Search

The prediction of rock fragmentation (Fr) is highly beneficial to the optimization of blasting operations in the mining industry. The characteristics of the rock mass, the blast geometry, and the explosive qualities are the primary elements influencing Fr. The methodical explosion of explosives within a rock mass results in the production of smaller rock pieces. This work is a step toward the prediction of the degree of Fr in opencast mining using advanced soft computing (SC) methods like: back propagation neural network (BPNN), k-nearest neighbor (KNN), multilayer perceptron (MLP), radial basis function (RBF), multi-variable regression (MVR), gene expression programming (GEP), Takagi-Sugeno fuzzy model (TSF), least-square-support vector machine (LS-SVM), and support vector machine (SVM). A dataset consisting of 219 blasting events with 10 influencing parameters: hole diameter (HDM), spacing (S), burden (B), maximum charge per delay (MCPD), stemming (ST), compressive strength (CS), powder factor (PF), specific drilling (SPD), number of holes (NH), and bench height (BH), were used in the present study. All models were assessed with the help of following performance parameters: RRSE, RSE, NRMSE, RRMSE, MAD, MAPE, MSE, RMSE, and R2. Based on loss function for Fr, scattered diagram, importance ranking, sensitivity analysis, rank analysis, and violin plot the top models were chosen. From the obtained results, it is seen that SVM produce better result compare to other models when predicting the Fr of rock. Under sensitivity analysis, spider diagrams and Tomado diagrams were plotted to determine the variation of input and output factors. The sensitivity analysis of the developed model shows that HDM has the least impact, whereas the parameters B and PF have the maximum impact on the Fr of rock. [ABSTRACT FROM AUTHOR]

With the rapid development of wearable technology, smart contact lenses (SCL) are gradually gaining attention as a breakthrough innovation. The emergence of these products suggests that smart glasses, which incorporate electronic components and visual aids, are expected to become the mainstream of human-computer interaction in the future. However, realizing this vision requires not only advanced electronics but also highly sophisticated manufacturing processes. Therefore, this paper provides an in-depth discussion on the process of manufacturing smart contact lenses using in-mold electronic decoration technology and focuses on the multi-objective problem of optimizing injection parameters such as melt temperature and holding pressure to achieve on micro-molecular displacements as well as residual stresses. First, the background and technical requirements of smart contact lenses are described in detail, emphasizing the prospect of SCL for a wide range of applications in augmented reality, healthcare, and smart assistance. Subsequently, the key role of IME technology in SCL manufacturing is discussed. Focusing on the optimization of melting temperature, holding pressure and holding time, the effects of these three key parameters on eyewear were systematically analyzed with the goal of improving the overall performance and biocompatibility of SCL. The multi-objective optimization of melting temperature and holding pressure was achieved by NSGA-III. Radial basis function interpolation was used as an auxiliary method to provide finer optimization results for NSGA-III. During the multi-objective optimization process, efforts were made to achieve uniform flow of melt temperature and optimal adjustment of holding pressure to maximize the transparency, stability and comfort of SCL. The final results obtained achieved an optimization rate of 95.60% and 93.47% for nodal displacement and residual stress, respectively, compared with the initially recommended process parameters. [ABSTRACT FROM AUTHOR]

When solar irradiation is uniform along with the array, the P–V curve represents a unique maximum power point (MPP). If the cells undergo shade conditions in the presence of bypass diodes, the solar array's power is decreased, and the P–V curve of the array represents multiple local MPPs (LMPP) and a global MPP (GMPP). LMPPs might mislead the maximum power point tracking (MPPT) algorithms because their characteristics are identical to the MPP. Various studies have been conducted on partial shading conditions. This study uses parallel distributed maximum power point tracking (DMPPT) due to the advantages of this structure. A high-gain converter is presented to resolve the high conversion gain required by the DC/DC converter in this structure. This study also presents MLP and RBF networks for MPP tracking and compares their efficiencies under the same irradiation and partial shading conditions. Since determining optimal weight coefficients in MLP neural networks and variances, means, and weights in RBF networks play an essential role in their performance, this study uses four optimization algorithms of particle swarm optimization (PSO), gray wolf optimization (GWO), grasshopper optimization algorithm (GOA), and Harris Hawks optimization (HHO). Finally, an adaptive fuzzy-PID controller controls the three-phase grid-connected inverter. A comparison of the results shows that the efficiency of MLP and RBFII is almost the same, about 98–99%. Moreover, the accuracy of MLP networks is higher than RBF, and RBF networks' only advantage is shorter training time. In addition, RBF networks require much more activation functions for proper performance. The simulation outcomes confirm the superior efficacy of the HHO algorithm in training neural networks when compared to alternative algorithms. [ABSTRACT FROM AUTHOR]

Accurate monitoring of atmospheric water vapor content is essential for the early warning of extreme weather events. As known, GNSS zenith troposphere delay (GNSS_ZTD) is an indispensable data source for retrieving precipitable water vapor (PWV). However, the newest GPT3 empirical model is not accurate enough to perform the ZTD (GPT3_ZTD) and PWV (GPT3_PWV) estimation in some regions, such as Japan. Thus, here, we introduce a radial basis function (RBF) neural network to establish ZTD forecast models based on the GPT3 model and use the predicted ZTD to retrieve PWV and adopt the retrieved PWV in forecasting precipitation. To thoroughly verify the accuracy of forecast results in 2021, we selected three external validation data: GNSS, radiosonde, and meteorological data. The GNSS_ZTD validation results show that the error compensation model of GPT3 based on RBF is superior to the GPT3 model and the model using a single RBF and back propagation (BP) neural network. The average RMSE of all GNSS stations is 50.7 mm, 53.7 mm, and 37.8 mm for GPT3_ZTD, RBF_BP_ZTD, and RBF_GPT3_ZTD, respectively. The GNSS_PWV and RO_PWV validation results show that the retrieved PWV with compensation of RBF_GPT3_ZTD is better than the uncompensated GPT3_ZTD, the average accuracy of RBF_GPT3_PWV of GNSS stations and radiosonde stations is improved by 40.4% and 25.8% against that of GPT3_PWV. For the precipitation forecast model results, the average forecast accuracy of all GNSS stations and radiosonde stations is 63.1% and 61.4%, according to the ERA5 precipitation. The average forecast accuracy is 66.3%, validated by meteorological precipitation records. The proposed model not only improves the GPT3 model but also forecasts the PWV value, which can improve the precipitation forecast in Japan, and is expected to expand to other regions. [ABSTRACT FROM AUTHOR]

The definition of an appropriate number of Radial Basis Functions and their parameters in Radial Basis Function networks is a non-trivial task. The fitting of its parameters has direct implications on model performance and generalization. Techniques such as cross-validation associated with error metrics, which frequently rely on iterative optimization, have been used to address this problem, requiring significant computational effort and uncertain solutions. We propose a method for determining the number and parameters of Radial Basis Functions based on the Dominating Set of the Gabriel graph, which represents the structure of the input data, such that no exterior/prior parameter estimation or heuristic methods are necessary. [ABSTRACT FROM AUTHOR]

The technological breakthrough and the availability of multispectral remote sensing data have given rise to an ambitious challenge for the classification of the multispectral images accurately to support administrative bodies in decision-making. In this paper, the multi-temporal medium resolution Sentinel-2 imagery of the densely populated urban area of Delhi-NCR is classified using SVM into five different land cover classes, namely water bodies, barren land, vegetative region, road network, and residential areas. Further, the effect of different kernel functions of SVM on land cover classification performance is contrasted and the radial basis function (RBF) leads to the best results. The experimental results are compared with the maximum likelihood classification (MLC) method on different evaluation metrics. The SVM with RBF kernel shows promising improvements in the overall accuracy by 10% relative to the polynomial kernel and by 3% compared to MLC. The analysis of multitemporal spectral imagery of the study area reflects the increase in a built-up area (road network, Buildings), water bodies, and decrement in the area of barren land and vegetation. [ABSTRACT FROM AUTHOR]

In this paper, we study the spectral properties of the periodized Radial Basis Function interpolation matrix as well as the related harmonic operators discretized using Radial Basis Functions. For Gaussian RBF, this procedure could be easily extended to an arbitrarily high dimensional space on a tensor-product grid as presented in the later parts of the paper. The experimental result of Boyd's condition number [1] is analytically well predicted in the context of periodized RBF. [ABSTRACT FROM AUTHOR]

The decanoic acid-palmitic acid composite phase change material compounds with SiO2 and TiO2 to prepare decanoic acid-palmitic acid/SiO2@TiO2 phase change microcapsules (D-P-SiO2@TiO2 PCM). The D-P-SiO2@TiO2 PCM could show efficient temperature regulation, remove pollutants through photocatalysis, and control air humidity. However, it is difficult to obtain the best experimental scheme directly using the traditional experimental setup due to the complicated photocatalytic-humidity performance. The radial basis function (RBF) model optimized the uniform experimental design parameters, and the D-P-SiO2@TiO2 PCM showed enhanced photocatalytic-humidity performance. The RBF-calculated preparation parameters were as follows: the molar ratio of decanoic acid-palmitic acid to tetraethyl silicate was 0.42, pH was 1.83, the molar ratio of deionized water to tetraethyl silicate was 98.15, while the molar rate of tetrabutyl titanate to tetraethyl silicate was 0.76. The degradation rate of gaseous formaldehyde by the RBF-optimized D-P-SiO2@TiO2 PCM was 69.57% after 6 h, and the moisture content was between 0.0923 and 0.0940 g·g−1 at 43.16–75.29% relative humidity (RH). The comparison between model optimization and the experiment sample prepared using the optimized parameters showed that the theoretical photocatalytic-humidity performance target value was 2.0502, and the tested target value was 2.0757. The error calculated from these two values was only 1.24%, and both were better than the best value of uniform experimental calculation. RBF mathematical model was proved to be an effective, convenient, and economic-saving method to simulate and predict D-P-SiO2@TiO2 PCM experimental design parameters. SEM and TEM analyses of the RBF-optimized D-P-SiO2@TiO2 PCM showed a uniform spherical structure, and the particle size analysis analyses was about 200 nm. The DSC analysis showed the phase transition temperature range was between 16.97 and 28.94 °C, within the comfort range of the human body. The UV–Vis investigations showed the absorption edge of the RBF-optimized D-P-SiO2@TiO2 PCM was 380 nm, in line with the band gap structure of the TiO2 anatase phase. The thermogravimetric investigations showed that this composite was stable at normal temperature and pressure. After a 100 times hot–cold cycle, the quality of the RBF-optimized D-P-SiO2@TiO2 PCM maintained its stability, as the photocatalytic-humidity performance was almost the same. The N2-adsorption analysis showed it had a high specific surface area and irregular pore structures, which could help it regulate air humidity. Considering these results, the D-P-SiO2@TiO2 PCM, a new ecological functional material, would be used in the construction industry to improve the architectural ecological environment. [ABSTRACT FROM AUTHOR]

Fruit growth patterns are often exploited in predictions of final fruit size and to inform planting and harvesting decisions. Ten local apricot (Prunus armeniaca) varieties with superior genotypes (two early-ripening, five mid-ripening and three late-ripening varieties) were assessed using 20 nonlinear regression models (NRMs) and a radial basis function (RBF) neural network model. Fruit diameter and weight measurements for each genotype were collected at four-day intervals from fruit set to commercial harvest. Patterns based on diameter and weights were attributed to each genotype. Among the NRM tested, only four were able to flawlessly predict apricot diameter and weight during the growing season. In addition, comparison of nonlinear regression methods with the neural network indicated than the RBF model displayed fewer prediction errors than the NRMs. The RBF model predicted fruit size with a coefficient of determination (R2 value) greater than 0.95. Therefore, predictions of growth patterns in fruit trees can be accomplished with neural network modeling. [ABSTRACT FROM AUTHOR]

In this study, an enhancement to the Multiquadric Radial Basis Function (MQ-RBF) method is presented for solving Elasticity Boundary Value Problems (E-BVPs). MQ-RBF is utilized to reproduce the system of governing partial differential equations and boundary conditions. Also the inefficacies of some existing methods in determining the Optimal Shape Parameter (OSP) when solving E-BVPs are shown. Afterward, a new algorithm is introduced based on Teaching–Learning-Based Optimization (TLBO). In addition, the lower and upper bounds of the shape parameter and a new objective function were introduced for the TLBO algorithm. Using the algorithm, one OSP is obtained for both MQ-RBF functions in 2D problems, resulting in reduced computational cost. Its efficiency is evidenced by solving several examples. Moreover, its accuracy is demonstrated by comparing its outcomes to those of analytical methods. Unlike the 1D problems, the OSP is found to be a function of loading and material properties as well as the geometry in 2D problems. [ABSTRACT FROM AUTHOR]

PV systems are widely installed and the building block to the solar system is the solar cell itself. In this paper, a new idea was presented, where Artificial Neural Network (ANN) was used to extract the five parameters of a solar module using only the basic voltage and current parameters. In addition, a large number of PV modules datasheet information was used to train an ANN to obtain the five parameters of any new module. In order to train the neural network (NN), particle swarm optimization technique was employed to obtain the input and output data set. Instead of five parameters of a solar module, only four parameters were used for the optimization technique and remaining was found from the direct equation.A comparison of Multilayer Perceptron and Radial Basis Function Neural Networks (NN) performance was provided. The proposed technique for solar cell parameters extraction was found robust, and accurate. Unlike parameter extraction for a specific panel, the proposed approach is a generalized approach that can extract the five parameters for any panel regardless its manufacturer. Using the mean square error and the squared error as measurement tools of the robustness of the proposed models, it has been found that the developed models are very accurate and robust. [ABSTRACT FROM AUTHOR]

The low-pressure pipe network inside buildings without effective leak detection means has become the most vulnerable link in the urban gas infrastructure network. Once the low-pressure gas pipeline fails and causes leakage, it is more likely to cause explosion and structural collapse. Considering the characteristics of low-pressure gas pipeline inside the building, the main difficulty in leakage identification comes from the pressure fluctuations caused by the gas usage in the pipeline. Based on steady-state calculation, we analyzed the steady-state hydraulic properties of the leakage and gas usage conditions. When the flow rates of the two conditions are close, the leakage fluctuation cannot be effectively identified based on the steady-state characteristics. We used Flowmaster to conduct transient hydraulic analysis under two conditions and found that the transient barotropic reflection wave caused by burner structure can be used as the basis for leakage detection. We built a simulation test system of low-pressure gas system for data acquisition and theoretical verification. On this basis, we proposed a set of intelligent low-pressure gas pipeline leak detection methods. The pressure signal was converted into a dynamic pressure signal, which can fully express the transient characteristics of pressure fluctuation and reduce the influence of steady-state characteristics. Wavelet packet and radial basis function neural network can intelligently learn the characteristics of dynamic pressure waves. When the leakage flow is the same as that of gas usage, the accuracy of leak detection can still be stabilized above 82.5%. [ABSTRACT FROM AUTHOR]

River Ganga is one of the most significant rivers in the country. This river is the adobe for numerous aquatic species and microorganisms. The color of the river suddenly changed to green due to the rise of algal bloom in the Varanasi and nearby regions of the river Ganga during May–June 2021. These algal blooms can be detrimental to the aquatic animals of the river. This study analyzes the occurrence and the possible reasons for the algal bloom generation in the river for the considered stretch. Several factors like nutrient accumulation in the river through agricultural run-off, warm river temperature, low flow condition of the river, thermal stratification, and less turbid river water can be considered as possible reasons for algal bloom development. In this work, the optical remote sensing-based Sentinel 2 datasets have been used for the duration of mid-May 2021 to mid-June 2021. These datasets have been processed in the Google Earth Engine (GEE) platform, and chlorophyll concentration has been calculated using different satellite-based indices or band ratios. The chlorophyll concentration measurements have quantified the algal bloom growth. These indices or band ratios have been analyzed using several artificial neural network (ANN) architectures like multilayer perceptron (MLP) and radial basis function (RBF) along with the in situ values. It has been found that chlorophyll concentration has been highest for the mid-June 2021 time period in the considered river stretch. [ABSTRACT FROM AUTHOR]

Multifidelity surrogates (MFSs) replace computationally intensive models by synergistically combining information from different fidelity data with a significant improvement in modeling efficiency. In this paper, a modified MFS (MMFS) model based on a radial basis function (RBF) is proposed, in which two fidelities of information can be analyzed by adaptively obtaining the scale factor. In the MMFS, an RBF was employed to establish the low-fidelity model. The correlation matrix of the high-fidelity samples and corresponding low-fidelity responses were integrated into an expansion matrix to determine the scaling function parameters. The shape parameters of the basis function were optimized by minimizing the leave-one-out cross-validation error of the high-fidelity sample points. The performance of the MMFS was compared with those of other MFS models (MFS-RBF and cooperative RBF) and single-fidelity RBF using four benchmark test functions, by which the impacts of different high-fidelity sample sizes on the prediction accuracy were also analyzed. The sensitivity of the MMFS model to the randomness of the design of experiments (DoE) was investigated by repeating sampling plans with 20 different DoEs. Stress analysis of the steel plate is presented to highlight the prediction ability of the proposed MMFS model. This research proposes a new multifidelity modeling method that can fully use two fidelity sample sets, rapidly calculate model parameters, and exhibit good prediction accuracy and robustness. [ABSTRACT FROM AUTHOR]

The titanium dioxide (TiO2) photocatalyst reactor is a widely used approach for eliminating non-biodegradable compounds in water and wastewater. Predicting the performance of the reactor in removing acid red 14 (AR14) was investigated using radial basis function (RBF), Adaptive Neuro-Fuzzy Inference System (ANFIS), and fuzzy regression analysis. The input variables in the models were radiation, irradiation time, the initial concentration of nanoparticle TiO2 to cement ratio (nTiO2/c), pH, initial concentration of AR14, flow rate, and initial concentration of oxidizing peroxydisulfate. In addition, for reducing the overfitting of the training process, we used K-fold cross-validation. The results of using the three methods indicated that the ANFIS performance was more suitable in comparison with the fuzzy regression models, such that the coefficient of determination (R2), index of agreement (IA), Nash–Sutcliffe efficiency (E) for model efficiency, Mean Squared Error (MSE), and Mean Bias Error (MBE) for training between the observed data and predicted data reached 0.965, 0.991, 0.73, 0.0132 and 0.019, respectively. Normalized input data improved slightly the training processes and performance of RBF compared with non-normalized input data. The ANFIS neural network for training process and model performance slightly achieved better results than the RBF neural network. The performance of the fuzzy regression approach was less suitable in comparison with RBF and ANFIS modeling. The sensitivity analysis of the RBF neural network depicted that the initial concentration of peroxydisulfate (S2O8) was the most effective factor (25.6%) in predicting the removal of AR14 from synthetic wastewater. After that, the effective factors were irradiation time (12.44%), initial concentration of AR14 (10.37%), pH (10.37), flow rate (8.55%), and the weight ratio of the initial concentration of nanoparticle TiO2 to cement ratio (nTiO2/c) (1.75%), respectively. [ABSTRACT FROM AUTHOR]

It is essential to study the behaviour of soot particles in the engine cylinder, including movement and physical properties during engine operation, for further analysis of performance characteristics and engine emissions. This research aims to determine the size distribution of soot particles near the cylinder wall, leading to soot in lubricating oil. The research was carried out numerically using in-house MATLAB routines to post-process CFD simulation results. In this analysis, the soot particle size was calculated using the Hiroyasu soot formation, Nagle & Strickland-Constable oxidation, and Wersborg soot coagulation models. In addition, the simulation time and computing effort were minimized by using a post-processed soot particle size estimate. According to the findings, soot particles smaller than 50 nm were found in the squish area, especially near the cylinder wall. The average value of the final location of the particles tracked from all zones in the trajectory analysis varies between 0.4 and 3.1% of the magnitude closer to the cylinder wall relative to the particles tracked without the thermophoresis factor. [ABSTRACT FROM AUTHOR]

Reducing the disturbance of parameter fluctuations to the system is very important to the robustness of stamping product quality. In this paper, a 6-sigma robust optimization design method based on residual error and radial basis interpolation is proposed to weaken the negative influence of parameter fluctuations. Not only the key controllable parameters but also the key noise parameters are considered by grey relational analysis (GRA) to screen the parameters that have an important influence on the stamping quality. In order to reflect the relationship between key parameters and forming quality more precisely, the polynomial response surface model is improved based on residual error and radial basis function (RBF), and solved by the multi-population co-evolutionary immune algorithm (MPCEIA). Moreover, the solution of deterministic optimization is designed with 6-sigma robust optimization to improve the robustness of the system to the parameter fluctuations. Finally, the method is applied to the design of a complex automobile cover part, proving its effectiveness and reliability. [ABSTRACT FROM AUTHOR]

In this paper, a new fault diagnosis and fault tolerant control algorithm for manipulators with actuator multiplicative fault is proposed. The dynamic model of the manipulator with disturbance is taken as the research object. When faults occur in the actuator, a nonlinear observer based on radial basis function (RBF) neural network is used to estimate the fault information. After the fault information is obtained, an adaptive back-stepping sliding mode controller is used to control the manipulator to reach the desired trajectory. At last, an illustrated example is given to demonstrate the efficiency of the proposed algorithm, and satisfactory results have been obtained. [ABSTRACT FROM AUTHOR]

Purpose: Establishing the spatial distribution of soil properties in agro-ecosystems is critical to increasing the productivity of intensive tillage soils and reducing environmental pressure. Thus, a comprehensive understanding of soil's heavy metal distribution characteristics is important for optimal management. The goal of this study was to determine the most accurate soil Cd interpolation method for the study area and analyze the variation and spatial distribution of Cd content with depth. Materials and methods: In this work, we employed two common spatial interpolation methods—inverse distance weighting (IDW) and radial basis function (RBF)—in conjunction with a geographic information system (GIS) to interpolate sampling data of the heavy metal cadmium (Cd) throughout Santai County; a locale in the lower part of Fujiang River, Mianyang, in the northeast of Sichuan Province. Results and discussion: The results of our investigation are as follows. (1) At depths of 0–50 cm, the Cd content decreased with depth while the coefficient of variation (Cv) increased with depth. The exponential model and the Gauss model adequately described the Cd content's spatial autocorrelation and the determination coefficient (R2) varied within 0.747–0.962, indicating that the model had high precision. (2) With respect to the accuracy of the IDW and RBF methods, IDW was more accurate than the RBF in the 0–20 cm soil layer, yet the reverse was found in the 20–50 cm soil layer. (3) Three-dimensional spatial distribution maps—constructed using the RBF and IDW methods at different soil depths—showed that the RBF method provided a greater smoothing effect than the IDW method. Conclusions: The spatial variability of soil Cd content shows that: (1) the Cd content's spatial autocorrelation was strong in the 0–50 cm soil layer. (2) The IDW method is suitable for areas that are greatly influenced by human or farming activities while the RBF method is appropriate for areas that are greatly influenced by internal soil factors. The RBF is superior for evaluating the spatial distribution of soil Cd content in this area. [ABSTRACT FROM AUTHOR]

In this work, a radial basis function (RBF)-based numerical scheme is developed which uses the Coimbra variable time fractional derivative of order 0 < α (t , x) < 1 . The Coimbra derivative can efficiently model a dynamical system whose fractional-order behavior changes with time and space locations. The RBF can effectively approximate spatial derivatives in multi-dimensions. The resulted numerical scheme is RBF–FD type and is validated for solute transport problems in 1D and 2D dimension domains. Various cases of variable-order 0 < α (t , x) < 1 have been discussed. The present numerical scheme can effectively approximate those variable-order models whose exact solution can not be obtained in a simple way. [ABSTRACT FROM AUTHOR]

The medical digital twin is emerging as a viable opportunity to provide patient-specific information useful for treatment, prevention and surgical planning. A bottleneck toward its effective use when computational fluid dynamics (CFD) techniques and tools are adopted for the high fidelity prediction of blood flow, is the significant computing cost required. Reduced order models (ROM) looks to be a promising solution for facing the aforementioned limit. In fact, once ROM data processing is accomplished, the consumption stage can be performed outside the computer-aided engineering software adopted for simulation and, in addition, it could be also implemented on interactive software visualization interfaces that are commonly employed in the medical context. In this paper we demonstrate the soundness of such a concept by numerically investigating the effect of the bulge shape for the ascending thoracic aorta aneurysm case. Radial basis functions (RBF) based mesh morphing enables the implementation of a parametric shape, which is used to build up the ROM framework and data. The final result is an inspection tool capable to visualize, interactively and almost in real-time, the effect of shape parameters on the entire flow field. The approach is first verified considering a morphing action representing the progression from an average healthy patient to an average aneurismatic one (Capellini et al. in Proceedings VII Meeting Italian Chapter of the European Society of Biomechanics (ESB-ITA 2017), 2017; Capellini et al. in J. Biomech. Eng. 140(11):111007-1–111007-10, 2018). Then, a set of shape parameters, suitable to consistently represent a widespread number of possible bulge configurations, are defined and accordingly generated. The concept is showcased taking into account the steady flow field at systolic peak conditions, using ANSYS®Fluent®and its ROM environment for CFD and ROM calculations respectively, and the RBF MorphTM software for shape parametrization. [ABSTRACT FROM AUTHOR]

Sediment yield is important for maintaining soil health, reservoir sustainability, environmental pollution, and conservation of natural resources. The main aim of the present work is to develop four machine learning models, artificial neural networks (ANNs), radial basis function (RBF), support vector machine (SVM) and multiple model (MM)-ANNs for forecasting daily sediment yield. These models were applied to the Shakkar and Manot watersheds covering 25 years (1990–2015) and 10 years (2000–2010) of rainfall and discharge data, respectively. Results showed that the MM-ANNs model satisfactorily predicted sediment yield and outperformed the other models providing the highest correlation coefficient (0.921, 0.883) and Nash-Sutcliffe efficiency (0.744, 0.763) and the lowest relative absolute error (0.360, 0.344) and root mean square error (23,609.5, 269,671.5) for the Shakkar and Manot during the test period, respectively. Hence, the MM-ANNs model can be successfully used for sediment prediction. [ABSTRACT FROM AUTHOR]

Surrogate model assisted evolutionary algorithms (SAEAs) are metamodel-based strategies usually employed on the optimization of problems that demand a high computational cost to be evaluated. SAEAs employ metamodels, like Kriging and radial basis function (RBF), to speed up convergence towards good quality solutions and to reduce the number of function evaluations. However, investigations concerning the influence of metamodels in SAEAs performance have not been developed yet. In this context, this paper performs an investigative study on commonly adopted metamodels to compare the ordinary Kriging (OK), first-order universal Kriging (UK1), second-order universal Kriging (UK2), blind Kriging (BK) and RBF metamodels performance when embedded into a single-objective SAEA Framework (SAEA/F). The results obtained suggest that the OK metamodel presents a slightly better improvement than the others, although it does not present statistically significant difference in relation to UK1, UK2, and BK. The RBF showed the lowest computational cost, but the worst performance. However, this worse performance is around 2% in relation to the other metamodels. Furthermore, the results show that BK presents the highest computational cost without any significant improvement in solution quality when compared to OK, UK1, and UK2. [ABSTRACT FROM AUTHOR]

Forest management should be directed towards multifunctional management and utilization of forest services (other than wood production) in order to achieve maximum utilization and minimum degradation. Artificial intelligence enables forest managers to plan for utilization of forest landscape aesthetic values. Visual quality evaluation is a stochastic problem in natural forest landscapes and it is influenced by forest characteristics. We aimed to landscape visual quality evaluation by expert/human-perception-based approach and application of artificial intelligence modeling techniques for the visual quality prediction of forest landscapes. Therefore, we recorded five landscape attributes in 100 forest landscapes. We developed the stochastic model to evaluate visual quality potential by artificial intelligence techniques. Comparing to multi-layer regression (R2 = 0.588) and multi-layer perceptron (R2 = 0.847), the radial basis function (RBF) (R2 = 0.887) model represents the highest value of R2 in the test data set. The water, shrubs, roads, rocky hills, and trees, in forest landscapes were introduced respectively as the most important attributes which influence the RBF model. The designed graphical user interface tool, as an environmental decision support system, evaluates landscape visual quality of forests, and it helps to solve stochastic problems such as visual quality value. [ABSTRACT FROM AUTHOR]

Accurate modeling of the rainfall-runoff process is still a challenging job despite the availability of various modeling methods, such as data-driven or knowledge-driven, developed by various researchers in their previous studies. Among these models, artificial neural network (ANN)-based rainfall-runoff models play an important role in the hydrology due to their capability of reproducing the highly nonlinear nature between the various factors involved in the hydrology of the watershed. In this paper, an attempt has been made to develop an ANN-based rainfall-runoff model for the Hoshangabad catchment of the Narmada River in Madhya Pradesh. Two different models, feed-forward back propagation (FFBP) and radial basis function (RBF) network models, were developed using several arrangements of input dataset then relate their capability of flow estimation for the period 2004–2013. The best model performance was selected based on various performance evaluation criteria, i.e., R2, MSE, and AARE. Based on this study, it is observed that the ANN model provides better outcomes for the dataset scaled between zero and one. For the Hoshangabad catchment, an input arrangement of present-day rainfall with antecedent rainfall up to 4 days and a 1-day antecedent runoff value provide the best outcomes for both models (FFBP and RBF). Out of this, the RBF network performs better as compared with the FFBP network with R2 value of 0.9964. The result of the present study suggests that ANN network models an essential tool for predicting the hydrological responses in agricultural watersheds and thus helps to provide sustainable measures for a watershed. [ABSTRACT FROM AUTHOR]

Combined weir-gate structure is one of the important structures which are control the water level, measure discharge and avoid sediment deposition behind the weir. In this study, first try to simulate four combined triangular weir-rectangular gates with different geometric conditions via 3D numerical software (Flow-3D) by using experimental data. Then, dimensionless analyses were done to find the non-dimension parameters affected the discharge coefficient of this structure. At the end, four different intelligent system models were used to estimate the discharge coefficient, evaluated and compared the results with each other. Results show that the Flow-3D software has a high capability to simulate the flow over the combined structure. Besides, the values of goodness of fit criteria show that the numerical solver, estimate the water head and discharge coefficient very well. Moreover, in all models, the results show that the discharge coefficient has an inverse relation with dimensionless parameters (h/b, h/d and h/y) and discharge coefficients in this study are between 0.3–0.9. On the other hand, results of accuracy analyses of four intelligent system models of MLP, RBF, GRNN and M5P show that the MLP model is the superior model in this study, and after that, the rest of models' sort as M5P, RBF and GRNN in this study. [ABSTRACT FROM AUTHOR]

A radial basis function generated finite-differences (RBF-FD) inspired technique for evaluating definite integrals over the volume of the ball in three dimensions is described. Such methods are necessary in many areas of applied mathematics, mathematical physics and myriad other application areas. Previous approaches needed restrictive uniformity in the node set, which the algorithm presented here does not require. By using RBF-FD approach, the proposed algorithm computes quadrature weights for N arbitrarily scattered nodes in only O (N log N) operations with high orders of accuracy. [ABSTRACT FROM AUTHOR]

The gas–oil and petrochemical industry plays one of the most powerful roles in the growth of the worldwide industrial and economic sectors. Huge amounts of investment are focused on these projects through various delivery methods. The majority of these projects are facing time delay challenges, cost overruns and quality weaknesses worldwide. Generally, the influence factors on project productivity performance would be subjected to changes in projects. In this research, the contribution of design change effects that imposed by frequency and severity on cost overrun and schedule delay was studied through a case study. The aim is to provide and introduce an artificial neural network through radial basis function (RBF) model to increase the level of perception and overall project performance during an early stages of project (design phase). The purpose is to create an insight evaluation on the level of impact of changes to the project during the design phase as an advanced supporting tool for decision makers. The output of RBF neural network will present the quantified measurement of all interacted change causes by means of cost and time expenditures. Experimental results show that the proposed prediction neural network is in good agreement with case observations. Consequently, this model of predominantly influencing factors will lead decision makers to deliver more optimized decisions with less complexity through wider insight on design change events. [ABSTRACT FROM AUTHOR]

Rockfall hazard is a very common phenomenon mainly occurring in mountainous slopes, coastal cliffs, volcanoes, riverside, and trenches. In the present research, a model is developed for hazard analysis of seismic rockfalls on a regional scale. For this purpose, three models including Adaptive Neuro-Fuzzy Inference System (ANFIS), multilayer perceptron artificial neural network (MLP), and radial basis function artificial neural network (RBF) were utilized. Firooz Abad-Kojour earthquake of 2004 was used as the benchmark and the model base. The rockfall-susceptible zones predicted by ANFIS, MLP, and RBF methods were compared with the database (distribution map) of seismic rockfalls. The results showed a good overlap between MLP-predicted rockfall hazard zones and database (distribution map) of seismic rockfalls. To evaluate the statistical results of ANFIS, MLP, and RBF models, the verification parameters with high accuracy such as density ratio, quality sum (Qs), and Receiver Operating Characteristic Curve were employed. By analyzing the hazard maps and considering the Qs index obtained by ANFIS (26.76) and MLP (49.19), and RBF (13.84), it could be observed that the calculated Qs of MLP were higher than those of ANFIS and RBF. Moreover, based on the obtained value of the area under the curve from ANFIS (0.984), MLP (0.986), and RBF methods (0.884), it is seen that the MLP network, compared to ANFIS and RBF models, provided a higher accuracy in hazard analysis of rockfalls caused by the earthquake of Firooz Abad-Kojour of 2004. [ABSTRACT FROM AUTHOR]

Vehicular ad hoc network (VANET), a special type of mobile ad-hoc networks (MANETs), is characterized by a very high mobility. Due to the dynamic nature of vehicle nodes and unstable wireless links, the design of an efficient and effective routing scheme is serious challenging for VANETs. Thus, proposing a reliable routing protocol for important data dissemination is essential for VANETs. In this paper, a clustering-based routing protocol is introduced where nodes are clustered according to movement information such as node degree and velocity of vehicles by means of imperialist competitive algorithm. Then, the cluster head is selected by means of radial basis function neural network algorithm according to amount of free buffer space and expected transmission count. Simulation results indicated that the proposed scheme improves the other related algorithms in terms of the parameters of packet delivery rate, throughput and end-to-end delay. [ABSTRACT FROM AUTHOR]

Time-domain analysis of high-frequency field-effect transistors is presented using the meshfree radial point interpolation method. The distributed modeling approach is followed in the linear and nonlinear regimes. The corresponding matrix Telegrapher's equation and terminal conditions are discretized using the weighted average method. Conditionally and unconditionally stable schemes are studied. For linear and conditionally stable nonlinear analysis, the leap-frog and Crank–Nicolson methods are used. To provide unconditional stability in the nonlinear regime, a hybrid backward/forward scheme is exploited. [ABSTRACT FROM AUTHOR]

Multicarrier code division multiple access (MC-CDMA) is a novel wireless communication technology with high spectral efficiency and system performance. However, all multiple access techniques including MC-CDMA were most likely to have multiple access interference (MAI). So, this paper mainly aims at designing a suitable receiver for MC-CDMA system to mitigate such MAI. The classical receivers like maximal-ratio combining and minimum mean square error fail to cancel MAI when the MC-CDMA is subjected to nonlinear distortions, which may occur due to saturated power amplifiers or arbitrary channel conditions. Being highly nonlinear structures, the neural network (NN) receivers such as multilayer perceptron and radial basis function networks could be better alternative for such a case. The possibility NN receiver for a MC-CDMA system under different nonlinear conditions has been studied with respect to both performance and complexity analysis. [ABSTRACT FROM AUTHOR]

Evaluating behavior of earth dams under dynamic loads is one of the most important problems associated with the initial design of such massive structures. This study focuses on prediction of deformation of earth dams due to earthquake shaking. A total number of 103 real cases of deformation in earth dams due to earthquakes that has occurred over the past years were gathered and analyzed. Using soft computing methods, including feed-forward back-propagation and radial basis function based neural networks, two models were developed to predict slope deformations in earth dams under variant earthquake shaking. Earthquake magnitude (Mw), yield acceleration ratio (ay/amax), and fundamental period ratio (Td/Tp) were considered as the most important factors contributing to the level of deformation in earth dams. Subsequently, a sensitivity analysis was conducted to assess the performance of the proposed model under various conditions. Finally, the accuracy of the developed soft computing model was compared with the conventional relationships and models to estimate seismic deformations of earth dams. The results demonstrate that the developed neural model can provide accurate predictions in comparison to the available practical charts and recommendations. [ABSTRACT FROM AUTHOR]

Heart is one of the essential operating organs of the human body and its failure is a major contributing factor toward the human deaths. Coronary heart disease may be asymptotic but can be anticipated through the medical tests and daily life routine of the subject. Diagnosis of the coronary heart disease needs a specialized medical resource with the plenty of experience. All over the world and particularly in the developing countries, there is a lack of such experts which make the diagnosis more difficult. In this paper, we present a clinical heart disease diagnostic system by proposing feature subset selection methodology with an object of achieving improved performance. The proposed methodology presents three algorithms for selecting candidate feature subsets: (1) mean Fisher score-based feature selection algorithm, (2) forward feature selection algorithm and (3) reverse feature selection algorithm. Feature subset selection algorithm is presented to select the most decisive subset from the candidate feature subsets. The features are added to the feature subsets on the basis of their individual Fisher scores, while the selection of a feature subset depends on its Matthews correlation coefficient score and dimension. The selected feature subset with the reduced dimension is fed to the RBF kernel-based SVM which results in binary classification: (1) heart disease patient and (2) normal control subject. The proposed methodology is validated through accuracy, specificity and sensitivity using four UCI datasets, i.e., Cleveland, Switzerland, Hungarian and SPECTF. The statistical results achieved using the proposed technique are shown in comparison with the existing techniques reflecting its better performance. It has an accuracy of 81.19, 84.52, 92.68 and 82.7% for Cleveland, Hungarian, Switzerland and SPECTF, respectively. [ABSTRACT FROM AUTHOR]

Curvature is an important geometric property in computer graphics that provides information about the character of object surfaces. The exact curvature can only be calculated for a limited set of surface descriptions. Most of the time, we deal with triangles, point sets or some other discrete representation of the surface. For those, curvature can only be estimated. However, surfaces can be fitted by some kind of interpolation function and from it, curvature can be calculated directly. This paper proposes a method for curvature estimation and normal vector re-estimation based on surface fitting using Hermite Radial Basis Function interpolation. Hermite variation uses not only control points, but normal vectors at those points as well. This leads to a better and more robust interpolation than if only control points are used. Once the interpolant is obtained, the curvature and other possible properties can be directly computed using known approaches. The proposed algorithm was tested on several explicit and implicit functions, and it outperforms current state-of-the-art methods if exact normals are available. For normals calculated directly from a triangle mesh, the proposed algorithm works on par with existing state-of-the-art methods. [ABSTRACT FROM AUTHOR]

In order to predict the landslide disaster effectively, the prediction system of landslide disaster based on intelligent algorithm is constructed. Lianziya and Gushuwu are used as the research object. The applicability of RBF and BP algorithm in landslide deformation displacement prediction is analyzed and compared. Based on the cumulative displacement curve of the landslide deformation, the deformation of the landslide is divided into three main stages: initial deformation, uniform deformation and acceleration deformation. When the classical intelligent algorithm BP is used to predict the landslide deformation, the selection of network structure parameters has a great influence on the prediction results. By optimizing the parameters, the optimal network structure can be constructed, and better prediction accuracy can be obtained. The results show that the improved algorithms overcome the defects of the standard BP algorithm to some extent. The accuracy of prediction is improved. LMBP has the best prediction effect. The RBF has more advantages than the LM-BP algorithm for predicting landslide deformation, which is in good agreement with the landslide curve. [ABSTRACT FROM AUTHOR]

Multiple models of a physical phenomenon are sometimes available with different levels of approximation. The high fidelity model is more computationally demanding than the coarse approximation. In this context, including information from the lower fidelity model to build a surrogate model is desirable. Here, the study focuses on the design of a miniaturized photoacoustic gas sensor which involves two numerical models. First, a multifidelity metamodeling method based on Radial Basis Function, the co-RBF, is proposed. This surrogate model is compared with the classical co-kriging method on two analytical benchmarks and on the photoacoustic gas sensor. Then an extension to the multifidelity framework of an already existing RBF-based optimization algorithm is applied to optimize the sensor efficiency. The co-RBF method does not bring better results than co-kriging but can be considered as an alternative for multifidelity metamodeling. [ABSTRACT FROM AUTHOR]

By coupling the low-fidelity (LF) model with the high-fidelity (HF) samples, the variable-fidelity model (VFM) offers an efficient way to overcome the expensive computing challenge in multidisciplinary design optimization (MDO). In this paper, a cooperative radial basis function (Co-RBF) method for the VFM is proposed by modifying the basis function of RBF. The RBF method is constructed on the HF samples, while the Co-RBF method incorporates the entire information of the LF model with the HF samples. In Co-RBF, the LF model is regard as a basis function of Co-RBF and the HF samples are utilized to compute the Co-RBF model coefficients. Two numerical functions and three engineering problems are adopted to verify the proposed Co-RBF method. The predictive results of Co-RBF are compared with those of RBF and Co-Kriging, which show that the Co-RBF method improves the efficiency, accuracy and robustness of the existing VFMs. [ABSTRACT FROM AUTHOR]

Radial basis function link neural network (RBFLN) and fuzzy-weights of evidence (fuzzy-WofE) methods were used to assess regional-scale prospectivity for chromite deposits in the Western Limb and the Nietverdiend layered mafic intrusion of the Bushveld Complex in South Africa. Five predictor maps derived from geological, geochemical and geophysical data were processed in a GIS environment and used as spatial proxy for critical processes that were most probably responsible for the formation of the chromite deposits in the study area. The RBFLN was trained using input feature vectors that correspond to known deposits, prospects and non-deposits. The training was initiated by varying the number of radial basis functions (RBFs) and iterations. The results of training the RBFLN provided optimum number of RBFs and iterations that were used for classification of the input feature vectors. The results show that the network classified 73% of the validation deposits into highly prospective areas for chromite deposit, covering 6.5% of the study area. The RBFLN entirely classified all the non-deposit validation points into low prospectivity areas, occupying 86.6% of the study area. In general, the efficiency of the RBFLN in classifying the validation deposits and non-deposits indicates the degree of spatial relationship between the input feature vectors and the training points, which represent chrome mines and prospects. The RBFLN and fuzzy-WofE analyses used in this study are important in guiding identification of regional-scale prospect areas where further chromite exploration can be carried out. [ABSTRACT FROM AUTHOR]

For high-speed trains, high precision of train positioning is important to guarantee train safety and operational efficiency. By analyzing the operational data of Beijing-Shanghai high-speed railway, we find that the currently used average speed model (ASM) is not good enough as the relative error is about 2.5 %. To reduce the positioning error, we respectively establish three models for calculating train positions by advanced neural computing methods, including back-propagation (BP), radial basis function (RBF) and adaptive network-based fuzzy inference system (ANFIS). Furthermore, six indices are defined to evaluate the performance of the three established models. Compared with ASM, the positioning error can be reduced by about 50 % by neural computing models. Then, to increase the robustness of neural computing models and real-time response, online learning methods are developed to update the parameters in the last layer of neural computing models by the gradient descent method. With the online learning methods, the positioning error of neural computing models can be further reduced by about 10 %. Among the three models, the ANFIS model is the best in both training and testing. The BP model is better than the RBF model in training, but worse in testing. In a word, the three models can reduce the half number of transponders to save the cost under the same positioning error or reduce the positioning error about 50 % in the case of the same number of transponders. [ABSTRACT FROM AUTHOR]

Accurate measurements of PM concentration over time and space are especially critical for reducing adverse health outcomes. However, sparsely stationary monitoring sites considerably hinder the ability to effectively characterize observed concentrations. Utilizing data on meteorological and land-related factors, this study introduces a radial basis function (RBF) neural network method for estimating PM concentrations based on sparse observed inputs. The state of Texas in the USA was selected as the study area. Performance of the RBF models was evaluated by statistic indices including mean square error, mean absolute error, mean relative deviation, and the correlation coefficient. Results show that the annual PM concentrations estimated by the RBF models with meteorological factors and/or land-related factors were markedly closer to the observed concentrations. RBF models with combined meteorological and land-related factors achieved best performance relative to ones with either type of these factors only. It can be concluded that meteorological factors and land-related factors are useful for articulating the variation of PM concentration in a given study area. With these covariate factors, the RBF neural network can effectively estimate PM concentrations with acceptable accuracy under the condition of sparse monitoring stations. The improved accuracy of air concentration estimation would greatly benefit epidemiological and environmental studies in characterizing local air pollution and in helping reduce population exposures for areas with limited availability of air quality data. [ABSTRACT FROM AUTHOR]

Riverbank filtration (RBF) is a process during which river water is subjected to subsurface flow prior to abstraction wells, often characterized by improved water quality. The induced infiltration of river water through the riverbed also creates a clogging layer. This decreases riverbed permeability and abstraction rates, particularly if the river water has high turbidity, as in Thailand. As Chiang Mai Province is one of the most favorable sites for future RBF construction in Thailand, two sites, Mae Rim and San Pa Tong, were selected to simulate clogging by using a channel experiment. The mobile experimental apparatus was set up at the bank of the river in order to use fresh river water. Riverbed sediment was used as channel bed and filling material for the columns. The aim was to simulate riverbed clogging using river water with high turbidity and determine the effect of clogging, which can be quantified using vertical hydraulic conductivity (Kv). An increase in channel flow velocity caused partial removal of a clogging layer in only the top 0.03 m of the sediment column. The combination of low channel flow and high turbidity leads to much more clogging than high channel flow and low turbidity. A complete manual removal of the external clogging layer led to an increase in Kv, but the initial Kv values were not recovered. The external clogging had a lower effect on Kv than internal clogging. For planning new RBF sites along high-turbidity rivers, reduction in Kv to estimate RBF well yield cannot be calculated based only on initial Kv but requires field experiments. [ABSTRACT FROM AUTHOR]