with the GPR models demonstrating superior performance. The optimal GPR RQ model consistently provided accurate predictions with an RMSE of 0.0050 and R2 of 0.9728. Using the void percentage as the only input parameter for the regression models significantly impacted the prediction accuracy, showing the importance of the void extraction method. This study shows the necessity of minimizing solder voids and offers a robust methodological framework for a better prediction of the reliability of IGBTs.