Basis Function Artificial Neural Network Research Articles

Aplicação de redes Neuro Fuzzy ao processamento de peças automotivas por meio de injeção de polímeros

O processamento de peças automotivas por meio de injeção de polímeros envolve vários fenômenos físicos que ocorrem simultaneamente e que possuem caráter não linear e multivariável. Softwares comerciais podem ser utilizados na previsão dos parâmetros do processo, o que pode ser caro e inviável. Pode-se determinar os parâmetros de forma analítica, mas o tratamento desse problema requer a aplicação de teorias clássicas dos fenômenos de transporte, de difícil equacionamento. As redes NeuroFuzzy são aplicáveis a esse problema porque reúnem a capacidade de aprender das redes neurais artificiais com a capacidade da lógica Fuzzy de transformar variáveis linguísticas em regras. Neste trabalho combinou-se uma rede neural artificial Multilayer Perceptron e uma rede neural artificial Radial Basis Function à lógica Fuzzy para construir-se um modelo de inferência que previu o tempo de ciclo de processos de injeção de polímeros. Os resultados obtidos confirmam as redes NeuroFuzzy como opção para esse tipo de problema.

A mixed-integer non-linear programming with surrogate model for optimal remediation design of NAPLs contaminated aquifer

A mixed-integer non-linear programming (MINLP) with surrogate model was introduced to derive the optimal surfactant enhanced aquifer remediation (SEAR) process (remediation cost minimisation and removal rate maximisation) at a nitrobenzene-contaminated site. First, a 3D multi-phase flow simulation model was developed to simulate the SEAR process; using a radial basis function artificial neural network (RBFANN), the surrogate model was built which was an approximation of the simulation model; a MINLP was built to identify the optimal remediation strategies and genetic algorithm (GA) and penalty function were combined to solve the model; at last, the optimal remediation strategies were obtained. The approximation result of RBFANN was compared with that of back-propagation artificial neural network (BPANN), mean absolute error, mean relative error and coefficient of determination of the developed RBFANN model were 0.01, 2.27% and 0.85 respectively, which indicated much higher approximation accuracy than BPANN. The MINLP with surrogate model is a powerful tool for non-aqueous phase liquids (NAPLs) contaminated site remediation optimisation problem and it can greatly improve computational efficiency.

Rapid Identification of Asteraceae Plants with Improved RBF-ANN Classification Models Based on MOS Sensor E-Nose.

Plants from Asteraceae family are widely used as herbal medicines and food ingredients, especially in Asian area. Therefore, authentication and quality control of these different Asteraceae plants are important for ensuring consumers' safety and efficacy. In recent decades, electronic nose (E-nose) has been studied as an alternative approach. In this paper, we aim to develop a novel discriminative model by improving radial basis function artificial neural network (RBF-ANN) classification model. Feature selection algorithms, including principal component analysis (PCA) and BestFirst + CfsSubsetEval (BC), were applied in the improvement of RBF-ANN models. Results illustrate that in the improved RBF-ANN models with lower dimension data classification accuracies (100%) remained the same as in the original model with higher-dimension data. It is the first time to introduce feature selection methods to get valuable information on how to attribute more relevant MOS sensors; namely, in this case, S1, S3, S4, S6, and S7 show better capability to distinguish these Asteraceae plants. This paper also gives insights to further research in this area, for instance, sensor array optimization and performance improvement of classification model.

Analysis of pull-in instability of geometrically nonlinear microbeam using radial basis artificial neural network based on couple stress theory.

The static pull-in instability of beam-type microelectromechanical systems (MEMS) is theoretically investigated. Two engineering cases including cantilever and double cantilever microbeam are considered. Considering the midplane stretching as the source of the nonlinearity in the beam behavior, a nonlinear size-dependent Euler-Bernoulli beam model is used based on a modified couple stress theory, capable of capturing the size effect. By selecting a range of geometric parameters such as beam lengths, width, thickness, gaps, and size effect, we identify the static pull-in instability voltage. A MAPLE package is employed to solve the nonlinear differential governing equations to obtain the static pull-in instability voltage of microbeams. Radial basis function artificial neural network with two functions has been used for modeling the static pull-in instability of microcantilever beam. The network has four inputs of length, width, gap, and the ratio of height to scale parameter of beam as the independent process variables, and the output is static pull-in voltage of microbeam. Numerical data, employed for training the network, and capabilities of the model have been verified in predicting the pull-in instability behavior. The output obtained from neural network model is compared with numerical results, and the amount of relative error has been calculated. Based on this verification error, it is shown that the radial basis function of neural network has the average error of 4.55% in predicting pull-in voltage of cantilever microbeam. Further analysis of pull-in instability of beam under different input conditions has been investigated and comparison results of modeling with numerical considerations shows a good agreement, which also proves the feasibility and effectiveness of the adopted approach. The results reveal significant influences of size effect and geometric parameters on the static pull-in instability voltage of MEMS.

Monthly Electric Energy Consumption Forecasting Using Multiwindow Moving Average and Hybrid Growth Models

Monthly electric energy consumption forecasting is important for electricity production planning and electric power engineering decision making. Multiwindow moving average algorithm is proposed to decompose the monthly electric energy consumption time series into several periodic waves and a long-term approximately exponential increasing trend. Radial basis function (RBF) artificial neural network (ANN) models are used to forecast the extracted periodic waves. A novel hybrid growth model, which includes a constant term, a linear term, and an exponential term, is proposed to forecast the extracted increasing trend. The forecasting results of the monthly electric energy consumption can be obtained by adding the forecasting values of each model. To test the performance by comparison, the proposed and other three models are used to forecast China's monthly electric energy consumption from January 2011 to December 2012. Results show that the proposed model exhibited the best performance in terms of mean absolute percentage error (MAPE) and maximal absolute percentage error (MaxAPE).

Using Radial Basis Function Artificial Neural Network for Predicting Asphalt Content of Asphalt Paving Mixture

The determination of the asphalt content of asphalt paving mixtures is a fundamentally important test for many authorities and researchers in highways field . This study seek to examined the application and use of radial basis function artificial neural network with Gaussian activation function by MATLAB software for predicting the asphalt content of the hot mix asphalt paving mixtures using their properties of Marshall test. The architecture of the study developed network consist of five input nodes representing five properties of Marshall test, with six hidden nods, while the output consist of one output node representing the asphalt content percent. The study results have show that the radial basis function network can be applied as a recommended and appropriate computational tool to accurately and quickly determine the asphalt content of asphalt mixtures as alternative to using traditional techniques.

DESIGN OF C-SLOTTED MICROSTRIP ANTENNA USING ARTIFICIAL NEURAL NETWORK MODEL

In this paper, neural network model has been used t o estimation of resonance frequency of a coaxial fe ed C-slotted Microstrip Antenna. The Multi-Layer Perceptron Feed forward ba ck Propagation (MLPFFBP) and Radial basis function Artificial Neural Network (RBFANN) have been used to implement the ne ural network model. A relative performance analysis of the proposed neural network for different training algorithms. N umber of neurons and number of hidden layer is also carried out for estimating the resonance frequency. The method of moment (MOM) based IE3D software was used to generate data dict ionary for training and validation set of ANN. The results obtain using ANN are compared with simulation feeding and found quite satisfactory and also it is concluded that RBFANN network is more ac curate and fast compared to MLPFFBP network algorit hm.

Surrogate model application to the identification of an optimal surfactant-enhanced aquifer remediation strategy for DNAPL-contaminated sites

A surrogate model is introduced for identifying the optimal remediation strategy for Dense Non-Aqueous Phase Liquids (DNAPL)-contaminated aquifers. A Latin hypercube sampling (LHS) method was used to collect data in the feasible region for input variables. A surrogate model of the multi-phase flow simulation model was developed using a radial basis function artificial neural network (RBFANN). The developed model was applied to a perchloroethylene (PCE)-contaminated aquifer remediation optimization problem. The relative errors of the average PCE removal rates between the surrogate model and simulation model for 10 validation samples were lower than 5%, which is high approximation accuracy. A comparison of the surrogate-based simulation optimization model and a conventional simulation optimization model indicated that RBFANN surrogate model developed in this paper considerably reduced the computational burden of simulation optimization processes.

Multi-wavelength high-performance liquid chromatography: An improved method for analysis of complex substances such as Radix Paeoniae herbs

Single-wavelength high performance liquid chromatography-diode-array-detector (HPLC-DAD) and multi-wavelength combined HPLC methods were researched and developed in order to compare their performance for the classification of complex substances. Thus, the aims of this work were: to compare the performance of the single- and multi-wavelength HPLC-DAD methods for analysis of complex substances – in this context, the Radix Paeoniae herbs, which were classified on the basis of their component species and geographical origin. Three classification methods, Linear discriminant analysis (LDA), Radial basis function artificial neural network (RBF-ANN) and Least squares-support vector machine (LS-SVM), were compared on the basis of their performance in discriminating the Radix Paeoniae samples. The results showed that the multi-wavelength data produced better classification results of the two HPLC methods. This was so, irrespective of the chemometrics method used. However, the LS-SVM models were significantly better in classifying the herb samples. Consequently, the multi-wavelength HPLC-DAD approach is a strong alternative to the more common single-wavelength method, and the LS-SVM was the method of choice for classification of the complex substances such as the Radix Paeoniae herbs.

Prediction of glucagon receptor antagonist activities of some substituted imidazoles using combined radial basis function neural network and density functional theory

QSAR study of human glucagon receptor (HGR) ligands has been made with the help of quantum descriptors, such as energy of HOMO, energy of LUMO, softness, hardness using combination of principal component analysis, and radial basis function artificial neural network (ANN). Quantum descriptors have been calculated via the DFT-B3LYP method, with the basis set 6-311G. The developed neural network QSAR model outperformed the principal component regression model in both fitting and predictive abilities. ANN analysis indicated that the estimated activities were in total agreement with the experimentally observed values (R2 = 0.869, RMSD = 0.186; predictive Q2 = 0.732, RMSEcv = 0.346). The developed models were further examined by means of an external prediction set. The modeling study also reflected the important role of quantum properties of molecules when they interact with the target, HGR. The developed neural network model is expected to be useful in the rational design of new chemical entities as ligands of HGR and also for directing the synthesis of potent molecules in the future.

Prediction of CO2 Solubility in Polymers by Radial Basis Function Artificial Neural Network Based on Chaotic Self‐adaptive Particle Swarm Optimization and Fuzzy Clustering Method

AbstractTo replace costly and time‐consuming experimentation in laboratory, a novel solubility prediction model based on chaos theory, self‐adaptive particle swarm optimization (PSO), fuzzy c‐means clustering method, and radial basis function artificial neural network (RBF ANN) is proposed to predict CO2 solubility in polymers, hereafter called CSPSO‐FC RBF ANN. The premature convergence problem is overcome by modifying the conventional PSO using chaos theory and self‐adaptive inertia weight factor. Fuzzy c‐means clustering method is used to tune the hidden centers and radial basis function spreads. The modified PSO algorithm is employed to optimize the RBF ANN connection weights. Then, the proposed CSPSO‐FC RBF ANN is used to investigate solubility of CO2 in polystyrene (PS), polypropylene (PP), poly(butylene succinate) (PBS) and poly(butylene succinate‐co‐adipate) (PBSA), respectively. Results indicate that CSPSO‐FC RBF ANN is an effective method for gas solubility in polymers. In addition, compared with conventional RBF ANN and PSO ANN, CSPSO‐FC RBF ANN shows better performance. The values of average relative deviation (ARD), squared correlation coefficient (R2) and standard deviation (SD) are 0.1071, 0.9973 and 0.0108, respectively. Statistical data demonstrate that CSPSO‐FC RBF ANN has excellent prediction capability and high‐accuracy, and the correlation between prediction values and experimental data is good.

A New Radial Basis Function Artificial Neural Network based Recognition for Kurdish Manuscript

During recent decades, recognizing letters was a considerable discussion for artificial intelligence researchers and recognize letters due to the variety of languages and different approaches have many challenges. The Artificial Neural Networks (ANNs) are framed based on particular application such as recognition pattern and data classification through learning process is configured. So, it is a proper approach to recognize letters. Kurdish language has two popular handwritings based on Arabic and Latin. In this paper, Radial Basis Function (RBF) of ANNs is used to recognize Kurdish-Latin manuscripts. Although, the authors' proposed method is also used to recognize the letters of all Latin languages which include English, Turkish and etc. are used. The authors implement RBF of ANNs in MATLAB environment. In this paper, the efficiency criteria is supposed to minimize the Mean Square Error (MSE) to recognize Kurdish letters and maximize recognition accuracy of Kurdish letters in training and testing stage of RBF of ANNs. The recognition accuracy in training and testing stages are 100% and 96.7742%, respectively.

Radial Basis Function Based Neural Network for Motion Detection in Dynamic Scenes

Motion detection, the process which segments moving objects in video streams, is the first critical process and plays an important role in video surveillance systems. Dynamic scenes are commonly encountered in both indoor and outdoor situations and contain objects such as swaying trees, spouting fountains, rippling water, moving curtains, and so on. However, complete and accurate motion detection in dynamic scenes is often a challenging task. This paper presents a novel motion detection approach based on radial basis function artificial neural networks to accurately detect moving objects not only in dynamic scenes but also in static scenes. The proposed method involves two important modules: a multibackground generation module and a moving object detection module. The multibackground generation module effectively generates a flexible probabilistic model through an unsupervised learning process to fulfill the property of either dynamic background or static background. Next, the moving object detection module achieves complete and accurate detection of moving objects by only processing blocks that are highly likely to contain moving objects. This is accomplished by two procedures: the block alarm procedure and the object extraction procedure. The detection results of our method were evaluated by qualitative and quantitative comparisons with other state-of-the-art methods based on a wide range of natural video sequences. The overall results show that the proposed method substantially outperforms existing methods with Similarity and F1 accuracy rates of 69.37% and 65.50%, respectively.

NON-DESTRUCTIVE EVALUATION OF THE PULL-OFF ADHESION OF CONCRETE FLOOR LAYERS USING RBF NEURAL NETWORK

The interlayer bond is one of the primary qualities assessed during an inspection of floor concrete workmanship. The measure of this bond is the value of pull-off adhesion f b determined in practice by the pull-off method. The drawback of this method is that the tested floor is damaged in each of the test points and then needs to be repaired. This drawback can be overcome by developing a way which will make it possible to test floors in any point without damaging them locally. In this paper it is proposed to evaluate the pull-off adhesion of the top layer to the base layer in concrete floors by means of the radial basis function (RBF) artificial neural network using the parameters evaluated by the non-destructive acoustic impulse response technique and the non-destructive optical laser triangulation method. Presented RBF neural network model is useful tool in the non-destructive evaluation of the pull-off adhesion of concrete floor layers without the need to damage the top layer fragment from the base.

Prediction of critical cooling rate for glass forming alloys by artificial neural network

A radial basis function artificial neural network (RBFANN) model was established for the simulation and prediction of critical cooling rate Rc of glass forming alloys. The RBFANN model was trained, learned and examined using the data from the published literature as well as own experimental data. The performance of RBFANN model is examined by the linearly dependent coefficient between the predicted Rc and the corresponding experimental/calculated one; the influence of the type of alloys and elements and the large and minor change of element content on the Rc. In addition, a group of Zr–Al–Ni–Cu metallic glasses were designed and their Rcs were predicted by the RBFANN model. The results show that the established RBFANN model is reliable and adequate and can be used to design the composition and predict the Rc of glass forming alloys since the predicted Rc is inherent with the experimental/calculated one.

Modeling the Flux Decline during Protein Microfiltration: A Comparison between Feed-Forward Back Propagation and Radial Basis Function Neural Networks

Flux decline under various operating parameters in cross-flow microfiltration of BSA (bovine serum albumin) has been studied. A hydrophobic PES (polyethersulfone) membrane with an average pore diameter of 0.2 µm was used in all experiments. The experiments were carried out to investigate the effect of protein solution concentration and pH, trans-membrane pressure (TMP), cross-flow velocity (CFV), and membrane pore size on the flux decline trend and membrane rejection at constant trans-membrane pressure and ambient temperature. Subsequently, the experimental data, as a relatively large data set, have been subjected to a modeling study using both feed-forward back-propagation (BP) and radial basis function (RBF) artificial neural network (ANN) models. It is shown that through appropriate selection of parameters, it is possible to model the process accurately. Furthermore, it is concluded that the prediction capacity of RBFNN is superior to the BPNN, especially in the case of membrane rejection prediction.

Partner selection in agile supply chains: a fuzzy intelligent approach

Partner selection is a fundamental issue in supply chain management as it contributes significantly to overall supply chain performance. However, such decision-making is problematic due to the need to consider both tangible and intangible factors, which cause vagueness, ambiguity and complexity. This paper proposes a new fuzzy intelligent approach for partner selection in agile supply chains by using fuzzy set theory in combination with radial basis function artificial neural network. Using these two approaches in combination enables the model to classify potential partners in the qualification phase of partner selection efficiently and effectively using very large amounts of both qualitative and quantitative data. The paper includes a worked empirical application of the model with data from 84 representative companies within the Chinese electrical components and equipment industry, to demonstrate its suitability for helping organisational decision-makers in partner selection.

A non-parametric wave type based model for real-time prediction of strong ground motion accelerogram

A wave type based method for real-time prediction of strong ground motion (SGM) accelerogram is developed. Real-time prediction of SGM is requested in predictive building control systems to trigger and control actuator systems achieving the goal of reduction of the structural deformations during an on-going earthquake. It is well known that SGM is a classic example of non-stationary stochastic process with temporal variation of both amplitude and frequency content. The developed non-parametric model considers the non-homogeneity of the seismic process which contains different wave types with the individual frequency contents and time-dependency amplitude distribution pattern. Therefore, an important part of the method is to detect dominant seismic wave phases. Prediction of seismic signal is undertaken by applying frequency adaptive windowing approach, which leads to predict the on-coming signal in time window t+Δt based on the measured data in the time window t. Besides use of the frequency adaptive windowing, constant windowing and semi-adaptive windowing approaches are deployed. The results show that use of the adaptive time windows relevant to dominant frequency of the signal will enable the model to catch and predict the most dominant frequencies. Performance of the proposed model is verified by the use of 97 free-field accelerograms, which were applied to train and validate the prediction model. The selected accelerograms were measured above the soil type C and D according Eurocode 8 and their Moment magnitude are ranging between 6.2 and 7.7. The learning capability of the radial basis function Artificial Neural Network is used to reconstruct the SGM accelerogram. The most significant advantage of the proposed model is the concept of wave type based modeling which has the advantage of a conceptual physical modeling of the seismic process. Comparison of the real-time predicted and the observed accelerograms shows a high correlation when the frequency adaptive approach is applied. This paper lays a foundation for more effective use of real-time predictive control systems and potential for future extension in active structural control as well as in real-time seismology.

Application of Artificial Neural Networks for Maximum Power Point Tracking of Photovoltaic Panels

Maximum power point tracking technique for PV panels with support of online learning artificial neural network is offered. Mathematical model of the system is implemented in Matlab/Simulink environment. Maximum power point tracking is performed using IncCond algorithm and radial basis function artificial neural network. Several criteria for estimation of system performance were derived. It is shown that ANN can increase overall system efficiency by 10%.DOI: http://dx.doi.org/10.5755/j01.eee.18.10.3065

RADIACIÓN GLOBAL DIARIA ESTIMADA CON MÉTODOS CONVENCIONALES Y REDES NEURONALES ARTIFICIALES EN EL DISTRITO DE RIEGO 075

The difficulty of obtaining daily global radiation data has led to the development of models for its estimation, like the Hargreaves equation that uses maximum and minimum temperature weather variables, and the Prescott model, derived from Angström, that requires data such as sunshine hours and Angot value. Artificial neural network models are efficient tools for processing in parallel multiple input variables and estimating non-linear variables with considerable accuracy. The aim of this work was to compare the daily global radiation measured at the Santa Rosa 1 AC weather sta¬tion in irrigation district 075 in los Mochis, Sinaloa State, Mexico, with that estimated by the Hargreaves, calibrated Hargreaves, Angström-Prescott, calibrated Angström-Prescott and the feedforward backpropagation and radial basis function artificial neural network models. The Angström-Prescott model with coefficients a and b calibrated with R2 of 0.82 and RMSE of 2.33, together with the backpropagation algorithm multilayer in 3 scenarios (input variables of Julian day, minimum and maximum temperature, vapour mean pressure, relative humidity, wind speed, sunshine hours, Angot value) with R2 of 0.87 and RMSE of 1.97, were the best estimators of daily global radiation.