Set Of Experimental Data Research Articles

Identificación del acento en hablantes de español mediante el análisis de atributos MFCC y aprendizaje supervisado

Speaker recognition has multiple real-life applications. The purpose of this study is to determine the feasibility of classifying samples of human speech, specifically Spanish speakers, based on their distinctive accents. In this work, Mel-Frequency Cepstral Coefficients (MFCC) combined with machine learning techniques were used to identify the nationality of Spanish-speaking individuals through voice recordings obtained from the Crowdsourcing Latin American Spanish for Low-Resource Text-to-Speech corpus. Data preprocessing was performed by extracting 50 MFCC from each recording, which formed the dataset for experimentation. Experiments were conducted with different subsets, and the best results were obtained with individuals from four Latin American countries, including both males and females. Neural networks were employed for the classification stage, achieving an accuracy of 99.84%.

Neural network–based transfer learning to improve stiffness modeling of industrial robots with small experimental data sets

AbstractStiffness modeling is an essential subject for the composition of robot control. Accurate stiffness modeling is helpful for improving the control accuracy of industrial robots, particularly under dynamic load circumstances. The classic virtual joint modeling (VJM) method is challenging in predicting the deformation of the end-effector throughout the full workspace due to the nonlinear deformation of the robot joint and its serial articulated structure. This paper proposes a full-space stiffness modeling method for robots based on the integration of a multi-layer perceptual (MLP) model and VJM. To provide enough training data for the MLP model, VJM is used to build a stiffness model with a small set of experimental data to generate 106,400 training data. A model-based transfer learning approach is proposed to improve the model’s accuracy and generalization regarding the difference between generated training data and actual experimental data. The VJM stiffness model is compared with the MLP stiffness model and the existing CNN-based transfer learning model based on the same experimental data. Considering the deformation prediction in the three directions in Cartesian space, the mean absolute error, standard deviation, and maximum error of the MLP model are decreased by at least 24.90%, 14.20%, and 8.50%, respectively, than the VJM. These prediction results demonstrate that the proposed modeling technique can significantly increase the accuracy of robot stiffness modeling, which is essential for position compensation in precise motion control of robots under dynamic load.

Parameter Identification in Triple-Diode Photovoltaic Modules Using Hybrid Optimization Algorithms

Identifying the parameters of a triple-diode electrical circuit structure in PV modules is a critical issue, and it has been regarded as an important research area. Accordingly, in this study, a differential evolution algorithm (DEA) is hybridized with an electromagnetism-like algorithm (EMA) in the mutation stage to enhance the reliability and efficiency of the DEA. A new formula is presented to adapt the control parameters (mutation factor and crossover rate) of the DEA. Seven different experimental data sets are used to improve the performance of the proposed differential evolution with an integrated mutation per iteration algorithm (DEIMA). The results of the proposed PV modeling method are evaluated with other state-of-the-art approaches. According to different statistical criteria, the DEIMA demonstrates superiority in terms of root mean square error and main bias error by at least 5.4% and 10%, respectively, as compared to other methods. Furthermore, the DEIMA has an average execution time of 27.69 s, which is less than that of the other methods.

Transferable and Interpretable Prediction of Site-Specific Dehydrogenation Reaction Rate Constants with NMR Spectra.

Accurate and efficient determination of site-specific reaction rate constants over a wide temperature range remains challenging, both experimentally and theoretically. Taking the dehydrogenation reaction as an example, our study addresses this issue by an innovative combination of machine learning techniques and cost-effective NMR spectra. Through descriptor screening, we identified a minimal set of NMR chemical shifts that can effectively determine reaction rate constants. The constructed model performs exceptionally well on theoretical data sets and demonstrates impressive generalization capabilities, extending from small molecules to larger ones. Furthermore, this model shows outstanding performance when applied to limited experimental data sets, highlighting its robust applicability and transferability. Utilizing the Sure Independence Screening and Sparsifying Operator (SISSO) algorithm, we also present an interpretable rate constant-temperature-NMR (k-T-NMR) relationship with a mathematical formula. This study reveals the great potential of combining machine learning with easily accessible spectroscopic descriptors in the study of reaction kinetics, enabling the rapid determination of reaction rate constants and promoting our understanding of reactivity.

Contact fatigue life prediction of PEEK gears based on CTAB-GAN data augmentation

The contact fatigue life of PEEK gears is a critical fundamental data for the design of load-carrying capacity, which is significantly influenced by many factors such as material type, operating temperature, and loading level. Temperature effects must be considered during contact fatigue life prediction of PEEK gears. This article presents a predictive model for contact fatigue life of PEEK gears, utilizing the CTAB-GAN data augmentation. The CTAB-GAN data augmentation model expands the experimental dataset of contact fatigue lives of oil-lubricated PEEK gears. Then, a predictive formula was derived to estimate contact fatigue life of oil-lubricated PEEK gears considering the temperature dependent effect. Results revealed that the predicted fatigue lives of PEEK gears under different loading and temperature conditions based on the enhanced dataset are within a two-fold scatter band. The mean square error of predicted lives reduces by 30% compared to the life formula based on the original experimental data set. Furthermore a quantitative analysis was conducted on the influence of operating temperature and stress on the fatigue life of PEEK gears.

Carbon Nanoparticle Oxidation by NO2and O2: Chemical Kinetics andReaction Pathways.

Carbon nanoparticle interactions with gases are central to many environmental and technical processes, but the underlying reaction kinetics and mechanisms are not well understood. Here, we investigate the oxidation and gasification of carbon nanoparticles by NO2and O2under combustion exhaust conditions. We build on a comprehensive experimental data set and use a kinetic multilayer model (KM-GAP-CARBON) to trace the uptake and release of gas molecules alongside the temporal evolution of particle size and surface composition. The experimental results are captured by a model mechanism that involves different types of carbon atoms (edge/plane-like) and the formation of a reactive oxygen intermediate (activated CO complex) as the rate-limiting step. A transition between distinct chemical regimes driven by NO2at lower temperatures and O2at higher temperatures is reflected by an increase in the observable activation energy from ~60 kJ/mol to ~130 kJ/mol. We derive energy profiles for three alternative reaction pathways that involve uni- or bimolecular decomposition of reactive oxygen intermediates.

Carbon Nanoparticle Oxidation by NO2 and O2: Chemical Kinetics and Reaction Pathways

Carbon nanoparticle interactions with gases are central to many environmental and technical processes, but the underlying reaction kinetics and mechanisms are not well understood. Here, we investigate the oxidation and gasification of carbon nanoparticles by NO2 and O2 under combustion exhaust conditions. We build on a comprehensive experimental data set and use a kinetic multilayer model (KM‐GAP‐CARBON) to trace the uptake and release of gas molecules alongside the temporal evolution of particle size and surface composition. The experimental results are captured by a model mechanism that involves different types of carbon atoms (edge/plane‐like) and the formation of a reactive oxygen intermediate (activated CO complex) as the rate‐limiting step. A transition between distinct chemical regimes driven by NO2 at lower temperatures and O2 at higher temperatures is reflected by an increase in the observable activation energy from ~60 kJ/mol to ~130 kJ/mol. We derive energy profiles for three alternative reaction pathways that involve uni‐ or bimolecular decomposition of reactive oxygen intermediates.

Investigation of Churn and Annular Flow Transition Boundaries in Vertical Upward Gas-Water Two-Phase Flow

ABSTRACT Flow patterns in two-phase flow in vertical upward pipes are crucial for predicting phase distribution, flow transitions, and stability. Existing studies on churn to annular flow transitions primarily focus on flow rates, physical parameters, pressure, and temperature, often overlooking pipe diameter. This study addresses this gap using a multiphase pipe flow experimental setup to examine transitions under varying pipe diameters, gas flow rates, and liquid flow rates. Results show that with a constant liquid flow rate, increasing pipe diameter requires a higher gas flow rate to form annular flow, leading to greater pressure gradients and smoother fluctuations. Conversely, with a constant gas flow rate, larger pipe diameters decrease the liquid flow rate needed for annular flow, resulting in smaller pressure gradients and smoother fluctuations. Smaller pipe diameters make annular flow formation easier, with smaller pressure gradients but larger fluctuations. A novel churn-annular flow transition boundary model was developed, considering factors like gas shear force, surface tension, gravity, viscous force, and pressure gradient. Evaluated with 804 experimental data sets, the new model’s prediction accuracy exceeds 98.19%, outperforming traditional models with 53.92%-86.27% accuracy, enhancing applicability across varying diameters.

Banana slice drying using waste heat from an air conditioner for enhancing performance and sustainability

AbstractAs waste heat dryers (WHD) exhibit conservation of energy with environmental and economic benefits, the direct use of waste hot air rejected from the outdoor unit of a window air conditioning (AC) unit for drying banana slices is demonstrated experimentally in the present research work. For this purpose, a drying chamber was directly coupled to an AC unit (AC‐WHD) and analyzed its performance during banana slice drying by revisiting the concepts of generalized drying curve and drying performance index (DPI). The experimental data set of moisture content ratio obtained with AC‐WHD matched well with the generalized drying curve and the DPI value was found to be 0.999. It was found that the waste air supplied by the AC unit to the drying chamber was hot (≈ 50°C) and 28% dry, which aided the drying process. The thermodynamic performance of AC‐WHD was also evaluated using warm and wet exergy to provide insight into the mechanism of AC‐WHD drying. The exergy based sustainability index was higher while the exergetic destruction coefficient and environmental impact factor were lower for AC‐WHD when compared to literature values of the solar drying method. The findings of this research work indicate the practicality of coupling AC units with dryer units to address sustainable development goals (SDG).

Toward a Circular Lithium Economy with Electrodialysis: Upcycling Spent Battery Leachates with Selective and Bipolar Ion-Exchange Membranes.

Recycling spent lithium-ion batteries offers a sustainable solution to reduce ecological degradation from mining and mitigate raw material shortages and price volatility. This study investigates using electrodialysis with selective and bipolar ion-exchange membranes to establish a circular economy for lithium-ion batteries. An experimental data set of over 1700 ion concentration measurements across five current densities, two solution compositions, and three pH levels supports the techno-economic analysis. Selective electrodialysis (SED) isolates lithium ions from battery leachates, yielding a 99% Li-pure retentate with 68.8% lithium retention, achieving relative ionic fluxes up to 2.41 for Li+ over transition metal cations and a selectivity of 5.64 over monovalent cations. Bipolar membrane electrodialysis (BMED) converts LiCl into high-purity LiOH and HCl, essential for battery remanufacturing and reducing acid consumption via acid recycling. High current densities reduce ion leakage, achieving lithium leakage as low as 0.03%, though hydronium and hydroxide leakage in BMED remains high at 11-20%. Our analysis projects LiOH production costs between USD 1.1 and 3.6 per kilogram, significantly lower than current prices. Optimal SED and BMED conditions are identified, emphasizing the need to control proton transport in BMED and improve cobalt-lithium separation in SED to enhance cost efficiency.

Computational multiphase mixture simulations of a two-phase R-744 ejector geometry in transcritical R-744 heat pump applications

Purpose Two-phase R-744 ejectors are critical components enabling energy recovery in R-744 heat pump and refrigeration systems, but despite their simple geometry, the flow physics involve complex multiphase mixing phenomena that need to be well-quantified for component and overall system improvement. This study aims to report on multiphase mixture simulations for a specific two-phase R-744 ejector with supercritical inlet conditions at the motive inlet side. Design/methodology/approach Four different operating conditions, which have motive inlet pressure range of 90.1 bar–101.1 bar, are selected from an existing experimental data set. A two-phase thermodynamic equilibrium (TPTE) model is used, where the fluid properties are described by a thermodynamic look-up table. Findings The results show that the TPTE model overpredicts mass flow rates at the motive inlet, resulting in a relative error ranging from 15.6% to 21.7%. For the mass flow rate at the suction inlet, the relative errors are found less than 1.5% for three cases, while the last case has an error of 12.4%. The maximum deviation of the mass entrainment ratio is found to be 8.0% between the TPTE model and the experimental data. Ejector efficiency ranges from 25.4% to 28.0%. A higher pressure difference between the ejector outlet and the diverging nozzle exit provides greater pressure lift. Research limitations/implications Based on the results, near future efforts will be to optimize estimation errors while enabling more detailed field analysis of pressure, density, temperature and enthalpy in the computational domain. Originality/value The authors have two main original contributions: 1) the presented thermodynamic look-up table is unique and provides unique computation for the real-scale ejector domain. It was created by the authors and has not been applied before as far as we know. 2) To the best of the authors’ knowledge, this study is the first study that applies the STAR-CCM+ multiphase mixture model for R-744 mixture phenomena in heat pumps and refrigeration systems.

Vehicle object detection and ranging in vehicle images based on deep learning

In order to improve the accuracy of vehicle target detection and the stability of ranging in driving environments, a vehicle target detection and ranging method based on deep learning is proposed. The YOLOX-S algorithm is used as the vehicle target detection framework for improvement: the CBAM attention module is introduced on the basis of the original algorithm to enhance the network feature expression ability, and the confidence loss function is replaced by Focal Loss to reduce the training weight of simple samples and improve the attention of positive samples. The vehicle ranging model is established according to the imaging principle and geometric relationship of the vehicle camera, and the ranging feature point coordinates and camera internal parameters are input to obtain the ranging results. The self-made Tlab dataset and BDD 100K dataset are used to train and evaluate the improved YOLOX-S algorithm, and a static ranging experimental scene is built to verify the vehicle ranging model. The experimental results show that the improved YOLOX-S algorithm has a detection speed of 70.14 frames per second on the experimental data set. Compared with the original algorithm, the precision, recall, F1 value, and mAP are improved respectively 0.86%、1.32%、1.09%、1.54% ; within the measurement range of 50 m in the longitudinal direction and 11.25 m in the lateral direction, the average ranging error is kept within 3.20% . It can be seen that the proposed method has good vehicle ranging accuracy and stability while meeting the real-time requirements of vehicle detection.

Bayesian Probabilistic Inference of Nonparametric Distance Distributions in DEER Spectroscopy.

Double electron-electron resonance (DEER) spectroscopy measures distance distributions between spin labels in proteins, yielding important structural and energetic information about conformational landscapes. Analysis of an experimental DEER signal in terms of a distance distribution is a nontrivial task due to the ill-posed nature of the underlying mathematical inversion problem. This work introduces a Bayesian probabilistic inference approach to analyze DEER data, assuming a nonparametric distance distribution with a Tikhonov smoothness prior. The method uses Markov Chain Monte Carlo sampling with a compositional Gibbs sampler to determine a posterior probability distribution over the entire parameter space, including the distance distribution, given an experimental data set. This posterior contains all of the information available from the data, including a full quantification of the uncertainty about the model parameters. The corresponding uncertainty about the distance distribution is visually captured via an ensemble of posterior predictive distributions. Several examples are presented to illustrate the method. Compared with bootstrapping, it performs faster and provides slightly larger uncertainty intervals.

UAV Vehicle Detection System Based on YOLOv8

Abstract To utilize the drone vehicle detection technology to observe the road traffic conditions in real-time, thereby improving the operational effectiveness and safety of the traffic system, this paper adopts a vehicle target detection algorithm based on YOLOv8. YOLOv8 is the latest iteration developed by Ultralytics, which has further improvements and enhancements compared to YOLOv5. In the backbone network, YOLOv8 employs a C2f structure with richer gradient flow, boosting feature selection capabilities, and enhancing target detection performance and accuracy; In the detection head, YOLOv8 utilizes a Decoupled Head design and uses an improved Anchor-Free method to reduce computational complexity while improving detection efficiency; In terms of loss function, YOLOv8 utilizes the The Task-Aligned Assigner positive sample allocation strategy, and adds the Distribution Focal Loss to improve the model’s generalization ability. This paper divides the vehicles data set into 5 types, trains it using a pre-training model, and optimizes the network parameters through multiple iterations to achieve better detection performance. The experimental results indicate that YOLOv8 reaches the mAP of over 97% on the experimental data set, which is an improvement of approximately 8% compared to YOLOv5, and achieves a good detection performance, and is highly practical.

MinoritySalMix and adaptive semantic weight compensation for long-tailed classification

In real-world datasets, the widespread presence of a long-tailed distribution often leads models to become overly biased towards majority class samples while ignoring minority class samples. We propose a strategy called MASW (MinoritySalMix and adaptive semantic weight) to improve this problem. First, we propose a data augmentation method called MinoritySalMix (minority-saliency-mixing), which uses significance detection techniques to select significant regions from minority class samples as cropping regions and paste them into the same regions of majority class samples to generate brand new samples, thereby amplifying images containing important regions of minority class samples. Second, in order to make the label value information of the newly generated samples more consistent with the image content of the newly generated samples, we propose an adaptive semantic compensation factor. This factor provides more label value compensation for minority samples based on the different cropping areas, thereby making the new label values closer to the content of the newly generated samples. Improve model performance by generating more accurate new label value information. Finally, considering that some current re-sampling strategies generally lack flexibility in handling class sampling weight allocation and frequently require manual adjustment. We designed an adaptive weight function and incorporated it into the re-sampling strategy to achieve better sampling. The experimental results on three long-tailed datasets show that our method can effectively improve the performance of the model and is superior to most advanced long-tailed methods. Furthermore, we extended MinoritySalMix’s strategy to three balanced datasets for experimentation, and the results indicated that our method surpassed several advanced data augmentation techniques.

CO2 solubility and physicochemical properties of 2-amino 2-methyl-1-propanol based solvents for post-combustion CO2 capture. Predictive modeling using machine learning

Global warming, caused by Carbon Dioxide (CO2) emissions from hard-to-abate industries, necessitates CO2 capture as one of the promising alternatives. Post-combustion CO2 capture (PCC) can be retrofitted for industries where chemical absorption using an amine-based solvent is an efficient method for CO2 capture. An essential factor for CO2 absorption in an amine mixture is the study of CO2 solubility and CO2 + Amine physicochemical properties to identify the more efficient solvents. Often experimental data are available in the literature, but precisemodeling is necessary to use such data for further utilization in a process-plant model. In this work, a blended solvent aqueous 2-amine-2methyl-1-Propanal (AMP) with Piperazine (PZ) and 1-(2-aminoethyl) piperazine (AEP) have been identified as a potential candidate for PCC application. The fundamental properties such as the density, viscosity and CO2 solubility in AMP + PZ and AMP + AEP blended solutions have been predicted by employing machine learning to get an exceptionally efficient model. This data-mining technique efficiently calculates solubility, density, and viscosity by using 691, 559, and 559 experimental data sets respectively, from the literature of AMP + PZ. Similarly, 120 data sets of solubility have been collected from the literature for AMP + AEP. Random Forest regression, Artificial Neural Network (ANN), and XGBoost regression algorithms are adapted to predict CO2 solubility in aqueous AMP blended solution. The effectiveness of these models has been confirmed through statistical analysis, leading to the conclusion that XGBoost outperforms the other two approaches. Moreover, an identical approach has been employed to develop a predictive model for the density and viscosity of AMP + PZ solvent. The outcomes imply that the machine learning algorithm performs better with experimental data and may be utilised as an effective simulation tool.

A multi-step calibration strategy for reliable parameter determination of salt rock mechanics constitutive models

The storage of renewable hydrogen in salt caverns requires fast injection and production rates to cope with the imbalance between energy production and consumption. This raises concerns about the mechanical stability of salt caverns under such operational conditions. The use of appropriate constitutive models for salt mechanics is an important step in investigating this issue, therefore many constitutive models with several parameters have been presented in the literature. However, a robust calibration strategy to reliably determine which model and parameter set represents the given rock, based on stress–strain data sets, remains an unsolved challenge. In this paper, for the first time in the community, we present a multi-step strategy to determine a single parameter set based on many deformation data sets for salt rocks. Towards this end, we first develop a comprehensive constitutive model able to capture all relevant nonlinear deformation physics of transient, reverse, and steady-state creep. The determination of the single set of representative material parameters is then achieved by framing the calibration process as an optimization problem, for which the global Particle Swarm Optimization algorithm is employed. To allow for dynamic data integration, a multi-step calibration strategy is developed for a situation where experiments are included one at a time, as they become available. Additionally, due to the existing mild heterogeneity in the experimental rock samples, our optimization strategy is made flexible to allow for this slight heterogeneity. The devised optimization strategy, based on the multi-physics comprehensive constitutive modeling framework, results in a single set of representative material properties of all the deformation data sets. As a rigorous mathematical analysis for the presented method and the lack of relevant experimental data sets, we consider a wide range of synthetic experimental data sets, inspired by the existing sparse relevant data in the literature. The results of our performance analyses show that the proposed calibration strategy is robust. Moreover, the results become increasingly more accurate as more data sets become available.

Adaptive point cloud thinning method based on point cloud saliency factor

Abstract Point cloud thinning is an important data pre-processing method for saving computing resources and improving accuracy of point cloud matching. Aiming at the problems of insufficient feature extraction, poor environmental adaptation, and high computational complexity in existing methods, this study proposes a simplified method based on point cloud salient factors. This method first preprocesses the point cloud data, then calculates the significant factors of the point cloud, and uses the local curvature variance to perform adaptive region division. In the local area, the improved feature farthest point sampling (CIFPS) algorithm is used to classify the point cloud. Perform thinning and finally obtain a simplified point cloud. In order to verify the effectiveness of the method, we conducted a large number of experiments on our own experimental platform and public data sets, and compared it with several related point cloud simplification methods. Experimental results show that the average thinning time of our algorithm on 16-line radar data is 23.67ms, 32-line radar is 50.74ms, and 64-line radar is 87.28ms; the point cloud matching error is relative to that based on farthest point sampling and voxel Sampling and Laplacian sampling were reduced by 44.3%, 46.3%, and 23.7% respectively.

A New Continuous Strength Method for Prediction of Strain-Hardening Performance of High-Strength Aluminum Alloy Cylindrical Columns

This paper aims to develop a new continuous strength method (CSM) to more accurately predict the strain-hardening characteristics of high-strength aluminum alloy circular hollow sections (CHSs) under axial compression. A total of 11 stub column specimens made of 7A04-T6 and 6061-T6 aluminum alloys underwent testing. Additionally, 16 sets of experiment data were gathered from open sources, encompassing various aluminum alloy types such as 6082-T6, 6061-T6, and 6063-T5. Validated by experimental result, the finite element (FE) model was applied in a series of comprehensive parameter studies, supplementing the limited test result of high-strength aluminum alloy stub columns. Based on the experiment and FE results, this paper proposes a new CSM relation to determine the cross-sectional resistance of high-strength non-slender CHS aluminum alloys under compression. The cross-sectional resistance obtained from tests are compared with predicted strengths determined using the European code, as well as the solution of the CSM proposed in a previous study and in this paper. The comparison illustrates that the strength predictions in the European code and the previous study are conservative. Compared with the European code and the previous study, the strength prediction formula proposed in this paper improves accuracy by 11% and 5%, respectively, while reducing scatter by 8.4% and 2%, respectively.

Evaluating a new intercrop model for capturing mixture effects with an extensive intercrop dataset

Cereal-legume intercrops have numerous advantages over monocultures. However, the intercrop’s performance depends on the plant genotypes, management, and environment. Process-based agro-ecosystem models are important tools to evaluate the performance of intercrop systems as field experiments are limited in the number of treatments. The objective of this study was to calibrate and evaluate a new process-based intercrop model using an extensive experimental data set and to test whether the model is suitable for comparing intercrop management strategies. The data set includes all combinations of 12 different spring wheat entries (SW, Triticum aestivum L.) with two faba bean (FB, Vicia faba L.) cultivars, at two sowing densities, in three different environments. The results show that the intercrop model was capable of simulating the absolute mixture (intercrop) effects (AME) for grain yield, above-ground biomass, and topsoil root biomass, for both crops. However, the intercrop model does not perform better than a benchmark that ignores the intercrop effects when simulating plant height, fraction of intercepted radiation, volumetric soil water content, and subsoil root biomass. The intercrop model predicted reasonably well the differences between species and between SW cultivars for grain yield and aboveground plant biomass. Overall, the tested process-based model can be a useful tool for designing and pre-evaluation multiple combinations of crop management, species, and cultivars suitable for intercropping in diverse conditions.