Search Results (1,580)

To address critical limitations of ultra-low viscosity supercritical CO₂ fracturing fluids, including excessive fluid loss and inadequate proppant transport capacity, a series of thickeners designed to significantly enhance CO₂ viscosity were synthesized. Initially, FT-IR and ¹H NMR characterization confirmed successful chemical reactions and incorporation of both solvation-enhancing and -thickening functional groups. Subsequently, dissolution and thickening performance were evaluated using a custom-designed high-pressure vessel featuring visual observation capability, in-line viscosity monitoring, and high-temperature operation. All thickener systems exhibited excellent solubility, with 5 wt% loading elevating CO₂ viscosity to 3.68 mPa·s. Ultimately, molecular simulations performed in Materials Studio elucidated the mechanistic basis, electrostatic potential (ESP) mapping, cohesive energy density analysis, intermolecular interaction energy, and radial distribution function comparisons. These computational approaches revealed dissolution and thickening mechanisms of polymeric thickeners in CO₂. Full article

This research develops a reliable methodology for estimating minimum temperature distribution in agricultural areas, focusing on frost conditions threatening crop production. The data was collected across the plain of Krya Vrysi in Central Macedonia. The approach uses linear regression equations between daily minimum temperatures from a central station and 12 autonomous temperature sensors with data loggers. Statistical analysis covered winter 2023–2024, with 2025 validation showing exceptional predictive capability—R² values of 0.97–0.99 and RMSE of 0.34–0.58 °C. Spatial interpolation employed the Radial Basis Function with thin plate splines, effective for agricultural microclimatic interpolation. This methodology provides an operational frost prediction tool, enabling targeted interventions, reducing production losses and enhancing agricultural resilience. Full article

Entropy-based analyses have emerged as a powerful tool for quantifying the complexity, regularity, and information content of complex biological signals, such as electroencephalography (EEG). In this regard, EEG-based lie detection offers the advantage of directly providing more objective and less susceptible-to-manipulation results compared to traditional polygraph methods. To this end, this study proposes a novel multi-scale entropy approach by fusing fuzzy entropy (FE), time-shifted multi-scale fuzzy entropy (TSMFE), and hierarchical multi-band fuzzy entropy (HMFE), which enables the multidimensional characterization of EEG signals. Subsequently, using machine learning classifiers, the fused feature vector is applied to lie detection, with a focus on channel selection to investigate distinguished neural signatures across brain regions. Experiments utilize a publicly benchmarked LieWaves dataset, and two parts are performed. One is a subject-dependent experiment to identify representative channels for lie detection. Another is a cross-subject experiment to assess the generalizability of the proposed approach. In the subject-dependent experiment, linear discriminant analysis (LDA) achieves impressive accuracies of 82.74% under leave-one-out cross-validation (LOOCV) and 82.00% under 10-fold cross-validation. The cross-subject experiment yields an accuracy of 64.07% using a radial basis function (RBF) kernel support vector machine (SVM) under leave-one-subject-out cross-validation (LOSOCV). Furthermore, regarding the channel selection results, PZ (parietal midline) and T7 (left temporal) are considered the representative channels for lie detection, as they exhibit the most prominent occurrences among subjects. These findings demonstrate that the PZ and T7 play vital roles in the cognitive processes associated with lying, offering a solution for designing portable EEG-based lie detection devices with fewer channels, which also provides insights into neural dynamics by analyzing variations in multi-scale entropy. Full article

Surface defect detection serves as one of the crucial auxiliary components in the quality control of varistors, and it faces real challenges such as the scarcity of defect samples, high labelling cost, and insufficient a priori knowledge, which makes unsupervised deep learning-based detection methods attract attention. However, existing unsupervised models have problems such as inaccurate defect localisation and a low recognition rate of subtle defects in the detection results. To solve the above problems, an unsupervised detection method (Var-MNDR) is proposed to reconstruct the normal distribution of surface defects of varistors under mask constraints. Firstly, on the basis of colour space as well as morphology, an image preprocessing method is proposed to extract the main body image of the varistor, and a mask-constrained main body pseudo-anomaly generation strategy is adopted so that the model focuses on the texture distribution of the main body region of the image, reduces the model’s focus on the background region, and improves the defect localisation capability of the model. Secondly, Kolmogorov–Arnold Networks (KANs) are combined with the U-Network (U-Net) to construct a segmentation sub-network, and the Gaussian radial basis function is introduced as the learnable activation function of the KAN to improve the model’s ability to express the image features, so as to realise more accurate defect detection. Finally, by comparing the four unsupervised defect detection methods, the experimental results prove the superiority and generalisation of the proposed method. Full article

Angle crack defects significantly affect compressor blade radial deformation characteristics, posing critical challenges for reliability assessment under operational uncertainties. This study proposes a novel osprey optimization algorithm (OOA)-optimized Kriging and radial basis function (RBF) method (OOA-KR) for the efficient reliability evaluation of blade radial clearance with angle crack defects. The approach integrates Kriging’s uncertainty quantification capabilities with RBF neural networks’ nonlinear mapping strengths through an adaptive weighting scheme optimized by OOA. Multiple uncertainty sources including crack geometry, operational temperature, and loading conditions are systematically considered. A comprehensive finite element model incorporating crack size variations and multi-physics coupling effects generates training data for surrogate model construction. Comparative studies demonstrate superior prediction accuracy with RMSE = 0.568 and R² = 0.8842, significantly outperforming conventional methods while maintaining computational efficiency. Reliability assessment achieves 97.6% precision through Monte Carlo simulation. Sensitivity analysis reveals rotational speed as the most influential factor (S = 0.42), followed by temperature and loading parameters. The proposed OOA-KR method provides an effective tool for blade design optimization and reliability-based maintenance strategies. Full article

Aiming at the synchronization problem of fractional time-varying perturbation systems, an improved WRBF neural network was proposed based on the wavelet function and radial basis function (RBF). Then, the adaptive controller and updated law are derived based on the WRBF network. It is used to approximate functions and adjust the corresponding parameters in the controller. Based on Lyapunov and Barbalat stability theory, the synchronization of a fractional system with time-varying perturbation is proved effectively. Full article

As the main load-bearing components of engineering structures, regular health assessment of reinforced concrete (RC) columns is crucial for improving the service life and overall performance of the structures. This study focuses on the health detection problem of in-service RC columns. By combining deep learning algorithms and acoustic emission (AE) technology, the AE sources of in-service RC columns are located, and the optimal sensor layout form for the health monitoring of in-service RC columns is determined. The results show that the data cleaning method based on the k-means clustering algorithm and the voting selection concept can significantly improve the data quality. By comparing the localization performance of the Back Propagation (BP), Radial Basis Function (RBF) and Support Vector Regression (SVR) models, it is found that compared with the RBF and SVR models, the MAE of the BP model is reduced by 7.513 mm and 6.326 mm, the RMSE is reduced by 9.225 mm and 8.781 mm, and the R² is increased by 0.059 and 0.056, respectively. The BP model has achieved good results in AE source localization of in-service RC columns. By comparing different sensor layout schemes, it is found that the linear arrangement scheme is more effective for the damage location of shallow concrete matrix, while the hybrid linear-volumetric arrangement scheme is better for the damage location of deep concrete matrix. The hybrid linear-volumetric arrangement scheme can simultaneously detect damage signals from both shallow and deep concrete matrix, which has certain application value for the health monitoring of in-service RC columns. Full article

Accurate cost estimation during the conceptual and feasibility phase of highway projects is essential for informed decision making by public contracting authorities. Existing approaches often rely on pavement cross-section descriptors, general project classifications, or quantity estimates of major work categories that are not reliably available at the early planning stage, while focusing on one or more key asset categories such as roadworks, bridges or tunnels. This study makes a novel contribution to both scientific literature and practice by proposing the first early-stage highway construction cost estimation model that explicitly incorporates roadworks, interchanges, tunnels and bridges, using only readily available or easily derived geometric characteristics. A comprehensive and practical approach was adopted by developing and comparing models across multiple machine learning (ML) methods, including Multilayer Perceptron-Artificial Neural Network (MLP-ANN), Radial Basis Function-Artificial Neural Network (RBF-ANN), Multiple Linear Regression (MLR), Random Forests (RF), Support Vector Regression (SVR), XGBoost Technique, and K-Nearest Neighbors (KNN). Results demonstrate that the MLR model based on six independent variables—mainline length, service road length, number of interchanges, total area of structures, tunnel length, and number of culverts—consistently outperformed more complex alternatives. The full MLR model, including its coefficients and standardized parameters, is provided, enabling direct replication and immediate use by contracting authorities, hence supporting more informed decisions on project funding and procurement. Full article

A new methodology is introduced that incorporates advanced higher moment evaluation in a new approach to the Portfolio Selection problem, supported by effective Computational Intelligence models. The Portfolio Intelligence (PI) model extracts hidden patterns from numerous accounting data and financial statements, filtering misleading effects such as noise or fraud offering an optimal portfolio selection method. The chaotic reflections of speculative behaviors of investors are analyzed in fractal distributions, as higher moments with fundamentals clear the turbulence of noise while the PI model, under its robust AI classifiers, provides optimal investment support. Full article

The construction industry is characterized by complexity and high risk, making effective risk management essential for project success. Traditional risk management methods, which often rely on expert judgment and historical data, are increasingly inadequate for addressing modern construction projects’ dynamic and multifaceted challenges. This study systematically reviews applications of artificial neural networks (ANNs) in construction risk management, covering studies published between 1990 and 2024. Following PRISMA 2020 guidelines, an initial TITLE-ABSTRACT-KEYWORD search in Scopus (1990–2024) yielded 4648 records. After applying subject area and publication-type filters, 2483 records remained. Following duplicate removal, title and abstract screening reduced the pool to 132. After a full-text eligibility assessment, 86 studies were retained. Two additional studies were identified through co-citation analysis, and after the exclusion of four retracted papers, 84 studies were included in the final synthesis. Relevant peer-reviewed studies were categorized to evaluate ANN models, their applications, and key findings. The results indicate that ANNs, including backpropagation and radial basis function networks, have been applied effectively in cost estimation, schedule prediction, safety assessment, and quality control tasks. They offer advantages compared with conventional approaches, such as improved pattern recognition, faster data processing, and more accurate risk evaluation. At the same time, critical challenges persist, including data quality, computational demands, and the interpretability of outputs. To address these issues, studies increasingly recommend integrating ANNs with hybrid approaches such as fuzzy logic, genetic algorithms, and Monte Carlo simulations, as well as leveraging real-time data through IoT and BIM frameworks. This review contributes to theory and practice by consolidating fragmented evidence, distinguishing theoretical and practical contributions, and offering practical recommendations for industry adoption. It also highlights future research directions, particularly the integration of hybrid models, explainable AI, and real-time data environments. Full article

Differentiation of various crops in small-scale crops is important for food security and economic development in many rural communities. Despite being the oldest and simplest classification technique, thresholding continues to gain popularity for classifying complex images. This study aimed to evaluate the effectiveness of a multilevel thresholding technique in differentiating various crop types in small-scale farms. Three (3) types of crops were identified in the study area, and these were cabbage, maize, and sugar bean. Analytical Spectral Devices (ASD) spectral reflectance data were used to detect subtle differences in the spectral reflectance of crops. Analysis of ASD reflectance data revealed reflectance disparities among the surveyed crops in the Green, red, near-infrared (NIR), and shortwave infrared (SWIR) wavelengths. The ASD reflectance data in the Green, red, and NIR were then used to define thresholds for different crop types. The multilevel thresholding technique was used to classify the surveyed crops on the unmanned aerial vehicle (UAV) imagery, using the defined thresholds as input. Three (3) other machine learning classification techniques were also used to offer a baseline for evaluating the performance of the MLT approach, and these were the multilayer perceptron (MLP) neural network, radial basis function neural network (RBFNN), and the Kohonen’s self-organizing maps (SOM). An analysis of crop cover patterns revealed variations in crop area cover as predicted by the MLT and selected machine learning techniques. The classification results of the surveyed crops revealed the area covered by cabbage crops to be 7.46%, 6.01%, 10.33%, 7.05%, 9.48%, and 7.04% as predicted by the MLT on Blue band, MLT on Green band, MLT on NIR, MLP, RBFNN, and SOM, respectively. The area covered by maize crops as predicted by the MLT on Blue band, MLT on Green band, MLT on NIR, MLP, RBFNN, and SOM were noted to be 13.62%, 26.41%, 12.12%, 11.03%, 12.19% and 15.11%, respectively. Sugar bean was noted to occupy 57.51%, 43.72%, 26.77%, 27.44%, 24.15%, and 16.33% as predicted by the MLT on Blue band, MLT on Green band, MLT on NIR, MLP, RBFNN, and SOM, respectively. Accuracy assessment results generally showed poor crop pattern prediction with all tested classifiers in categorizing the surveyed crops, with the kappa index of agreement (KIA) values of 0.372, 0.307, 0.488, 0.531, 0.616, and 0.659 for the MLT on Blue band, MLT on Green band, MLT on NIR, MLP, RBFNN, and Kohonen’s SOM, respectively. Despite recommendations by recent studies, we noted that the MLT was noted to be unsuitable for classifying complex features such as spectrally overlapping crops. Full article

Background/Objectives: We recently discovered that tumors rely on blood fatty acids as an energy source for growth. Therefore, we investigated biomarkers in the lipid fractions of plasma from patients with pancreatic ductal adenocarcinoma (PDAC) for the screening diagnosis of PDAC. Methods: We screened common fatty acid types in human (normal 99, PDAC 103) and mouse (normal 7, KPC 22) plasma samples using a non-targeted approach. Subsequently, we identified targets in human plasma (set A: normal 68, PC 102) that could distinguish between healthy individuals and patients with cancer. Next, we verified whether the identified targets were useful in a new human set (set B: 96 normal, 78 PC). We combined sets A and B to create set C and further divided it into a training set (7:3 ratio; normal 115, pancreatic cancer 126) and a validation set (normal 49, PC 54). The identified targets were used to train three statistical models (logistic regression (LR), random forest (RF), and support vector machine (SVM) with a radial basis function (RBF) kernel). Results: The comparison of human and mouse plasma identified eight common lipid metabolites. We further identified four platforms containing these metabolites for target analysis: acylcarnitines, phospholipids, fatty acid amides, and sphingolipids. We analyzed the four platforms using sets A, B, and C and found 20 lipids (1 acylcarnitine, 1 sphingolipid, and 18 phospholipids) that met the criterion of AUC ≥ 0.75 in all three sets. Based on an average AUC for LR models with 11 or more phospholipids, the separation performance between healthy individuals and patients with cancer was 0.9207 (sensitivity, 90.74%; specificity, 86.22%; PPV, 87.90%; NPV, 89.42%), and the AUC of the validation set for CA19-9 in the same groups was 0.7354. The addition of CA19-9 to the LR models resulted in a separation performance of 0.9427 (90.74%; 88.01%; 89.32%; 89.61%) for the validation set. Conclusions: We identified 18 candidate fatty acid metabolites that could serve as biological markers in the serum lipid fractions of pancreatic cancer patients and confirmed that all of them decreased in patients. Additionally, we developed an algorithm utilizing these markers, which demonstrated a 25% increase in discriminatory power compared to the AUC value of CA19-9, an FDA-approved biomarker for pancreatic cancer. In summary, we identified candidate metabolites and algorithms that could serve as biomarkers in the lipid fractions of plasma from patients with pancreatic cancer. Full article

A simple Lagrangian travelling slice model has been successfully used to predict the relations between the process parameters and the strand temperatures in the continuous casting of steel. The present paper aims to include a simple macrosegregation, grain structure and mechanical stress and deformation model on top of the thermal slice framework. The basis of all the mentioned models is the slice heat-conduction model that considers the complex heat extraction mechanisms in the mould, with the sprays, rolls, and through radiation. Its main advantage is the fast calculation time, which is suitable for the online control of the caster. The macroscopic thermal and species transfer models are based on the continuum mixture theory. The macrosegregation model is based on the lever rule microsegregation model. The thermal conductivity and species diffusivity of the liquid phase are artificially enhanced to consider the convection of the melt. The grain structure model is based on cellular automata and phase-field concepts. The calculated thermal field is used to estimate the thermal contraction of the solid shell, which, in combination with the metallostatic pressure, drives the elastic-viscoplastic solid-mechanics models. The solution procedure of all the models is based on the meshless radial basis function generated finite difference method on the macroscopic scale and the meshless point automata concept on the grain structure scale. Simulation results point out the areas susceptible to hot tearing. Full article

Parameter uncertainties and fluctuating disturbances have posed significant challenges to the smooth and precise control of Flexible Joint Robots (FJRs) in industrial environments. To mitigate such disturbances, Disturbance Observers (DOBs) are commonly employed; however, the model uncertainties inherent in FJR systems make accurate dynamic modeling challenging, and the efficacy of DOBs hinges heavily on the accuracy of the dynamic model, which limits their applicability to FJR control. This paper presents a hybrid RBFNN-based Disturbance Observer (RBFNNDOB) state feedback controller for FJRs. By combining a nominal model-based DOB with an RBFNN, this method effectively addresses the unknown dynamics of FJRs while simultaneously compensating for external time-varying disturbances. In this framework, an adaptive neural network weight update law is formulated using Lyapunov stability theory. This enables the RBFNN to selectively estimate the unmodeled uncertainties in FJR dynamics, thereby minimizing computational redundancy in model estimation while allowing dynamic compensation for residual uncertainties beyond the nominal model and DOB estimation errors—ultimately enhancing computational efficiency and achieving robust compensation for rapidly changing disturbances. The boundedness of the tracking error is proven using the Lyapunov approach, and experimental validation is conducted on the FJR system to confirm the efficacy of the proposed control method. Full article

In this paper, an adaptive sliding mode fault-tolerant control (FTC) scheme is proposed for small Quadrotor Unmanned Aerial Vehicles (QUAVs) used in education and teaching formation in the presence of systematic unknown external disturbances with actuator failures. A radial basis function neural network (RBFNN) is employed to handle the nonlinear interaction function, and a fault-tolerant-based NN extended state observer (NNESO) is designed to estimate the unknown external disturbance. Meanwhile, an adaptive fault observer is developed to estimate and compensate for the fault parameters of the system. To achieve satisfactory trajectory tracking performance for the QUAV, an adaptive sliding mode control (SMC) strategy is designed. This strategy mitigates the strong coupling effects among the design parameters within the QUAV formation. The stability of the closed-loop system is rigorously demonstrated by Lyapunov analysis, and the controlled QUAV formation can achieve the desired tracking position. Simulation results verify the effectiveness of the proposed control method. Full article