Rheological behavior characteristics of TiO2-MWCNT/10w40 hybrid nano-oil affected by temperature,concentration and shear rate: An experimental study and a neural network simulating

In this article, rheological behavior of TiO2-MWCNT (45–55%)/10w40 hybrid nano-oil was studied experimentally. The nano- oils were tested at temperature ranges of 5–55 °C and in shear rates up to 11,997 s⁻¹. With respect to viscosity, shear stress and shear rate variations it was cleared that either of the base oil and nano-oil were non-Newtonian fluids. New equations which were based on thickness of the fluid were presented for different temperature values, R-squared values were between 0.9221 and 0.9998 (the precise of correlation changes depend on temperature). Also to predict the nano-oil behavior, neural network method was utilized. an artificial neural network (MLP type) were used to predict the viscosity in terms of temperature, solid volume fraction and shear stress. to compare the prediction precise of neural network and correlation the results of these two were compared with together. ANN showed more accurate results in comparison with correlation results. R² and (MSE) were 0.9979 and 0.000016 respectively for the ANN.     

An artificial neural networks approach to model and predict the relationship between the grounding resistance and length of buried electrode in the soil

This paper presents a technique based on the development of an artificial neural network (ANN) model for modeling and predicting the relationship between the grounding resistance and length of an electrode buried in the soil based on experimental data. The results indicate the strong agreement between the model prediction and experimental values. The statistical analysis shows that the R² values were 0.995 and 0.925 for the training and testing sets, respectively.     

Comparison of ANN and DoE for the prediction of laser-machined micro-channel dimensions

This paper presents four models developed for the prediction of the width and depth dimensions of CO₂ laser-formed micro-channels in glass. A 3³ statistical design of experiments (DoE) model was built and conducted with the power (P), pulse repetition frequency (PRF), and traverse speed (U) of the laser machine as the selected parameters for investigation. Three feed-forward, back-propagation artificial neural networks (ANNs) models were also generated. These ANN models were varied to investigate the influence of variations in the number and the selection of training data. Model A was constructed with 24 data randomly selected from the experimental results, leaving three data points for model testing; Model B was constructed with the eight corner points of the experimental data space, and seven other randomly selected data, leaving 12 data points for testing; and Model C was constructed with 15 randomly selected data leaving 12 data points for testing. These models were developed separately for both micro-channel width and depth prediction. These ANN models were constructed in LabVIEW coding. The performance of these ANN models and the DoE model were compared. When compared with the actual results two of the ANN models showed greater average percentage error than the DoE model. The other ANN model showed an improved predictive capability that was approximately twice as good as that provided from the DoE model.     

Integrating artificial neural networks and empirical correlations for the prediction of water-subcooled critical heat flux

The critical heat flux (CHF) is an important parameter for the design of nuclear reactors, heat exchangers and other boiling heat transfer units. Recently, the CHF in water-subcooled flow boiling at high mass flux and subcooling has been thoroughly studied in relation to the cooling of high-heat-flux components in thermonuclear fusion reactors. Due to the specific thermal-hydraulic situation, very few of the existing correlations, originally developed for operating conditions typical of pressurized water reactors, are able to provide consistent predictions of water-subcooled-flow-boiling CHF at high heat fluxes. Therefore, alternative predicting techniques are being investigated. Among these, artificial neural networks (ANN) have the advantage of not requiring a formal model structure to fit the experimental data; however, their main drawbacks are the loss of model transparency (‘black-box’ character) and the lack of any indicator for evaluating the accuracy and reliability of the ANN answer when ‘never-seen’ patterns are presented. In the present work, the prediction of CHF is approached by a hybrid system which couples a heuristic correlation with a neural network. The ANN role is to predict a datum-dependent parameter required by the analytical correlation; this parameter was instead set to a constant value obtained by usual best-fitting techniques when a pure analytical approach was adopted. Upper and lower boundaries can be possibly assigned to the parameter value, thus avoiding the case of unexpected and unpredictable answer failure. The present approach maintains the advantage of the analytical model analysis, and it partially overcomes the ‘black-box’ character typical of the straight application of ANNs because the neural network role is limited to the correlation tuning. The proposed methodology allows us to achieve accurate results and it is likely to be suitable for thermal-hydraulic and heat transfer data processing.     

A novel enzyme-assisted ultrasonic approach for highly efficient extraction of resveratrol from Polygonum cuspidatum

Resveratrol is a promising multi-biofunctional phytochemical, which is abundant in Polygonum cuspidatum. Several methods for resveratrol extraction have been reported, while they often take a long extraction time accompanying with poor extraction yield. In this study, a novel enzyme-assisted ultrasonic approach for highly efficient extraction of resveratrol from P. cuspidatum was developed. According to results, the resveratrol yield significantly increased after glycosidases (Pectinex® or Viscozyme®) were applied in the process of extraction, and better extraction efficacy was found in the Pectinex®-assisted extraction compared to Viscozyme®-assisted extraction. Following, a 5-level-4-factor central composite rotatable design with response surface methodology (RSM) and artificial neural network (ANN) was selected to model and optimize the Pectinex®-assisted ultrasonic extraction. Based on the coefficient of determination (R²) calculated from the design data, ANN model displayed much more accurate in data fitting as compared to RSM model. The optimum conditions for the extraction determined by ANN model were substrate concentration of 5%, acoustic power of 150 W, pH of 5.4, temperature of 55 °C, the ratio of enzyme to substrate of 3950 polygalacturonase units (PGNU)/g of P. cuspidatum, and reaction time of 5 h, which can lead to a significantly high resveratrol yield of 11.88 mg/g.     

Effect of self-absorption correction on LIBS measurements by calibration curve and artificial neural network

In this paper, laser-induced breakdown spectroscopy (LIBS) technique is used for concentration prediction of six elements of Mn, Si, Cu, Fe, Zn, and Mg in seven Al samples by two approaches of artificial neural network (ANN) and standard calibration curve. ANN is utilized as a new technique for determination and classification of various materials and elements in LIBS method. In this study, a few spectra of six aluminum standards with known concentrations are used for training of ANN. It should be noted that the mentioned network is not on trial and error basis, but it is a self-organized network. Calibration curve method, which is implemented in represented paper, determines certain relation between concentration and intensity. Then, the calibration curve and ANN methods obtained by six samples are used for prediction of the elements of the seventh standard sample in order to check the accuracy of these methods and make a comparison. In both approaches, a self-absorption correction is applied for high concentrations species and an improvement in prediction of two methods is seen. Results illustrate that at high concentrations except for Si, ANN shows a better prediction with a lower relative error compared to calibration curve approach after self-absorption correction. Primitive study without any self-absorption correction shows that ANN and calibration curve predictions with the best result are related to Fe with R ² = 0.99 % having the minimum errors.     

Employing response surface methodology and neural network to accurately model thermal conductivity of TiO2–water nanofluid using experimental data

In this research, the thermal conductivity of the H₂O–titania nanofluid is modeled versus the particle concentration and temperature via the Artificial Neural Network (ANN) and Response Surface Methodology (RSM). The experimental data include six particle concentrations and five temperatures from 30 to 70 °C. The thermal conductivity augments by the increment in nanoparticle concentration and temperature, such that the maximum thermal conductivity increment happens at the highest temperature and nanoparticle concentration (i.e., T = 70 °C and φ = 1%). It is observed that the impact of temperature on the thermal conductivity is more noticeable than the influence of particle concentration, however, the thermal conductivity demonstrates a more non-linear trend versus nanoparticle volume fraction compared with the temperature. The best structure of the neural network has 2 hidden layers with 2 and 4 neurons, respectively in the 1^st and 2^nd hidden layers. The results show that the prediction precision of the ANN correlation is better than that of the RSM correlation.     

Dual-modality and dual-energy gamma ray densitometry of petroleum products using an artificial neural network

The prediction of volume fractions in order to measure the multiphase flow rate is a very important issue and is the key parameter of multi-phase flow meters (MPFMs). Currently, the gamma ray attenuation technique is known as one of the most precise methods for obtaining volume fractions. The gamma ray attenuation technique is based on the mass attenuation coefficient, which is sensitive to density changes; density is sensitive in turn to temperature and pressure fluctuations. Therefore, MPFM efficiency depends strongly on environmental conditions. The conventional solution to this problem is the periodical recalibration of MPFMs, which is a demanding task. In this study, a method based on dual-modality densitometry and artificial intelligence (AI) is presented, which offers the advantage of the measurement of the oil–gas–water volume fractions independent of density changes. For this purpose, several experiments were carried out and used to validate simulated dual modality densitometry results. The reference density point was established at a temperature of 20 °C and pressure of 1 bar. To cover the full range of likely density fluctuations, four additional density sets were defined (at changes of ±4% and ±8% from the reference point). An annular regime with different percentages of oil, gas and water at different densities was simulated. Four features were extracted from the transmission and scattered detectors and were applied to the artificial neural network (ANN) as inputs. The input parameters included the ²⁴¹Am full energy peak, ¹³⁷Cs Compton edge, ¹³⁷Cs full energy peak and total scattered count, and the outputs were the oil and air percentages. A multi-layer perceptron (MLP) neural network was used to predict the volume fraction independent of the oil and water density changes. The obtained results show that the proposed ANN model achieved good agreement with the real data, with an estimated root mean square error (RMSE) of less than 3.     

An approximation to the cross sections of Z l boson production at CLIC by using neural networks

In this work, the possible dynamics associated with leptophilic Z _l boson at CLIC (Compact Linear Collider) have been investigated by using artificial neural networks (ANNs). These hypotetic massive boson Z _l have been shown through the process e ⁺ e ^?→µ⁺µ^?. Furthermore, the invariant mass distributions for final muons have been consistently predicted by using ANN. For these highly non-linear data, we have constructed consistent empirical physical formulas (EPFs) by appropriate feed-forward ANN. These ANNEPFs can be used to derive further physical functions which could be relevant to studying Z _l .     

Wooden windows: Sound insulation evaluation by means of artificial neural networks

Windows are the weakest part of a façade in terms of acoustic performance: the weighted sound insulation index (R_w), measured according to ISO 140-3, is the fundamental parameter to evaluate the façade acoustic insulation.The paper aims at developing an artificial neural network (ANN) model to estimate the R_w value of wooden windows based on a limited number of windows parameters; this is a new approach because acoustic phenomena are non-linear and affected by a plurality of factors and, therefore, usually investigated through experimentation.Data set is taken from experimental campaigns carried out at the Laboratory of Acoustics, University of Perugia. A multilayer feed-forward approach was chosen and the model was implemented in MATLAB. On the basis of the results obtained by means of a preliminary training and test campaign of several ANN architectures, five main parameters were selected as network inputs: window typology, frame and shutters thickness, number of gaskets, R_w of glazing; R_w value of the window is the network output. Different ANN configurations were trained and a root mean-square error less than 3% was obtained, comparable to measurement uncertainty.This approach allows to develop a model which, with input parameters varying within appropriate ranges, can easily estimate the acoustic performance of wooden windows without experimental campaign on prototypes, saving both money and time. If the training data set is large enough, the presented approach could be very useful for design and optimization of acoustic performance of new products.     

Correlation of volumetric properties of binary mixtures of some ionic liquids with alcohols using equation of state

In our previous paper, we extended the Tao and Mason equation of state (TM EOS) to pure ionic liquids. Here we apply TM EOS based on statistical?Cmechanical perturbation theory to binary mixtures of ionic liquids. Three temperature-dependent quantities are needed to use the equation of state: the second virial coefficient, B ₂, effective van der Waals co-volume, b, and a scaling factor, ??. The second virial coefficients are calculated from a correlation that uses the normal boiling temperature and normal boiling density. ?? and b can also be calculated from the second virial coefficient by scaling. In this procedure, the number of input parameters, for calculation of B ₂, ??, and b reduced from 5 (i.e., critical temperature, critical pressure, acetric factor, Boyle temperature T _B, and the Boyle volume ?? _B) to 2 (i.e., T _bp and ?? _bp). At close inspection of the deviations given in this work, the TM EOS predicts the densities with a mean AAD of 1.69%. The density of selected system obtained from the TM EOS has been compared with those calculated from perturbed-hard-sphere equation of state. Our results are in favor of the preference of the TM EOS over another equation of state. The overall average absolute deviation for 428 data points that calculated by perturbed-hard-sphere equation of state is 2.60%.     

色貌模型的人工神经网络方法的研究

色貌模型(CAM)主要解决不同观察条件、不同背景和不同环境下的颜色真实再现问题。采用人工神经网络(ANN)的方法来实现目前最新的色貌模型CIECAM02的预测,包括正向预测(从色度参数到色貌属性参数)和逆向预测(从色貌属性参数到色度参数),应用自然色系统(NCS)中的部分色样作为神经网络的训练和测试样本。由于正向输出色貌属性参数空间不是均匀的,对于网络预测精度用特殊方法评估,而对于逆向模型则可直接利用LAB色差公式评价。测试的结果表明:用神经网络对CIECAM02模型的预测达到了较高的精度。     

Artificial neural network (ANN) approach for modeling and formulation of phenol adsorption onto activated carbon

In this study, a three-layer feed-forward back propagation network with Levenberg-Marquardt (LM) learning algorithm was applied to predict adsorption of phenol onto activated carbon (AC). Batch experiments were carried out to obtain experimental data. The neural network was trained considering the amount of adsorbent, initial concentration of phenol, temperature, contact time and pH as input parameters and the final concentration of phenol as a desired parameter. Different transfer functions for hidden and output layers and different number of neurons in a hidden layer were tested to optimize the network structure. An empirical equation for final concentration of phenol was developed by using the weights of optimized network. Accuracy of the developed ANN model was also measured using statistical parameters, such as mean absolute error (MAE), mean square error (MSE), root mean square error (RMSE) and correlation coefficient (R²). Results showed that MAE, MSE, RMSE, and R² values of the ANN model were 0.1540, 0.0565, 0.2378, and 0.9998, respectively, which indicate high accuracy of the ANN model. In the equilibrium study, predicted results of the ANN model were also compared with experimental data and the results of other conventional isotherm models.     

人工神经网络用于有机环境污染紫外光谱库检索

本文将人工神经网络（ＡＮＮ）用于有机环境污染物紫外光谱库检索。对该神经网络的参数优化作了讨论。并用ＡＮＮ对噪声、杂质等因素的影响作了详细的考察。为了提高紫外光谱的分辨，本文提出用光谱作ＡＮＮ训练和检索，使网络的收敛速度明显加快，对检验光谱中杂质的容允程度明显增加。本文还将ＡＮＮ与传统的相关系数法作了比较。结果表明，ＡＮＮ法在抗噪声和杂质等方面明显优于相关系数法。     

人工神经网络用于有机环境污染物紫外光谱库检索

本文将人工神经网络（ＡＮＮ）用于有机环境污染物紫外光谱库检索。对该神经网络的参数优化作了讨论。并用ＡＮＮ对噪声、杂质等因素的影响作了详细的考察。为了提高紫外光谱的分辨，本文提出用导数光谱作ＡＮＮ训练和检索，使网络的收敛速度明显加快，对检验光谱中杂质的容允程度明显增加。本文还将ＡＮＮ与传统的相关系数法作了比较。结果表明，ＡＮＮ法在抗噪声和杂质等方面明显优于相关系数法。     

Noninvertibility and resonance in discrete-time neural networks for time-series processing

We present a computer-assisted study emphasizing certain elements of the dynamics of artificial neural networks (ANNs) used for discrete time-series processing and nonlinear system identification. The structure of the network gives rise to the possibility of multiple inverses of a phase point backward in time; this is not possible for the continuous-time system from which the time series are obtained. Using a two-dimensional illustrative model in an oscillatory regime, we study here the interaction of attractors predicted by the discrete-time ANN model (invariant circles and periodic points locked on them) with critical curves. These curves constitute a generalization of critical points for maps of the interval (in the sense of Julia-Fatou); their interaction with the model-predicted attractors plays a crucial role in the organization of the bifurcation structure and ultimately in determining the dynamic behavior predicted by the neural network.     

An a priori evaluation of a principal component and artificial neural network based combustion model in diesel engine conditions

A principal component analysis (PCA) and artificial neural network (ANN) based chemistry tabulation approach is presented. ANNs are used to map the thermochemical state onto a low-dimensional manifold consisting of five control variables that have been identified using PCA. Three canonical configurations are considered to train the PCA-ANN model: a series of homogeneous reactors, a nonpremixed flamelet, and a two-dimensional lifted flame. The performance of the model in predicting the thermochemical manifold of a spatially-developing turbulent jet flame in diesel engine thermochemical conditions is a priori evaluated using direct numerical simulation (DNS) data. The PCA-ANN approach is compared with a conventional tabulation approach (tabulation using ad hoc defined control variables and linear interpolation). The PCA-ANN model provides higher accuracy and requires several orders of magnitude less memory. These observations indicate that the PCA-ANN model is superior for chemistry tabulation, especially for modelling complex chemistries that present multiple combustion modes as observed in diesel combustion. The performance of the PCA-ANN model is then compared to the optimal estimator, i.e. the conditional mean from the DNS. The results indicate that the PCA-ANN model gives high prediction accuracy, comparable to the optimal estimator, especially for major species and the thermophysical properties. Higher errors are observed for the minor species and reaction rate predictions when compared to the optimal estimator. It is shown that the prediction of minor species and reaction rates can be improved by using training data that exhibits a variation of parameters as observed in the turbulent flame. The output of the ANN is analysed to assess mass conservation. It is observed that the ANN incurs a mean absolute error of 0.05% in mass conservation. Furthermore, it is demonstrated that this error can be reduced by modifying the cost function of the ANN to penalise for deviation from mass conservation.     

Electron/pion identification in the CBM TRD applying a ω<Stack><Subscript><Emphasis Type="Italic">n</Emphasis></Subscript><Superscript><Emphasis Type="Italic">k</Emphasis></Superscript></Stack> goodness-of-fit criterion

The problem of electron/pion identification in the CBM experiment based on the measurements of energy losses and transition radiation in the TRD detector is discussed. Earlier we analyzed a possibility to solve such a problem using an artificial neural network (ANN) [1]. Here we consider an approach based on a nonparametric ω _n ^k goodness-of-fit criterion, and comparison with the ANN method is also performed. We show that both methods provide a comparable level of pion suppression and electron identification, the ω _n ^k test is more simple for practical applications, the ANN method provides the needed level of pions suppression only if “clever” variables are used. We demonstrate that application of the ω _n ^k -criterion to the J/Ψ reconstruction provides a high level of pion background suppression and significantly improves a signal-to-background ratio. The text was submitted by the authors in English.