Generalized Feed Forward Research Articles

Predicting grout’s uniaxial compressive strength (UCS) for fully grouted rock bolting system by applying ensemble machine learning techniques

This study proposes a novel system for accurately predicting grout’s uniaxial compressive strength (UCS) in fully grouted rock bolting systems. To achieve this, a database comprising 73 UCS values with varying water-to-grout (W/G) ratios ranging from 22 to 42%, curing times from 1 to 28 days, the admixture of fly ash contents ranging from 0 to 30%, and two Australian commercial grouts, Stratabinder HS, and BU-100, was built after conducting comprehensive series of experimental tests. After building the dataset, a metaheuristic technique, the jellyfish search (JS) algorithm was employed to determine the weight of base models in the ensemble system. This system combined various data and modelling techniques to enhance the accuracy of the UCS predictions. What sets this technique apart is the comprehensive database and the innovative use of the JS algorithm to create a weighted averaging ensemble model, going beyond traditional methods for predicting grout strength. The proposed ensemble model was called the weighted averaging ensemble model (WAE-JS), in which the obtained results of several soft computing models such as multi-layer perceptron (MLP), Bayesian regularized (BR) neural networks, generalized feed-forward (GFF) neural networks, classification and regression tree (CART), and random forest (RF) were weighted based on JS and the new results were then generated. Eventually, the result of WAE-JS was compared to other models, including MLP, BR, GFF, CART, and RF, based on some statistical parameters, such as R-squared coefficients, RMSE, and VAF as indices for evaluating the performance and capability of the proposed model. The results suggested the superiority of the ensemble WAE-JS system over the base models. In addition, the proposed WAE-JS model effectively improved the predicting accuracy achieved from the MLP, BR, GFF, CART, and RF. Furthermore, the sensitivity analysis revealed that the W/G had the most significant impact on the grout’s UCS values.

Online Hour-Ahead Load Forecasting Using Appropriate Time-Delay Neural Network Based on Multiple Correlation–Multicollinearity Analysis in IoT Energy Network

To meet up fluctuations of the real-time electric load demands, many electricity markets have gone for the real-time-market-based operation. To do so, online forecasting of the real-time load demand is necessary. Due to changes in the relation between impacting variables and output over time, a continuous-learning-based approach is highly desired. A fixed data set-based training may perform accurately for a certain amount of time, but as the load pattern and impact of different external variables change, the performance of such a model may decrease. Thus, to overcome such an inherent problem of fixed-sized databased forecasting model development, in this work, a novel simultaneous online learning and feature-engineering-based appropriate time-delay neural network has been proposed. The data for developing a forecasting model are collected through different sensors available in the Internet of Things (IoT)-based networks. To develop an optimal-cost-effective IoT network and a parsimonious model for load forecasting, variable type-dependent correlation considering the multicollinearity has been performed in online training. The proper choice of the model has been also proved using the numerical analysis with the help of time-delay embedding theory. Interestingly, it is found that, with the proper choice of inputs and their lagged variables, the proposed model performs better over general feedforward, general regression neural networks, and several deep learning and advanced models, including recurrent neural networks, fully connected deep neural networks, and dendritic neuron model.

Soil erosion modeling using erosion pins and artificial neural networks

Assessment of soil erosion is crucial for any long-term soil conservation plan. Traditional in-situ measurements provide a precise amount of erosion rate; however, the procedure is costly and time-consuming when applied over an extensive area. This study aimed to investigate the use of erosion pins and artificial neural networks (ANNs) to assess the spatial distribution of annual soil erosion rates in the mountainous areas of the north of Iran. First, annual surface erosion and splash erosion were measured using two types of erosion pins. Next, the variables affecting soil erosion (vegetation canopy, the shape of slope, slope gradient, slope length, and soil properties) were identified and estimated through field studies and analysis of a digital elevation model (DEM) and the data set were divided into three subsets of training, cross-validation, and testing. Seven artificial neural network algorithms were used and evaluated to estimate the annual soil erosion rates for the areas without recorded erosion data. Finally, the modeled values were mapped in GIS, and the longitudinal profiles of soil erosion were extracted. Findings showed that (1) Consideration should be given to the generalized feed forward (GFF) network, given the high accuracy rate (NMSE:0.1; R-sqr:0.9) compared to other tested ANN algorithms. (2) Vegetation canopy was found to be the most significant variable in annual soil erosion rate (R: −0.75 to −0.85) compared to other input variables. And (3) Annual measurements of erosion pins revealed that the splash erosion is higher (contributing 62 percent to total erosion) compared to surface runoff erosion (contributing 38 percent to total erosion).

Recognition of Off-Line Handwritten Rajasthani Characters using Generalized Feed Forward Classifier

The offline handwritten identification in the area of pattern recognition was a heavy and difficult task. Because of its application in different areas, a set of work is being done and the results are continuing to be strengthened by different methods. We suggested in this paper a handwritten model for individual character recognition using generalized neural networks for feed forward. We take 17 character samples handwritten in scanned image format for experimental purposes; Rajasthani knows 850 different samples of handwritten characters. HOG extraction methods are used to construct pattern vectors for all training sets. These features are recognition classifier for generalized feed forward. We obtained an overall classification with GFF classifier accuracy rate of 85.21% from the proposed scheme for the identification of Rajasthani characters.

Artificial neural networks and multiple linear regression as potential methods for modeling seed yield of safflower (Carthamus tinctorius L.)

Modeling relationships between safflower seed yield (SY) and its components would be helpful to understand the most important trait-(s) with significant effects on seed yield, which in turn helps safflower breeders in the indirect selection of high-performance cultivars. For this purpose, we evaluated performance of five different artificial neural network (ANN) models, including Generalized feed forward (GFF), multilayer perceptron (MLP), Jordan/Elman (JE), Principal component analysis (PCA), and Radial basis function (RBF) with different learning algorithms, transfer functions, hidden layers and neuron in each layer, along with multi-linear regression model to predict seed yield of safflower. A panel of 25 lines/cultivars was used during two years (2012–2014) to evaluate some of the important traits related to seed yield. PCA was used to perform feature extraction in the first stage of modeling. Based on PCA and first two components that incorporated approximately 79% of the total variability, five agro morphological and phenological traits, including plant height (PH), number of branches per plant (NBP), number of capsules per plant (NCP), thousand seed weight (TSW), and number of seeds per capsule (NSC), due to high values of Eigen vectors were selected as input. The performance of the models in predicting seed yield was determined using some of the statistical quality parameters, including coefficient of determination (R2), root mean square error (RMSE) and mean absolute error (MAE). The MLP model with the Levenberg–Marquardt learning algorithm, sigmoid axon transfer function, two hidden layers (i.e. topology 5-5-4-1) and 1500 training epochs was selected as the best model to predict seed yield of safflower. Sensitivity analysis indicated that the NCP is the most influential trait to predict SY in the MLP/ANN model.

Comparative Analysis of Neural Network Models for Petroleum Products Pipeline Monitoring

In recent years, Neural Network (NN) has gained popularity in proffering solution to complex nonlinear problems. Monitoring of variations in Petroleum Products Pipeline (PPP) attributes (flow rate, pressure, temperature, viscosity, density, inlet and outlet volume) which changes with time is complex due to existence of non linear interaction amongst the attributes. The existing works on PPP monitoring are limited by lack of capabilities for pattern recognition and learning from previous data. In this paper, NN models with pattern recognition and learning capabilities are compared with a view of selecting the best model for monitoring PPP. Data was collected from Pipelines and Products Marketing Company (PPMC), Port Harcourt, Nigeria. The data was used for NN training, validation and testing with different NN models such as Multilayer Perceptron (MLP), Radial Basis Function (RBF), Generalized Feed Forward (GFF), Support Vector Machine (SVM), Time Delay Network (TDN) and Recurrent Neural Network (RNN). Neuro Solutions 6.0 was used as the front-end-engine for NN training, validation and testing while My Structured Query Language (MySQL) database served as the back-end-engine. Performance of NN models was measured using Mean Squared Error (MSE), Mean Absolute Error (MAE), Correlation Coefficient (r), Akaike Information Criteria (AIC) and Minimum Descriptive Length (MDL). MLP with one hidden layer and three processing elements performed better than other NN models in terms of MSE, MAE, AIC, MDL and r values between the computed and the desired output.

GFF Classifier for Detection of Diabetic Retinopathy in Retinal Images

The emerging situation in today’s world suggests diabetic retinopathy may be a major problem in the medical world. Diabetic retinopathy is dangerous because it cannot be identified in its earlier stages and leads to vision loss. Hence, detection of diabetic retinopathy in early stage is very much important. This paper focuses on Generalized Feed Forward Neural Network (GFFNN) to detect diabetic retinopathy in retinal images. In this paper the authors present the GFFNN as a classifier to classify retinal images as normal and abnormal. 64-point Discrete Cosine Transform (DCT) and 09 statistical parameters such as Entropy, Mean, Standard deviation, Average, Euler number, Contrast, Correlation, Energy and Homogeneity are extracted from fundus retinal images to form a feature vector. The feature vector is used to train and test the GFFNN. The training and cross validation recognition rate by the GFFNN are 100% and 95.45% respectively for detection of normal and abnormal retinal images. General Terms Neural Network, Classification of retinal images

Deployment of artificial neural network for short-term forecasting of evapotranspiration using public weather forecast restricted messages

Near-future irrigation demand forecasting is important information for anticipating decisions on crop irrigation scheduling and planning water allocation in large irrigation command areas of Texas. The key determinant that is required for estimating irrigation demand in advance is toward the evapotranspiration forecast. Normally, in rich data environment, current reference evapotranspiration (ETo) is estimated by the well-known FAO56 PM method which requires bunch of observed climatic data. In poor data environment for either current or future estimation, this well-known method application is restricted. Indeed, the correctness of ETo forecast remains a challenging computational task, since inaccurate weather variables can alter the forecast accuracy. Therefore, this study aims to employ artificial neural network (ANN) methodology for forecasting near future ETo values by using restricted climate information messages retrieved from public weather forecast source. Four ANNs learning algorithms including the Generalized Feedforward (GFF), Linear Regression (LR), Multilayer Perceptron (MLP) and Probabilistic Neural Network (PNN) are applied with three sets of inputs combination composed of minimum (Tmin) and maximum (Tmax) daily air temperatures, extraterrestrial radiation (Ra) and net solar radiation (Rs) to forecast ETo in Dallas. The coefficient of correlation (CC), mean square error (MSE), normalized mean square error (NMSE), mean absolute error (MAE) and mean square error skill score (MSESS) were used for the models evaluation. Statistically, in comparison with FAO56 PM, the performances of ANNs models using only Tmax and Tmin predictors were inferior to those of Tmax, Tmin and Ra. With Tmax, Tmin and Rs input-sets, MLP yielded the highest accuracies (CC=0.926; MSE=0.770mm/day, NMSE=0.143mm/day; MAE=0.708mm/day). Tmax is an important ETo forecast predictor, and the performance improvement relies mostly on Rs accuracy. With precise weather forecast information, ANN made ETo forecast possible (Average CC=0.860, MSESS=0.738). These results can assist irrigation districts to accommodate in advance their crop water demand to near-future irrigation requirement.

A Generalized Feed Forward Dynamic Adaptive Mesh Refinement and Derefinement Finite Element Framework for Metal Laser Sintering—Part I: Formulation and Algorithm Development

A novel multiscale thermal analysis framework has been formulated to extract the physical interactions involved in localized spatiotemporal additive manufacturing processes such as the metal laser sintering. The method can be extrapolated to any other physical phenomenon involving localized spatiotemporal boundary conditions. The formulated framework, named feed forward dynamic adaptive mesh refinement and derefinement (FFD-AMRD), reduces the computational burden and temporal complexity needed to solve the many classes of problems. The current study is based on application of this framework to metals with temperature independent thermal properties processed using a moving laser heat source. The melt pool diameters computed in the present study were compared with melt pool dimensions measured using optical micrographs. The strategy developed in this study provides motivation for the extension of this simulation framework for future work on simulations of metals with temperature-dependent material properties during metal laser sintering.

BLADDER CANCER RECOGNITION: NEW TREND

Magnetic Resonance Imaging (MRI) has been widely applied to various medical procedures. The daily growth of medical data leads to human mistakes in manual analysis; and increases the need for automatic analysis. Therefore, applying tools to collect, classify, and analyse medical data automatically is essential. Medical imaging issues are extremely complex, due to the high importance of correct diagnosis and treatment of diseases in healthcare systems. For these reasons, automatic medical image analysis algorithms are used to help increase the reliability and accurate understanding of medical images. The objective of this paper is to investigate the use of Artificial Neural Network (ANN) algorithms, such as multilayer perceptron (MLP), Jordan/Eleman network, Self-Organizing Feature Map (SOFM) and Support Vector Machine (SVM), to early detect bladder cancer (diagnosis), to determine tumour staging (for the sake of prognosis), and to assess the accuracy of MRI in T staging bladder cancer. A set of functional images, taken by Magnetic Resonance (MR), was used. It was found that multilayer perceptron (MLP) neural network gave better results than all other algorithms. We developed a model that defines cancer level in order to enhance its treatment. Experimental results show that the devised approach increases the accuracy of bladder cancer diagnosis to 81.8% using Generalized Feed Forward (GFF) after processing for more than 40 hours.

Strongly Secure Scan Design Using Generalized Feed Forward Shift Registers

Strongly Secure Scan Design Using Generalized Feed Forward Shift Registers

English

The modeling of rainfall-runoff relationship in a watershed is very important in designing hydraulic structures, controlling flood and managing storm water. Artificial Neural Networks (ANNs) are known as having the ability to model nonlinear mechanisms. This study aimed at developing a Generalized Feed Forward (GFF) network model for predicting annual flood (depth) of Johor River in Peninsular Malaysia. In order to avoid over training, cross-validation technique was performed for optimizing the model. In addition, predictive uncertainty index was used to protect of over parameterization. The governing training algorithm was back propagation with momentum term and tangent hyperbolic types was used as transfer function for hidden and output layers. The results showed that the optimum architecture was derived by linear tangent hyperbolic transfer function for both hidden and output layers. The values of Nash and Sutcliffe (NS) and Root mean square error (RMSE) obtained 0.98 and 5.92 for the test period. Cross validation evaluation showed 9 process elements is adequate in hidden layer for optimum generalization by considering the predictive uncertainty index obtained (0.14) for test period which is acceptable.

Predicting Diel, Diurnal and Nocturnal Dynamics of Dissolved Oxygen and Chlorophyll‐a Using Regression Models and Neural Networks

AbstractHuman‐induced and natural interruptions with continuous streams of observational data necessitate the development of gap‐filling and prediction strategies towards better understanding, monitoring and management of aquatic systems. This study quantified the efficacy of multiple non‐linear regression (MNLR) versus artificial neural network (ANN) models as well as the temporal partitioning of diurnal versus nocturnal data for the predictions of chlorophyll‐a (chl‐a) and dissolved oxygen (DO) dynamics. The temporal partitioning increased the predictive performances of the best MNLR models of diurnal DO by 45% and nocturnal DO by 4%, relative to the best diel MNLR model of diel DO ($r_{{\rm adj}}^{2} = 68.8\%$). The ANN‐based predictions had a higher predictive power than the MNLR‐based predictions for both chl‐a and DO except for diurnal DO dynamics. The best ANNs based on independent validations were multilayer perceptron (MLP) for diel chl‐a, generalized feedforward (GFF) for diurnal and nocturnal chl‐a, MLP for diel DO, GFF for diurnal DO, and MLP for nocturnal DO.

Generalized Feed Forward Shift Registers and Their Application to Secure Scan Design

SUMMARY In this paper, we introduce generalized feed-forward shift registers (GF 2 SR) to apply them to secure and testable scan design. Previously, we introduced SR-equivalents and SR-quasi-equivalents which can be used in secure and testable scan design, and showed that inversioninserted linear feed-forward shift registers (I 2 LF 2 SR) are useful circuits for the secure and testable scan design. GF 2 SR is an extension of I 2 LF 2 SR and the class is much wider than that of I 2 LF 2 SR. Since the cardinality of the class of GF 2 SR is much larger than that of I 2 LF 2 SR, the security level of scan design with GF 2 SR is much higher than that of I 2 LF 2 SR. We consider how to control/observe GF 2 SR to guarantee easy scan-in/out operations, i.e., state-justification and state-identification problems are considered. Both scan-in and scan-out operations can be overlapped in the same way as the conventional scan testing, and hence the test sequence for the proposed scan design is of the same length as the conventional scan design. A program called WAGSR (Web Application for Generalized feed-forward Shift Registers) is presented to solve those problems.

Artificial neural network modeling of process and product indices in deep bed drying of rough rice

This study aimed to model the performance indices of deep bed drying of rough rice using artificial neural networks (ANNs), compare the ANN approach to the multivariate regression method, and determine the sensitivity of the ANN model to the input variables. The effects of air temperature, air velocity, and air relative humidity on drying kinetics, product output rate (POR), evaporation rate (ER), and percentage of kernel cracking (KC) were investigated. To predict the dependent parameters, 3 well-known networks, namely the multilayer perceptron, generalized feed forward (GFF), and modular neural network, were examined. The GFF networks with the Levenberg-Marquardt learning algorithm, hyperbolic tangent activation function, and 4-15-1, 3-4-4-1, 3-7-1, and 3-11-1 topologies provided superior results, respectively, for predicting moisture content, POR, ER, and CK. The values of all of the drying indices predicted by the ANN were closer to the experimental data than linear and logarithmic regression models. The output variables were significantly affected by the dependent variables. However, air temperature and air relative humidity showed the maximum and the minimum influence on the network outputs, respectively.

English

Epilepsy is a common neurological disorder that is characterized by recurrent unprovoked seizures. About 40 to 50 million people worldwide are reported to have epilepsy. In this paper the authors present clinical decision support system (DSS) for the diagnosis of epilepsy. The DSS is developed by using Multilayer Perceptron (MLP), Generalized Feed Forward Neural Network (GFF-NN) and Elman Neural Network (E-NN). The validity of neural networks to diagnose the epilepsy is checked and the most suitable neural network is recommended for the diagnosis of epilepsy. Also the different feature enhancement techniques like principal component analysis (PCA), FFT and statistical parameters are used for the input dimensionality reduction. Epilepsy diagnosis is modeled as the classification of normal EEG, interictal EEG and ictal EEG. With the different input dimensionality reduction methods performance parameters of MLP, GFF-NN and E-NN are measured and compared. For the GFF-NN, number of free parameter is reduced up to 92.22% when PCA is used for input dimensionality reduction with the overall accuracy of 98.61%.

Global Stabilization of Feedforward Nonlinear System Based on Nested Saturated Control

In this paper,we discuss global stabilization procedure for convergence of a more general feedforward nonlinear systems.Our stabilizer consists of a nested saturation function,which is a nonlinear combination of saturation functions.We extend the existing stabilization results and prove that our stabilizer is exponential convergent.

Global Stabilization of Feedforward Nonlinear System Based on Nested Saturated Control

Abstract In this paper, we discuss global stabilization procedure for convergence of a more general feedforward nonlinear systems. Our stabilizer consists of a nested saturation function, which is a nonlinear combination of saturation functions. We extend the existing stabilization results and prove that our stabilizer is exponential convergent.

Induction Machine Fault Detection Using Generalized Feed Forward Neural Network

Industrial motors are subject to incipient faults which, if undetected, can lead to motor failure. The necessity of incipient fault detection can be justified by safety and economical reasons. The technology of artificial neural networks has been successfully used to solve the motor incipient fault detection problem. This paper develops inexpensive, reliable, and noninvasive NN based incipient fault detection scheme for small and medium sized induction motors. Detailed design procedure for achieving the optimal NN model and Principal Component Analysis for dimensionality reduction is proposed. Overall thirteen statistical parameters are used as feature space to achieve the desired classification. GFFD NN model is designed and verified for optimal performance in fault identification on experimental data set of custom designed 2 HP, three phase 50 Hz induction motor.

Induction machine fault detection using support vector machine based classifier

Industrial motors are subject to various faults which, if unnoticed, can lead to motor failure. The necessity of incipient fault detection can be justified by safety and economical reasons. The technology of artificial neural networks has been successfully used to solve the motor fault detection problem. This paper develops inexpensive, reliable, and noninvasive NN based fault detection scheme for small and medium sized induction motors. Detailed design procedure for achieving the optimal NN model and Principal Component Analysis for dimensionality reduction is proposed. Overall thirteen statistical parameters are used as feature space to achieve the desired classification. Generalized Feed Forward (GFFDNN) and Support Vector Machine (SVM) NN models are designed and verified for optimal performance in fault identification on experimental data set of custom designed 2 HP, three phase 50 Hz induction motor.