Data-driven efficient prediction and design optimisation for high-temperature performance of asphalt mixture

High temperatures, combined with heavy traffic loads, increase the risk of permanent deformation and reduce the useful lifespan of pavements, leading to a greater need for maintenance and rehabilitation activities throughout the pavement's service life. This study aims to evaluate the effectiveness of recycled thermoplastic additives, specifically Multi-Layer Plastic Packaging (MPP) materials, in enhancing the high-temperature performance of asphalt mixtures. MPP materials, including types such as Polyester, Polyethylene, Nylon, and Metalized Polyester, were used for binder modification in asphalt mixtures. The modification was conducted using both wet and dry methods. The physical and rheological properties of the modified binders were assessed at high temperatures to determine their impact on stiffness and resistance to permanent deformation. The results indicate that MPP materials significantly improve the high-temperature performance of asphalt mixtures by enhancing binder stiffness and resistance to permanent deformation, with the wet method proving more effective than the dry method. These findings highlight the need for developing practical guidelines for incorporating MPP materials in asphalt production in the future and underscore the importance of comprehensive studies on maintenance, rehabilitation, and overall pavement lifecycle costs.

In order to study the performance of asphalt mixture with PR.S, Marshall test and rutting, low temperature bending, water stability test were done. The results indicated that the additive PR.S played an important role in improving high-temperature anti-rutting performance of asphalt mixture because of the cementation, reinforcement, inter-lock and adsorption function. With the amount of PR.S increasing, high-temperature performance of asphalt mixture increased gradually and low-temperature performance declined a little. In order to decide the appropriate amount of the additive PR.S, the low-temperature anti-cracking performance should be mainly considered when asphalt mixture was designed. The other performance of asphalt mixture with the amount of 0.45% PR.S could also meet the requirements of the specification. Considering its great contribution to anti-rutting, PR.S asphalt mixture was more properly adopted in the middle layer of asphalt pavement.

Soot blowing optimization is a key, but challenging question in the health management of coal-fired power plant boiler. The monitoring and prediction of ash fouling for heat transfer surfaces is an important way to solve this problem. This study provides a hybrid data-driven model based on advanced machine-learning techniques for ash fouling prediction. First, the cleanliness factor is utilized to represent the level of ash fouling, which is the original data from the distributed control system. The wavelet threshold denoising algorithm is employed as the data preprocessing approach. Based on the empirical mode decomposition (EMD), the denoised cleanliness factor data is decoupled into a series of intrinsic mode functions (IMFs) and a residual component. Second, the support vector regression (SVR) model is used to fit the residual, and the Gaussian process regression (GPR) model is applied to estimate the IMFs. The cleanliness factor data of ash accumulation on the heat transfer surface of diverse devices are deployed to appraise the performance of the proposed SVR + GPR model in comparison with the sole SVR, sole GPR, SVR + EDM and GPR + EDM models. The illustrative results prove that the hybrid SVR + GPR model is superior to other models and can obtain satisfactory effects both in one-step- and the multistep-ahead cleanliness factor predictions.

The morphological properties of coarse aggregates, such as shape, angularity, and surface texture, have a great influence on the mechanical performance of asphalt mixtures. This study aims to investigate the effect of coarse aggregate morphological properties on the high-temperature performance of asphalt mixtures. A modified Los Angeles (LA) abrasion test was employed to produce aggregates with various morphological properties by applying abrasion cycles of 0, 200, 400, 600, 800, 1000, and 1200 on crushed angular aggregates. Based on a laboratory-developed Morphology Analysis System for Coarse Aggregates (MASCA), the morphological properties of the coarse aggregate particles were quantified using the index of fractal dimension. The high-temperature performances of the dense-graded asphalt mixture (AC-16), gap-graded stone asphalt mixture (SAC-16), and stone mastic asphalt (SMA-16) mixtures containing aggregates with different fractal dimensions were evaluated through the dynamic stability (DS) test and the penetration shear test in laboratory. Good linear correlations between the fractal dimension and high-temperature indexes were obtained for all three types of mixtures. Moreover, the results also indicated that higher coarse aggregate angularity leads to stronger high-temperature shear resistance of asphalt mixtures.

Prediction of load-bearing capacity of FRP-steel composite tubed concrete columns: using explainable machine learning model with limited data

Supervised machine learning models and their algorithms play major roles in extrapolative data analysis and getting valuable information of the data. One emerging robust supervised machine learning based model is the Gaussian process regression (GPR) model. A vital strength of the GPR model is its capacity to adaptively model multipath linear/nonlinear functional approximation, dimension reduction and classification problems. Nevertheless, there exist a plethora of approximation (prediction) methods and hyperparameters that systematically impact GPR kernel function for effective modeling and learning capacity. Some of the key approximation methods includes the Exact Gaussian process regression, Fully independent conditional, Subset of data points approximation, and Subset of regressors. Nonetheless, the problem of knowing how to identify or select the best from these approximation methods during data training and learning with GPR model so as to avoid the predictive variance problem is a challenge. In this contribution, we propose a stepwise selection algorithm to tackle the challenge. GPR modelling with stepwise selection algorithm has been tested for extrapolative regression analysis of live cell availability data obtained from an operational telecom service provider. The GPR extrapolative based evaluation results show that the proposed approach is not only tractable, but also yield low errors in terms of accurate service availability quantification and estimation. Also, in terms of mean absolute error, the regression results of the proposed GPR extrapolative based evaluation combined with stepwise kernels selection algorithm were also far better compared with the ones obtained using support vector regression.

Agricultural productivity highly depends on the cost of energy required for cultivation. Thus prior knowledge of energy consumption is an important step for energy planning and policy development in agriculture. The aim of the present study is to evaluate the application potential of multiple linear regression (MLR) and machine learning tools such as support vector regression (SVR) and Gaussian process regression (GPR) to forecast the agricultural energy consumption of Turkey. In the development of the models, widespread indicators such as agricultural value‐added, total arable land, gross domestic product share of agriculture, and population data were used as input parameters. Twenty‐eight‐year historical data from 1990 to 2017 were utilized for the training and testing stages of the models. A Bayesian optimization method was applied to improve the prediction capability of SVR and GPR models. The performance of the models was measured by various statistical tools. The results indicated that the Bayesian optimized GPR (BGPR) model with exponential kernel function showed a superior prediction capability over MLR and Bayesian optimized SVR model. The root mean square error, mean absolute deviation, mean absolute percentage error, and coefficient of determination (R2) values for the BGPR model were determined as 0.0022, 0.0005, 0.2041, and 0.9999 in the training phase and 0.0452, 0.0310, 7.7152, and 0.9677 in the testing phase, respectively. As a result, it can be concluded that the proposed BGPR model is an efficient technique and has the potential to predict agricultural energy consumption with high accuracy.

Performance prognosis of FRCM-to-concrete bond strength using ANFIS-based fuzzy algorithm

The proliferation of photovoltaic (PV) power generation in power distribution grids induces increasing safety and service quality concerns for grid operators. The inherent variability, essentially due to meteorological conditions, of PV power generation affects the power grid reliability. In order to develop efficient monitoring and control schemes for distribution grids, reliable forecasting of the solar resource at several time horizons that are related to regulation, scheduling, dispatching, and unit commitment, is necessary. PV power generation forecasting can result from forecasting global horizontal irradiance (GHI), which is the total amount of shortwave radiation received from above by a surface horizontal to the ground. A comparative study of machine learning methods is given in this paper, with a focus on the most widely used: Gaussian process regression (GPR), support vector regression (SVR), and artificial neural networks (ANN). Two years of GHI data with a time step of 10 min are used to train the models and forecast GHI at varying time horizons, ranging from 10 min to 4 h. Persistence on the clear-sky index, also known as scaled persistence model, is included in this paper as a reference model. Three criteria are used for in-depth performance estimation: normalized root mean square error (nRMSE), dynamic mean absolute error (DMAE) and coverage width-based criterion (CWC). Results confirm that machine learning-based methods outperform the scaled persistence model. The best-performing machine learning-based methods included in this comparative study are the long short-term memory (LSTM) neural network and the GPR model using a rational quadratic kernel with automatic relevance determination.

The performance of an asphalt mixture is significantly affected by the properties of its asphalt mortar, which consists of an asphalt binder, mineral fillers, fine aggregates and air voids. The aim of this work was to evaluate the correlations between the high-temperature performance of an asphalt mixture and the rheological properties of its corresponding asphalt mortar. The multisequence repeated loading (MSRL) test was used to estimate the high-temperature performance of the asphalt mixture. Six different gradations, AC-13, SMA-13, SUP-13, AC-20, SUP-20 and AC-25, and two styrene–butadiene–styrene (SBS)-modified asphalt binders were considered and used to prepare the asphalt mixture specimens. The gradations and asphalt types of asphalt mortars were consistent with their asphalt mixtures. A modified multiple-stress creep–recovery (MSCR) test was proposed for evaluating the rheological properties of asphalt mortar with a dynamic shear rheometer (DSR). Based on the basic form of the Hirsh model, a multiple regression model was established, and its coefficient of determination (R-square) was 0.96. The rheological response of the asphalt mortar presented great correlation with the high-temperature behaviour of the asphalt mixture. In addition, the MSCR indicators (nonrecoverable compliance and percent recovery) obtained at 12.8 kPa creep stress represented the rheological status of asphalt mortar in asphalt mixture well. Therefore, the mechanical behaviours of asphalt mixture at high temperature could be accurately predicted by the MSCR indicators of asphalt mortar and its coarse aggregate parameters.

The present study used three machine learning models, including Least Square Support Vector Regression (LSSVR) and two non-parametric models, namely, Quantile Regression Forest (QRF) and Gaussian Process Regression (GPR), to quantify uncertainty and precisely predict the side weir discharge coefficient (Cd) in rectangular channels. So, 15 input structures were examined to develop the models. The results revealed that the machine learning models used in the study offered better accuracy compared to the classical equations. While the LSSVR and QRF models provided a good prediction performance, the GPR slightly outperformed them. The best input structure that was developed included all four dimensionless parameters. Sensitivity analysis was conducted to identify the effective parameters. To evaluate the uncertainty in the predictions, the LSSVR, QRF, and GPR were used to generate prediction intervals (PI), which quantify the uncertainty coupled with point prediction. Among the implemented models, the GPR and LSSVR models provided more reliable results based on PI width and the percentage of observed data covered by PI. According to point prediction and uncertainty analysis, it was concluded that the GPR model had a lower uncertainty and could be successfully used to predict Cd.

Supercritical carbon dioxide (Sc-CO2) has thus been proposed as an appropriate solvent for diluting the pharmaceuticals to increase particle size. The use of supercritical fluids (SCFs) in various industrial applications, such as extraction, chromatography, and particle engineering, has attracted considerable interest. Recognizing the solubility behavior of various drugs is an essential step in the pharmaceutical industry's pursuit of the most effective supercritical approach. In this work, four models were used to predict the solubility of Azathioprine in supercritical carbon dioxide, including Ridge regression (RR), Huber regression (HR), Random forest (RF), and Gaussian process regression (GPR). The R-squared scores of all four models are 0.974, 0.6518, 0.966, and 1.0 for Ridge regression (RR), Huber regression (HR), Random forest (RF), and Gaussian process regression (GPR) models, respectively. The RMSE error rates of 2.843 ×10-13, 7.036 ×10-12, 5.673 ×10-13, and 1.054 ×10-30 for the RR, HR, RF, and GPR models, respectively. MAE metrics of 1.205 ×10-6, 2.151 ×10-6, 5.997 ×10-7 and 9.419 ×10-16 errors were also found in the RR, HR, RF, and GPR models, respectively. It was found that Ridge regression (RR), Random forest (RF), and Gaussian process regression (GPR) models can be used to predict any compound's solubility in supercritical carbon dioxide.

Gaussian process-based Bayesian optimization (GPBO) is used to search parameters in machine learning, material design, etc. It is a method for finding optimal solutions in a search space through the following four procedures. (1) Develop a Gaussian process regression (GPR) model using observed data. (2) The GPR model is used to obtain the estimated mean and estimated variance for the search space. (3) The point where the sum of the estimated mean and the weighted estimated variance (upper confidence bound, UCB) is largest is the next search point (in the case of a maximum search). (4) Repeat the above procedures. Thus, the generalization performance of the GPR is directly related to the search performance of the GPBO. In procedure (1), the kernel parameters (KPs) of the GPR are tuned via gradient descent (GD) using the log-likelihood as the objective function. However, if the number of iterations of the GD is too high, there is a risk that the KPs will overfit the observed data. In this case, because the estimated mean and variance output by the GPR model are inappropriate, the next search point cannot be properly determined. Therefore, overtuned KPs degrade the GPBO search performance. However, this negative effect can be mitigated by changing the parameters of the GPBO. We focus on the weight of the estimated variances (exploration weight) of the UCB as one of these parameters. In a GPBO with a large exploration weight, the observed data appear in various regions in the search space. If the KP is tuned using such data, the GPR model can estimate the diverse regions somewhat correctly, even if the KP overfits the observed data, i.e., the negative effect of overtuned KPs on the GPR is mitigated by setting a larger exploration weight for the UCB. This suggests that the negative effect of overtuned KPs on the GPBO search performance may be related to the UCB exploration weight. In the present study, this hypothesis was tested using simple numerical simulations. Specifically, GPBO was applied to a simple black-box function with two optimal solutions. As parameters of GPBO, we set the number of KP iterations of GD in the range of 0–500 and the exploration weight as {1,5}. The number of KP iterations expresses the degree of overtuning, and the exploration weight expresses the strength of the GPBO search. The results indicate that, in the overtuned KP situation, GPBO with a larger exploration weight has better search performance. This suggests that, when searching for solutions with a small GPBO exploration weight, one must be careful about overtuning KPs. The findings of this study are useful for successful exploration with GPBO in all situations where it is used, e.g., machine learning hyperparameter tuning.

Static Poisson's ratio (νs) is an essential property used in petroleum calculations, namely fracture pressure (FP). The νs is often determined in the laboratory; however, due to time and cost constraints, quicker and cheaper alternatives are sought, such as data-driven models. However, existing methods lack the accuracy needed for critical applications, necessitating the need to explore more accurate methods. In addition, the previous studies used limited datasets and they do not show the relationships between the inputs and output. Therefore, this study developed a reliable model to predict the νs accurately using the nineteen most common learning methods. The proposed models were created based on a large data of 1691 datasets from different countries. The best-performing model of the nineteen models was selected and further enhanced using various approaches such as trend analysis to improve the model's performance and robustness as some models show high accuracy but show incorrect relationships between the inputs and output because the machine learning model only built based on the data and do not consider the physical behavior of the model. The proposed Gaussian process regression (GPR) model was also compared with published models. After the proposed GPR model was developed, the FP was determined based on the proposed GPR νs model and the previous νs models to evaluate their accuracy on the FP determinations. The best approach out of the published and proposed methods was GPR with a coefficient of determination (R2) and average-absolute-percentage-relative-error (AAPRE) of 0.95 and 2.73%. The GPR model showed proper trends for all inputs. The cross-plotting and group error analyses also confirmed that the proposed GPR approach had high precision and surpassed other methods within all practical ranges. The GPR model decreased the residual error of FP from 87% to 26%. It is believed that such a significant improvement in the accuracy of the GPR model will have a significant effect on realistic FP determination.

Improved lithium-ion battery health prediction with data-based approach