Recognition and classification of single and concurrent unnatural patterns in control charts via neural networks and fitted line of samples

The correct, prompt recognition and analysis of unnatural and significant patterns in Schewhart’s control charts are very important since they remind out-of-control conditions. In fact, pattern extraction increases the sensitivity of charts when identifying out of control conditions. Artificial neural networks have been used to identify unnatural patterns in many research studies due to their high efficiency in pattern recognition. In most of such studies, there is a significant risk of misclassification of highly sensitive patterns. To put it more clearly, the proposed models offered for the recognition of patterns with low parametric coefficients are mistaken. This study, offers a model for the recognition and analysis of basic patterns in process control charts using LVQ and MLP networks along with a fitted line analysis. In this model, not only does risk of misclassification at different levels of sensitivity decrease remarkably, but there will also be the possibility for recognition and analysis when basic pattern occur simultaneously. The efficiency and effectiveness of the model are shown by conducting tests based on simulation.     

基于变分贝叶斯算法和MLP网络的后非线性混合盲源分离方法研究

传统的后非线性模型往往要求其后非线性函数是可逆的,否则无法进行源信号的分离。然而在实际中,这一要求并不完全满足。针对此不足,结合变分贝叶斯推论和多层感知器网络,提出一种改进的多层感知器后非线性模型,它通过多层感知器来模拟后非线性函数,实现对不可逆后非线性函数混合的盲分离。仿真和实验结果表明该方法是有效的。     

Development and analyses of data-driven models for predicting the bed depth profile of solids flowing in a rotary kiln

Soft computing data-driven modeling (DDM) techniques have attracted the attention of many researchers across the globe as they do not require deep knowledge of the complex physical process. In the present research, data-driven based models have been developed using support vector regression (SVR), multilayer perceptron neural network (MLP), radial basis function neural network (RBFNN) and general regression neural networks (GRNN) techniques for predicting the bed depth profile of solids flowing in a rotary kiln. The performances of the developed models were compared and evaluated against the experimental results in terms of statistical measures such as coefficient of determination (R²), and average absolute relative error (AARE). The obtained results and findings from this research have revealed that data-driven models can predict the bed depth profile of solids flowing in a rotary kiln quite accurately. The SVR-based model exhibited the lowest AARE value of 1.72% and highest R² value of 0.9981 while GRNN, RBFNN, and MLP models gave corresponding values of AARE as 3.69%, 55.13%, 98.15% and those of R² as 0.9898, 0.0052 and 0.0081, respectively. Moreover, the developed DDM-based models i.e. GRNN, RBFNN, and MLP models overcame the limitations of the existing solutions which involved iterative numerical procedure entailing high degree of computational complexity.     

Pore size prediction of polyethersulfone ultrafiltration membranes using artificial neural networks

This paper reports the performance of two different artificial neural networks (ANN), Multi Layer Perceptron (MLP) and Radial Basis Function (RBF) compared to conventional software for prediction of the pore size of the asymmetric polyethersulfone (PES) ultrafiltration membranes. ANN has advantages such as incredible approximation, generalization and good learning ability. The MLP are well suited for multiple inputs and multiple outputs while RBF are powerful techniques for interpolation in multidimensional space. Three experimental data sets were used to train the ANN using polyethylene glycol (PEG) of different molecular weights as additives namely as PEG 200, PEG 400 and PEG 600. The values of the pore size can be determined manually from the graph and solve it using mathematical equation. However, the mathematical solution used to determine the pore size and pore size distribution involve complicated equations and tedious. Thus, in this study, MLP and RBF are applied as an alternative method to estimate the pore size of polyethersulfone (PES) ultrafiltration membranes. The raw data needed for the training are solute separation and solute diameter. Values of solute separation were obtained from the ultrafiltration experiments and solute diameters ware calculated using mathematical equation. With the development of this ANN model, the process to estimate membrane pore size could be made easier and faster compared to mathematical solutions.     

Employing virtual samples to build early high-dimensional manufacturing models

Machine learning algorithms are widely applied to extract useful information, but the sample size is often an important factor in determining their reliability. The key issue that makes small dataset learning tasks difficult is that the information that such datasets contain cannot fully represent the characteristics of the entire population. The principal approach of this study to overcome this problem is systematically adding artificial samples to fill the data gaps; this research employs the mega-trend-diffusion technique to generate virtual samples to extend the data size. In this paper, a real, small dataset learning task in the array process of a thin-film transistor liquid-crystal display (TFT-LCD) panel manufacturer is proposed, where there are only 20 samples used for learning the relationship between 15 inputs and 36 output attributes. The experiment results show that the approach is effective in building robust back-propagation neural network (BPN) and support vector regression (SVR) models. In addition, a sensitivity analysis is implemented with the 20 samples by using SVR to extract the relationship between the 15 factors and the 36 outputs to help engineers infer process knowledge.     

Wavelet neural network (WNN) approach for calibration model building based on gasoline near infrared (NIR) spectra

In this paper we have compared the abilities of two types of artificial neural networks (ANN): multilayer perceptron (MLP) and wavelet neural network (WNN) — for prediction of three gasoline properties (density, benzene content and ethanol content). Three sets of near infrared (NIR) spectra (285, 285 and 375 gasoline spectra) were used for calibration models building. Cross-validation errors and structures of optimized MLP and WNN were compared for each sample set. Four different transfer functions (Morlet wavelet and Gaussian derivative – for WNN; logistic and hyperbolic tangent – for MLP) were also compared. Wavelet neural network was found to be more effective and robust than multilayer perceptron.     

Hybrid discrete-time neural networks

Hybrid dynamical systems combine evolution equations with state transitions. When the evolution equations are discrete-time (also called map-based), the result is a hybrid discrete-time system. A class of biological neural network models that has recently received some attention falls within this category: map-based neuron models connected by means of fast threshold modulation (FTM). FTM is a connection scheme that aims to mimic the switching dynamics of a neuron subject to synaptic inputs. The dynamic equations of the neuron adopt different forms according to the state (either firing or not firing) and type (excitatory or inhibitory) of their presynaptic neighbours. Therefore, the mathematical model of one such network is a combination of discrete-time evolution equations with transitions between states, constituting a hybrid discrete-time (map-based) neural network. In this paper, we review previous work within the context of these models, exemplifying useful techniques to analyse them. Typical map-based neuron models are low-dimensional and amenable to phase-plane analysis. In bursting models, fast-slow decomposition can be used to reduce dimensionality further, so that the dynamics of a pair of connected neurons can be easily understood. We also discuss a model that includes electrical synapses in addition to chemical synapses with FTM. Furthermore, we describe how master stability functions can predict the stability of synchronized states in these networks. The main results are extended to larger map-based neural networks.     

Hybrid Sine Cosine and Stochastic Fractal Search for Hemoglobin Estimation

The sample's hemoglobin and glucose levels can be determined by obtaining a blood sample from the human body using a needle and analyzing it. Hemoglobin (HGB) is a critical component of the human body because it transports oxygen from the lungs to the body's tissues and returns carbon dioxide from the tissues to the lungs. Calculating the HGB level is a critical step in any blood analysis job. The HGB levels often indicate whether a person is anemic or polycythemia vera. Constructing ensemble models by combining two or more base machine learning (ML) models can help create a more improved model. The purpose of this work is to present a weighted average ensemble model for predicting hemoglobin levels. An optimization method is utilized to get the ensemble's optimum weights. The optimum weight for this work is determined using a sine cosine algorithm based on stochastic fractal search (SCSFS). The proposed SCSFS ensemble is compared to Decision Tree, Multilayer perceptron (MLP), Support Vector Regression (SVR) and Random Forest Regressors as model-based approaches and the average ensemble model. The SCSFS results indicate that the proposed model outperforms existing models and provides an almost accurate hemoglobin estimate.     

Applying ANN,ANFIS and LSSVM Models for Estimation of Acid Solvent Solubility in Supercritical CO₂

In the present work, a novel machine learning computational investigation is carried out to accurately predict the solubility of different acids in supercritical carbon dioxide. Four different machine learning algorithms of radial basis function, multi-layer perceptron (MLP), artificial neural networks (ANN), least squares support vector machine (LSSVM) and adaptive neuro-fuzzy inference system (ANFIS) are used to model the solubility of different acids in carbon dioxide based on the temperature, pressure, hydrogen number, carbon number, molecular weight, and the dissociation constant of acid. To evaluate the proposed models, different graphical and statistical analyses, along with novel sensitivity analysis, are carried out. The present study proposes an efficient tool for acid solubility estimation in supercritical carbon dioxide, which can be highly beneficial for engineers and chemists to predict operational conditions in industries.     

Characterization of random rough surfaces from scattered intensities by neural networks

Abstract

Optical scatterometry, a non-invasive characterization method, is used to infer the statistical properties of random rough surfaces. The Gaussian model with rms-roughness [sgrave] and correlation length σ is considered in this paper but the employed technique is applicable to any representation of random rough surfaces. Surfaces with wide ranges of Λ and σ, up to 5 wavelengths (λ), are characterized with neural networks. Two models are used: self-organizing map (SOM) for rough classification and multi-layer perceptron (MLP) for quantitative estimation with nonlinear regression. Models infer Λ and σ from scattering, thus involving the inverse problem. The intensities are calculated with the exact electromagnetic theory, which enables a wide range of parameters. The most widely known neural network model in practise is SOM, which we use to organize samples into discrete classes with resolution ΔΛ = Δσ = 0.5λ. The more advanced MLP model is trained for optimal behaviour by providing it with known parts of input (scattering) and output (surface parameters). We show that a small amount of data is sufficient for an excellent accuracy on the order of 0.3λ and 0.15λ for estimating Λ and σ, respectively.