Test Data Research Articles

Complete pipeline for Oxford Nanopore Technology amplicon sequencing (ONT-AmpSeq): from pre-processing to creating an operational taxonomic unit table.

Amplicon sequencing has long served as a robust method for characterising microbial communities, and despite inherent resolution limitations, it remains a preferred technique, offering cost- and time-effective insights into bacterial compositions. Here, we introduce ONT-AmpSeq, a user-friendly pipeline designed for processing amplicon sequencing data generated from Oxford Nanopore Technology (ONT) devices. Our pipeline enables efficient creation of taxonomically annotated operational taxonomic unit (OTU) tables from ONT sequencing data, with the flexibility to multiplex amplicons on the same barcode. The pipeline encompasses six main steps-statistics, quality filtering, alignment, clustering, polishing, and taxonomic classification-integrating various state-of-the-art software tools. We provide a detailed description of each step, along with performance tests and robustness evaluations using both test data and a ZymoBIOMICS® Microbial Community Standard mock community dataset. Our results demonstrate the ability of ONT-AmpSeq to effectively process ONT amplicon data, offering valuable insights into microbial community composition. Additionally, we discuss the influence of polishing tools on taxonomic insight and the impact of taxonomic annotation methods on the derived microbial composition. Overall, ONT-AmpSeq represents a comprehensive solution for analysing ONT amplicon sequencing data, facilitating streamlined and reliable microbial community analysis. The pipeline, along with test data, is freely available for public use.

Wire rope isolator identification and dynamic modeling for small amplitude vibrations

Wire rope isolators are used in variety of applications to protect sensitive equipment from vibration. The nonlinear hysteretic behavior of steel wires provides advantages when compared to linear vibration isolators. This study proposes an amplitude dependent stiffness and damping model for low amplitude vibrations under axial loading, where the parameters can be determined with an experimental procedure. Comprehensive experimental results of forced vibration tests with varying loading, frequency and preload were considered in the model identification. Amplitude dependent stiffness and loss energy models were determined from the test data, and the effect of the preload and loading frequency on the model parameters were studied. It is shown, that the effect of preload and frequency is not evidently clear, while the effect of vibration amplitude is more significant. The mathematical model was further verified against measurements from base excitation loading. The proposed model can be used to study the effectiveness of the selected wire rope isolator configurations in chosen application, and to effectively perform dynamic design studies.

The performance comparison of machine learning methods for solar PV power prediction

Purpose The energy generation process through photovoltaic (PV) panels is contingent upon uncontrollable variables such as wind patterns, cloud cover, temperatures, solar irradiance intensity and duration of exposure. Fluctuations in these variables can lead to interruptions in power generation and losses in output. This study aims to establish a measurement setup that enables monitoring, tracking and prediction of the generated energy in a PV energy system to ensure overall system security and stability. Toward this goal, data pertaining to the PV energy system is measured and recorded in real-time independently of location. Subsequently, the recorded data is used for power prediction. Design/methodology/approach Data obtained from the experimental setup include voltage and current values of the PV panel, battery and load; temperature readings of the solar panel surface, environment and the battery; and measurements of humidity, pressure and radiation values in the panel’s environment. These data were monitored and recorded in real-time through a computer interface and mobile interface enabling remote access. For prediction purposes, machine learning methods, including the gradient boosting regressor (GBR), support vector machine (SVM) and k-nearest neighbors (k-NN) algorithms, have been selected. The resulting outputs have been interpreted through graphical representations. For the numerical interpretation of the obtained predictive data, performance measurement criteria such as mean absolute error (MAE), mean squared error (MSE), root mean squared error (RMSE) and R-squared (R2) have been used. Findings It has been determined that the most successful prediction model is k-NN, whereas the prediction model with the lowest performance is SVM. According to the accuracy performance comparison conducted on the test data, k-NN exhibits the highest accuracy rate of 82%, whereas the accuracy rate for the GBR algorithm is 80%, and the accuracy rate for the SVM algorithm is 72%. Originality/value The experimental setup used in this study, including the measurement and monitoring apparatus, has been specifically designed for this research. The system is capable of remote monitoring both through a computer interface and a custom-developed mobile application. Measurements were conducted on the Karabük University campus, thereby revealing the energy potential of the Karabük province. This system serves as an exemplary study and can be deployed to any desired location for remote monitoring. Numerous methods and techniques exist for power prediction. In this study, contemporary machine learning techniques, which are pertinent to power prediction, have been used, and their performances are presented comparatively.

Semisupervised Learning to Boost hERG, Nav1.5, and Cav1.2 Cardiac Ion Channel Toxicity Prediction by Mining a Large Unlabeled Small Molecule Data Set.

Predicting drug toxicity is a critical aspect of ensuring patient safety during the drug design process. Although conventional machine learning techniques have shown some success in this field, the scarcity of annotated toxicity data poses a significant challenge in enhancing models' performance. In this study, we explore the potential of leveraging large unlabeled small molecule data sets using semisupervised learning to improve drug cardiotoxicity predictive performance across three cardiac ion channel targets: the voltage-gated potassium channel (hERG), the voltage-gated sodium channel (Nav1.5), and the voltage-gated calcium channel (Cav1.2). We extensively mined the ChEMBL database, comprising approximately 2 million small molecules, and then employed semisupervised learning to construct robust classification models for this purpose. We achieved a performance boost on highly diverse (i.e., structurally dissimilar) test data sets across all three targets. Using our built models, we screened the whole ChEMBL database and a large set of FDA-approved drugs, identifying several compounds with potential cardiac ion channel activity. To ensure broad accessibility and usability for both technical and nontechnical users, we developed a cross-platform graphical user interface that allows users to make predictions and gain insights into the cardiotoxicity of drugs and other small molecules. The software is made available as open source under the permissive MIT license at https://github.com/issararab/CToxPred2.

Research on the static and dynamic tensile damage mechanism of concrete based on fuzzy sets

The quantitative analysis of microscopic damage mechanisms in concrete and the formulation of damage evolution equations using mathematical theory and computerized tomography (CT) imaging are still emerging fields. The test data from CT scans have yet to be fully exploited. This study proposes a method for quantitatively characterizing mesoscale damage evolution in concrete. By segmenting CT images from static and dynamic tensile tests, we define metrics such as the aggregate rate, mortar rate, and porosity to examine the damage progression under both stress conditions. We then establish a damage evolution model based on porosity changes. Our findings reveal that crack growth in concrete is slower under static tension, with cracks navigating around aggregates towards weaker regions. Conversely, under dynamic loads, cracks propagate more rapidly and directly through aggregates, following paths of quicker energy dispersion. Static loading results in less aggregate damage than does dynamic loading, with a notable increase in porosity and a more varied change in the mortar rate. The damage to the finer components of concrete increases as the crack-adjacent region diminishes. Damage is generally more extensive under static loading than under dynamic loading. The proposed porosity-based damage evolution model effectively describes concrete damage under static tension, leveraging CT technology to enhance the universality of the research and offering a microscopic-level theoretical foundation for studying concrete mechanical properties.

The role of manufacturing-induced texture on the tribological performance of cold work tool steels

Manufacturing operations produce surface characteristics that, although stochastic, can significantly affect functionality, especially in forming tools, impacting contact and lubrication conditions during operation. This study investigates the influence of stochastic microtextures resulting from milling on the tribological performance of cold work tool steels with two different carbon contents (0.8 and 2 wt%). Different surface textures were observed resulting from the different C contents, with 3D roughness parameters indicating rougher surfaces for the 2.0% wt. C steel. Tribological behavior was assessed using the strip drawing test to analyze friction, wear coefficients, and wear mechanisms. Surface analysis before and after testing was executed employing SEM, EDX, and CLSM, with CLSM also used to determine 3D roughness parameters of the worn tracks. Post-test macrographic analyses and 2D roughness measurements were conducted on the pulled sheets. Tribological test data revealed lower friction and wear coefficients for the 2.0 wt% C tool steel, with susceptibility to abrasion wear, while the 0.8 wt% C tool exhibited a higher tendency towards adhesion wear. Post-test analysis suggested smoother surfaces for the 2.0 wt% C steel compared to the 0.8 wt% C steel. Macrographic analysis showed no visible wear marks on sheets tested with the 2.0 wt% C steel, contrasting with wear grooves visible on sheets pulled against the 0.8 wt% C steel. Additionally, 2D roughness measurements indicated higher roughness after pulling against the 0.8 wt% C tool compared to the 2.0 wt. C tool. Overall, the study demonstrates that manufacturing-induced textures without the need of post-manufacturing texturing influence the tribological performance of the evaluated steels, opening an avenue to be explored to improve the tribological performance of forming tools.

Innovative Reward-Based Crowdfunding Decision Model

Due to the risky nature of newly creative projects for entrepreneurs, reward-based crowdfunding is currently an alternative fundraising channel for those who need seed funding to finance the creation of their prototype. The objectives of this research are to explore the success factors, including entrepreneurial, project and campaign factors, in project fundraising under a reward-based crowdfunding platform. We propose to develop a model for predicting the success of crowdfunding projects by machine learning. The datasets have been retrospectively gathered from historical records of campaigns in the Kickstarter website. The study's findings show that the logistic regression and decision tree models, respectively, had accuracy rates of 88.2% and 88.8%. The highest accuracy percentage of 94.1% originates from new testing data that has been externally validated for the technology industry. The practical implication of our research is that entrepreneurs can apply the proposed prediction model to identify the most influential topical features embedded in campaigns.

An optimized multiaxial fatigue life prediction model based on the subcritical plane method

Since a more rigorous requirement is raised for the performance of aero-engine increases, the multiaxial fatigue problem of its components turns out to be increasingly severe. In this study, a multiaxial fatigue life prediction model is developed by using the subcritical plane method, where the effect of non-proportional additional hardening under non-proportional loading is considered by both the deflection angle of the subcritical plane and the combined parameters of strain on the subcritical plane. Moreover, the cyclic strain hardening index in the traditional model is substituted with an additional strengthening material coefficient that is characterized by the static strength of the material, which reduces the difficulty in calibrating the parameters of the multiaxial life model. The effectiveness of the proposed model is validated by comparing the predicted fatigue life with some fatigue test data, and the errors between the predicted and the testing data fall into a factor of two error bands. In general, the proposed multiaxial life prediction model simplifies the calibration of model parameters, and the prediction accuracy is still ensured, which promotes the model to facilitate engineering application.

IMPLEMENTASI TEKNIK SMOTE UNTUK MENGATASI IMBALANCE CLASS DALAM KLASIFIKASI SENTIMEN MENGENAI CHATGPT

ChatGPT is a chatbot or computer program in the form of a virtual robot that can simulate human-like conversations. ChatGPT is widely used in various fields in academia. The impact of the use of ChatGPT on academia and public perception of this technology is significant. Sentiment analysis can be used to determine the polarity of a text or opinion that is positive or negative. In this research, social media is used as a data source to collect public opinion regarding ChatGPT instantly. The methods used in this reserach are the KNN algorithm and Naive Bayes algorithm. The aim of this research is to find the best algorithm model for sentiment classification in terms of public opinion for ChatGPT which contains English text. Before testing the algorithm model, a text processing stage was carried out which included the processes of case folding, tokenizing, stopword removal, and stemming. Word weighting using TF-IDF was carried out before the data was ready to be processed. Splitting data used in this research includes 80% of the dataset as training data and 20% of the dataset as testing data. The application of the SMOTE technique to the KNN and Naive Bayes algorithms to overcome the imbalance class of the public opinion dataset regarding ChatGPT. The research results show that combining SMOTE and Naive Bayes algorithm gives the best results with an accuracy value of 85.00%, a precision value of 87.64%, a recall value of 84.78% and an f1-score of 86.18%.

Advancements In Machine Learning Algorithms for Enhanced Fault Analysis and Categorization in Power Systems

With the growing number of sources, complexity, and dynamic elements of power systems that are prone to disturbance and electrical faults, conventional methods of fault detection are lagging in power system protection. Major faults are exposed to transmission lines due to several factors, such as climate change, heat, lightning strokes, sudden spikes of current and voltage etc. Protecting these lines from faults is cumbersome using traditional methods. However, with the increasing complexity of transmission lines in power systems, advanced technologies such as machine learning come into play for prior fault detection and identification. ML is becoming a popular and intelligent technique for monitoring and predicting health and diagnosing faults in transmission lines of power systems. Since ML is evolving at a faster pace and given the need for and growth of advanced technologies like ML and artificial intelligence, this paper emphasizes comparative analysis of various ML algorithms for transmission line fault analysis. Comparisons for various line faults (LL, LLG, LG, LLL, LLLG, no fault) in power systems using machine learning techniques have been done in terms of accuracy, precision, and other performance measures. The results of the accuracy of various algorithms are represented as tabulated facts in this paper. The Python programming language has been utilized over a dataset to calculate results. For each algorithm, the confusion matrix shows the accuracy of predicting faults over the testing data after training the model or algorithm on the training data. Upon completion of analysis and comparison of MLT's like LR, SVM, DT, RF, K-NN, and gradient descent for analysis of transmission line fault detection, RF and DT machine learning algorithms provided the best accuracy and results. Finally, this paper gives a brief overview of various line faults, ML algorithms accuracy for each transmission fault in a confusion matrix and tabular format and concluded the best machine learning algorithms for power system transmission line fault identification.

Effectiveness of Video Assisted Teaching on Knowledge and Pre-procedure Anxiety Level of Patients Undergoing Upper GI Endoscopy

An endoscopy is a procedure in which doctor uses specialized instruments to view and operate on the internal organs and vessels.People often found anxious towards hospital environment itself. There no intervention was provided to re-leave from anxiety among the patients. So the researchers aimed to assess the effectiveness of video assisted teaching regarding upper GI endoscopy to improve knowledge and thereby reduce the anxiety among the patients in Government General Hospital, Puducherry.The present study was adopted quantitative quasi experimental research design. The 50 samples were selected by purposive sampling technique with inclusion and exclusion criteria. Pre test data was collected by using self-administered structures questionariesand video assisted teaching programme regarding endoscopy to the samples. The video contains introduction, functions, procedure and visualization of the endoscopy.The researchers accompany each sample during the endoscopy. Post test data was collected after endoscopy with the same questionnaire.The results revealed that, pre and post test mean score of knowledge was 5.26±2.562 and 11.12±0.480. The calculated paired ‘t’ value is 16.421 which was statistically significant at the p level 0.000. The pre and post test mean score of anxiety was 39.64±9.021 and 68.14±3.995. The calculated paired ‘t’ test value was 18.475 which was statically significant at the p level 0.000. The test results infer that, the video assisted programme was effective in improving knowledge and relive in pre procedure anxiety among the clients undergoing endoscopy.The chi-square value reveals that the level of knowledge was significant to gender and educational status at the p level 0.005 respectively and the level of anxiety was significant to educational status at the p level 0.005. The researchers recommended all the therapeutic and diagnostic procedures must be educated among the patients will help to gain complete success.

Damage identification of reinforced concrete slabs using experimental modal testing and finite element model updating

Condition assessment and health monitoring of reinforced concrete structures is an essential part of their effective maintenance throughout their service life. Vibration-based damage detection and health monitoring is one of the most popular global approaches where experimentally measured modal parameters, such as frequencies, mode shapes, etc. are compared to their corresponding finite element counterparts to estimate possible stiffness loss. The present investigation implements a vibration-based parameter estimation algorithm for updating a finite-element (FE) model for damage assessment of reinforced concrete slabs using measured frequency response functions (FRF) data through a mixed numerical-experimental technique. An inverse eigen-sensitivity algorithm has been implemented to minimize the error function realized as the weighted differences in the mode shapes and frequencies of the FE model and the modal test data. Three reinforced concrete slabs were subjected to real physical damage conditions to verify the efficacy of the proposed methodology. The technique is found to be resilient in the presence of modal and coordinate sparsity. In general, damage can be assessed even if the portions of the structure are inaccessible to the users, thereby extending the possibility of its application to most of the practical configurations of reinforced concrete slabs.

Artificial neural network hyperparameters optimization for predicting the thermal conductivity of MXene/graphene nanofluids

BackgroundThe critical role of thermal conductivity (TC) as a significant thermo-physical property in MXene/graphene-based nanofluids for photovoltaic/thermal systems has motivated recent research into developing precision predictive models. The multilayer perceptron neural network (MLPNN) has emerged as an eminent AI algorithm for this task. MethodsThis study employs Bayesian optimization, random search (RS), and grid search (GS) to fine-tune MLPNN hyperparameters—hidden layers, neurons, activation functions, standardization, and regularization—to elevate TC modeling efficiency. The proposed methodology unfolds in sequential phases: data analysis, data pre-processing, and introduction of MLPNN, GS, RS, Bayesian approach, and their integration algorithm. The next phase entails developing predictive models and presenting optimal cases. Lastly, the final models undergo statistical evaluation and graphical comparison for a thorough analysis. FindingsResults manifest that the GS-MLPNN model excels, achieving the lowest testing data error (MAPE = 0.5261%) and high conformity with empirical data (R = 0.99941). Meanwhile, the RS method adjusts hyperparameters with negligible precision loss (MAPE = 0.6046%, R = 0.99887). Contrarily, Bayesian optimization lags, increasing errors (MAPE = 3.1981%) and lower correlation (R = 0.98099), suggesting its relative inefficacy for this specific application. The optimized models provide efficient predictions, significantly reducing the financial/computing costs associated with experimental/numerical analysis.

Stroke Research Using Administrative Claims Database in Japan: A Narrative Review.

Although administrative claims databases have recently been used for clinical research in Japan, no detailed description of their utilization in stroke research is available. We reviewed stroke studies using the Diagnosis Procedure Combination (DPC), the National Database of Health Insurance Claims and Specific Health Checkups (NDB), and several commercial databases sourced from social health insurance associations, focusing on their applications and limitations. Original articles on stroke published by April 2024 using the DPC, NDB, and commercial databases were identified in Ovid MEDLINE. The characteristics of each database were compared in terms of comprehensiveness, traceability, baseline information, and outcome assessment in stroke research. A total of 114 studies were included (83 for DPC, 6 for NDB, and 25 for commercial databases). The number of stroke studies using administrative databases in Japan is still approximately 10 per year, although there is a slowly increasing trend. The DPC database was utilized for short-term outcome studies because of its detailed baseline and outcome information, although the inability to track patients once they changed facilities limits their use in long-term studies. The NDB database is potentially useful for long-term studies because of its comprehensiveness and traceability, but difficulties in data access restrict its usage. The most commonly used commercial database utilizes baseline information on lifestyle and blood test data, although the lack of coverage for those over 75 years old may limit its generalizability. Administrative claims databases are beginning to be used in stroke research in Japan but are not yet fully utilized. Researchers need to understand their applications and limitations.

Comparison of Seasonal ARIMA and Support Vector Machine Forecasting Method for International Arrival in Lombok

Seasonal Autoregressive Integrated Moving Average is a statistical model designed to analyze and forecast data with that shows seasonal patterns and trends. Support Vector Machine (SVM) is a machine learning-based technique that can be used to forecast time series data. SVM uses the kernel tricks to overcome non-linearity problems, whereas The SARIMA model is well-suited for data that exhibit seasonal fluctuations that repeat over time. Lombok International Airport is the main gateway to West Nusa Tenggara and has become a symbol of tourism growth in the region. Time series analysis is a very useful tool in determining patterns and forecasting the number of international arrivals at Lombok International Airport within a certain period. This study aims to compare the SARIMA model and SVM which can read non-linear patterns in the number of international arrivals at Lombok International Airport. After obtaining the SARIMA and SVM models, the two models are evaluated using test data based on the smallest RMSE value. The SVM model with a linear kernel trick provides the smallest RMSE when compared to SARIMA with SVM RMSE is 238,655. While the best model in Seasonal ARIMA is SARIMA (3,1,0)(1,0,0)12, the forecasting results show SARIMA is better in the forecasting process for the next 10 months.

Shaking table test on progressive collapse of the double-layer cylindrical reticulated shell with infilled wall

Double-layer cylindrical reticulated shell has made rapid development in recent decades in China, which is widely used in the large-scale public buildings such as gymnasiums, airplane terminal, Shopping mall, etc. Stability and progressive collapse mechanism have been attracting a lot of attention because the large span space structure is largely affected by the geometry nonlinear, which makes it sensitive to the local damages. The infilled wall has a greater impact on the stiffness of the lower supporting structure due to sparse column grid. However, the influence of infilled walls on the collapse mechanism of large-span spatial structures is very complex due to the coupling of upper and lower structures. Therefore, the effects of infilled walls on double-layer cylindrical reticulated shell was investigated by a series of progressive collapse tests with shaking table. In order to evaluate the seismic performance and the capacity of resisting progressive collapse, the structural dynamic characteristics and response laws were obtained under different ground motions and input principal directions. The results indicated that the existence of infilled wall made the coupling between the lower supporting structure and the upper roof structure more complicated. The infilled walls influenced the structural damage evolution and development, mitigated the damage from the earthquake and changed the progressive collapse pattern, which made the catenary collapse mechanism occur when the structure collapsed. Although the further quantitative study is still needed, this paper provides lots of useful test data and conclusions, which gives references to the infilled wall design and seismic performance evaluation for double-layer cylindrical reticulated shell.

Detection of Hunting Pits using Airborne Laser Scanning and Deep Learning

ABSTRACT Forests worldwide contain unique cultural traces of past human land use. Increased pressure on forest ecosystems and intensive modern forest management methods threaten these ancient monuments and cultural remains. In northern Europe, older forests often contain very old traces, such as millennia-old hunting pits and indigenous Sami hearths. Investigations have repeatedly found that forest owners often fail to protect these cultural remains and that many are damaged by forestry operations. Current maps of hunting pits are incomplete, and the locations of known pits have poor spatial accuracy. This study investigated whether hunting pits can be automatically mapped using national airborne laser data and deep learning. The best model correctly mapped 70% of all the hunting pits in the test data with an F1 score of 0.76. This model can be implemented across northern Scandinavia and could have an immediate effect on the protection of cultural remains.

Examining the mediating and moderating roles of social marketing behavioural enhancers regarding consistent condom use.

Since their importance was highlighted by Ayikwa and De Jager, social marketing behavioural enhancers (SMBEs) have been investigated in terms of their causal and correlational relationships with the intention to use condoms (IUC) and consistent condom use (CCU), as people often fail to act on their intentions. However, scrutiny of their mediating and moderating roles could provide insights pertinent to the design of effective HIV and AIDS preventative programmes. This study examined whether perceived behavioural control (PBC) and IUC interact with exposure to HIV and AIDS information (EI), ease of access to condoms (EAC) and level of related knowledge (KN) in determining CCU. It also investigated whether PBC and IUC predict CCU through increasing EI, EAC, and KN. A quantitative approach was adopted and data were collected from 607 participants, aged at least 18, living in Gauteng Province, South Africa. The questionnaire administered included pre-existing items, validated through exploratory and confirmatory factor analysis procedures. Regression analyses of the data for mediation and moderation testing were performed using PROCESS macro software for SPSS. The results indicated that none of the SMBEs mediated the non-significant PBC-CCU relationship: B = -0.0258, SE = 0.0199, p = 0.195. Nor did they mediate the significant IUC-CCU relationship: B = 0.0395, SE = 0.0195, p = 0.043. Similarly, none of the SMBEs were found to moderate the PBC-CCU relationship (EI*PBC: B = 0.0034, SE = 0.0056, p = 0.540; KN*PBC: B = -0.0006, SE = 0.0064, p = 0.931; EAC*PBC: B = 0.0011, SE = 0.0059, p = 0.854) as IUC-CCU relationship (EI*IUC: B = 0.0036, SE = 0.0054, p = 0.513; KN*IUC: B = -0.0096, SE = 0.0060, p = 0.111; EAC*IUC: B = 0.0044, SE = 0.0061, p = 0.469). A recommendation is made to scrutinise the mediating and moderating roles of SMBEs in the context of health behavioural models other than the theory of planned behaviour, which was considered in this study.

CHINESE YUAN EXCHANGE RATE AGAINST THE INDONESIAN RUPIAH PREDICTION USING SUPPORT VECTOR REGRESSION

This study aims to forecast the exchange rate between the Chinese Yuan (CNY) and the Indonesian Rupiah (IDR) using Support Vector Regression (SVR), a machine-learning technique that can handle nonlinear and complex data. The authors utilize the monthly selling exchange rate of CNY against IDR from January 2012 to October 2023 sourced from the “investing” platform. The optimal SVR model is obtained by splitting the data into 113 training samples and 28 testing samples and using the Radial Basis Function (RBF) kernel. The model achieves high accuracy, with a Mean Absolute Percentage Error (MAPE) of 1.738%, a Root Mean Squared Error (RMSE) of 50.661 for the training data and a MAPE of 2.516%, and an RMSE of 64.735 for the testing data. The results of this paper can provide valuable insights for policymakers, investors, and traders who are interested in the CNY/IDR exchange rate dynamics and the economic implications of the Belt and Road Initiative (BRI). The study aligns with the Sustainable Development Goals (SDGs), specifically SDG 8, aiming to promote sustained, inclusive, and sustainable economic growth.

PC-Kriging-powered parallelizing Bayesian updating for stochastic vehicle-track dynamical system with contact force measurements and Gaussian process discrepancy model

High-precision vehicle-track coupled dynamical models play a vital role in assessing the running safety of the vehicle, tracking fatigue behavior, and establishing a digital twin for high-speed railways. This study established a high-fidelity vehicle-track coupled dynamical model as a forward solver, which was then updated in a Bayesian inference framework based on the field tests of rail irregularities and wheel/rail normal contact forces. The prediction errors corresponding to different time steps were represented through a Gaussian process (GP) model characterized by a covariance function with the Gaussian kernel to incorporate the correlation between different time steps. To cope with the considerable expense of the likelihood function calculations with large-scale loop operations and the computational burden due to the large batch of repetitive evaluations of the high-fidelity forward model involved in the stochastic sampling from the posterior probability distribution, a metamodel-powered parallelizing stochastic sampling scheme was adopted. The posterior distribution could be formulated by replacing the expensive explicit model evaluation involved in the likelihood function with a Polynomial-Chaos-Kriging (PC-Kriging) predictor providing a surrogate mapping between the wheel/rail normal contact forces and physical parameters. Finally, a parallelizing computing scheme running across multiple CPU cores on high-performance computers was realized to sample from the vectorized formula of the emulator-powered posterior distribution. Field test data from a high-speed railway in China was adopted to demonstrate that the Bayesian inference scheme can quantify the uncertainties of the core physical parameters of the high-fidelity vehicle-track coupled dynamical system by efficiently trading off accuracy and computational efficiency.