Artificial Adaptive System Research Articles

Polynomial Based Functional Link Artificial Recurrent Neural Network adaptive System for predicting Indian Stocks

AbstractA low complexity Polynomial Functional link Artificial Recurrent Neural Network (PFLARNN) has been proposed for the prediction of financial time series data. Although different types of polynomial functions have been used for low complexity neural network architectures earlier for stock market prediction, a comparative study is needed to choose the optimal combinations of the nonlinear functions for a reasonably accurate forecast. Further a recurrent version of the Functional link neural network is used to model more accurately a chaotic time series like stock market indices with a lesser number of nonlinear basis functions. The proposed PFLARNN model when trained with the well known gradient descent algorithm produces reasonable accuracy with a choice of range of weight parameters of the network. However, to improve the accuracy of the forecast further, the weight parameters of the recurrent functional neural network are optimized using an evolutionary learning algorithm like the differential evo...

The constellation of dietary factors in adolescent acne: a semantic connectivity map approach.

Different lifestyle and dietetic factors have been linked with the onset and severity of acne. To assess the complex interconnection between dietetic variables and acne. This was a reanalysis of data from a case-control study by using a semantic connectivity map approach. 563 subjects, aged 10-24 years, involved in a case-control study of acne between March 2009 and February 2010, were considered in this study. The analysis evaluated the link between a moderate to severe acne and anthropometric variables, family history and dietetic factors. Analyses were conducted by relying on an artificial adaptive system, the Auto Semantic Connectivity Map (AutoCM). The AutoCM map showed that moderate-severe acne was closely associated with family history of acne in first degree relatives, obesity (BMI ≥ 30), and high consumption of milk, in particular skim milk, cheese/yogurt, sweets/cakes, chocolate, and a low consumption of fish, and limited intake of fruits/vegetables. Our analyses confirm the link between several dietetic items and acne. When providing care, dermatologists should also be aware of the complex interconnection between dietetic factors and acne.

Predicting mechanical properties of elastomeric modified nylon blend using adaptive neuro-fuzzy interference system and neural network

Artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) are employed to predict the final mechanical properties of Polyamide 6 (PA6)/ ethylene-propylene-rubber (EPR)-grafted blend at different compositions. An individual ANFIS model is developed to predict the mechanical property at different processing condition. The results of this study are suggestive that both ANFIS and ANN could be employed quite effectively for this aim in such a polymer blend system. Therefore, the mechanical properties such as yield strength, modulus, and Izod impact strength are well predicted employing this method including an acceptable range of errors.

Comparison of artificial neural network and adaptive neuro-fuzzy inference system for predicting the wrinkle recovery of woven fabrics

The aim of this study was to compare the artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) models for predicting the wrinkle recovery of polyester/cotton woven fabrics. The prediction models were developed using experimental data-set of 115 fabric samples of different constructions. Warp and weft yarn linear densities, ends/25 mm and picks/25 mm, were used as input/predictor variables, and warp and weft crease recovery angles (CRA) as output/response variables. It was found that the prediction accuracy of the ANN models was slightly better as compared with that of ANFIS models developed in this study. However, the ANFIS models could characterize the relationships between the input and output variables through surface plots, which the ANN models could not. The developed models may be used to optimize the fabric construction parameters for maximizing the wrinkle recovery of polyester/cotton woven fabrics.

Comparision of some models for estimation of reflectance of hypericum leaves under stress conditions

Abstract Lack of water resources and high water salinity levels are among the most important growth-restricting factors for plants species of the world. This research investigates the effect of irrigation levels and salinity on reflectance of Saint John’s wort leaves (Hypericum perforatum L.) under stress conditions (water and salt stress) by multiple linear regression (MLR), artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS). Empirical and heuristics modeling methods were employed in this study to relate stress conditions to leaf reflectance. It was found that the constructed ANN model exhibited a high performance than multiple regression and ANFIS in estimating leaf reflectance accurately.

Prediction of Compressional and Shear Slowness from Conventional Well Log Data: Using Intelligent Systems

Compressional slowness (DTCO) and shear slowness (DTSM) are two fundamental parameters that have many applications in petrophysical, geophysical, and geomechanical operations. These two parameters can be obtained using a dipole sonic imaging tool, but unfortunately this tool is run in just a few wells of a field. Therefore, it is important to predict compressional and shear slowness indirectly from the other conventional well logs that have strong correlation with these two parameters when the dipole sonic log is missing in a given well. The overriding impetus of this work is to construct intelligent systems, including artificial neural network, fuzzy logic, and adaptive neuro fuzzy inference system, for prediction of compressional and shear slowness. Finally, the results are combined with a committee machine algorithm using a simple averaging technique. A total number of 2,715 data points from Asmari formation, which has compressional and shear slowness, are used. These data are divided into two groups: 2,172 data points for construction of intelligent systems and 543 data points used for model testing. The results showed that, despite a difference in concept, all of the intelligent techniques were successful for estimation of compressional and shear slowness.

Synchro Phasor Measurement Based Fault Analysis of a Parallel Transmission Line

This paper presents a new method to classify transmission line shunt faults and determine the fault location using phasor data of the transmission system. Most algorithms employed for analyzing fault data require that the fault type to be classified. The older fault-type classification algorithms are inefficient because they are not effective under certain operating conditions of the power system and may not be able to accurately select the faulted transmission line if the same fault recorder monitors multiple lines. An intelligent techniques described in this paper is used to precisely detect all ten types of shunt faults that may occur in an electric power transmission system (double-circuit transmission lines) with the help of data obtained from phasor measurement unit. This method is virtually independent of the mutual coupling effect caused by the adjacent parallel circuit and insensitive to the variation of source impedance. Thousands of fault simulations by MATLAB have proved the accuracy and effectiveness of the proposed algorithm. This paper includes the analysis of fault identification techniques using Artificial Neural Network and Adaptive Neuro-Fuzzy Inference System based protection schemes. The performances of the techniques are examined for different faults on the parallel transmission line and compared with the conventional relay scheme. The results obtained shows that ANFIS based fault identification gives better performance than other techniques.

Artificial Neural Networks: An Overview and their Use in the Analysis of the AMPHORA-3 Dataset

The Artificial Adaptive Systems (AAS) are theories with which generative algebras are able to create artificial models simulating natural phenomenon. Artificial Neural Networks (ANNs) are the more diffused and best-known learning system models in the AAS. This article describes an overview of ANNs, noting its advantages and limitations for analyzing dynamic, complex, non-linear, multidimensional processes. An example of a specific ANN application to alcohol consumption in Spain, as part of the EU AMPHORA-3 project, during 1961–2006 is presented. Study's limitations are noted and future needed research using ANN methodologies are suggested.

Estimating Some Physical Properties of Sour and Sweet Cherries Based on Combined Image Processing and AI Techniques

Abstract Physical properties of agricultural products are considered as important factors in optimization of storage conditions, packaging, transportation, water adsorption/desorption, heat, pesticides, and foodstuff moving out and also their breathing. This paper presents a time and cost economizing method to determine these important attributes of sour and sweet cherries by combining image processing and two common artificial intelligence techniques, artificial neural networks (ANNs) and adaptive neuro-fuzzy inference system (ANFIS). The measuring technique consisted of a charge-coupled device camera for image acquisition, fluorescent illuminants, capture card, and MATLAB for image analysis. Several networks were designed, trained, and generalized with a back-propagation algorithm using “trainlm” as training function. Several ANFIS models were designed with different number and type of membership functions (MFs) for each input. Generally, “gaussian” and “pi-shaped” MFs showed better results for estimating output variables among others. Considering statistical analysis, ANFIS showed better results than ANN.

Comparative Modeling and Analysis of the Mass Transfer Coefficient in a Turbulent Bed Contactor using Artificial Neural Network and Adaptive Neuro-Fuzzy Inference Systems

This study deals with predicting the gas film volumetric mass transfer coefficient (Koga) in a turbulent bed contactor (TBC), using both artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) techniques. The networks have been trained and evaluated with the experimental data available in the literature. Input variables to the networks are process variables such as gas and liquid phase concentration, gas and liquid superficial velocities and also specific area of packings. The results obtained the ability of developed ANN and ANFIS for prediction of Koga. Although it was observed that both ANN and ANFIS models provided a good statistical prediction in terms of coefficient of determination (R2), mean relative error (MRE) and root mean square error (RMSE), the accuracies of ANFIS predictions were better than those of ANN predictions.

Functional abilities and cognitive decline in adult and aging intellectual disabilities. Psychometric validation of an Italian version of the Alzheimer's Functional Assessment Tool (AFAST): analysis of its clinical significance with linear statistics and artificial neural networks

(a) A psychometric validation of an Italian version of the Alzheimer's Functional Assessment Tool scale (AFAST-I), designed for informant-based assessment of the degree of impairment and of assistance required in seven basic daily activities in adult/elderly people with intellectual disabilities (ID) and (suspected) dementia; (b) a pilot analysis of its clinical significance with traditional statistical procedures and with an artificial neural network. AFAST-I was administered to the professional caregivers of 61 adults/seniors with ID with a mean age (± SD) of 53.4 (± 7.7) years (36% with Down syndrome). Internal consistency (Cronbach's α coefficient), inter/intra-rater reliabilities (intra-class coefficients, ICC) and concurrent, convergent and discriminant validity (Pearson's r coefficients) were computed. Clinical significance was probed by analysing the relationships among AFAST-I scores and the Sum of Cognitive Scores (SCS) and the Sum of Social Scores (SOS) of the Dementia Questionnaire for Persons with Intellectual Disabilities (DMR-I) after standardisation of their raw scores in equivalent scores (ES). An adaptive artificial system (AutoContractive Maps, AutoCM) was applied to all the variables recorded in the study sample, aimed at uncovering which variable occupies a central position and supports the entire network made up of the remaining variables interconnected among themselves with different weights. AFAST-I shows a high level of internal homogeneity with a Cronbach's α coefficient of 0.92. Inter-rater and intra-rater reliabilities were also excellent with ICC correlations of 0.96 and 0.93, respectively. The results of the analyses of the different AFAST-I validities all go in the expected direction: concurrent validity (r=-0.87 with ADL); convergent validity (r=0.63 with SCS; r=0.61 with SOS); discriminant validity (r=0.21 with the frequency of occurrence of dementia-related Behavioral Excesses of the Assessment for Adults with Developmental Disabilities, AADS-I). In our sample age and gender do not correlate with the scale and comparing the distribution of the AFAST-I and DMR-SCS and DMR-SOS expressed as ES, it appears that memory disorders and temporal and spatial disorientation (SCS) precede the loss of functional abilities, whereas changes in social behaviour (SOS) are less specific in detecting cognitive deterioration sufficient to provoke functional disability and vice versa. The results of AutoCM analysis reveal that the hub (core) of the entire network is represented by the functional domain 'personal/oral hygiene' in the entire study sample and 'use of toilet' in a subgroup of subjects who obtained an ES equal to 0 at DMR-SCS. These results confirm the reliability and validity of AFAST-I and emphasise the complexity of the relationship among functional status, cognitive functioning and behaviour also in adults/seniors with ID.

Adaptive voltage regulation of self excited induction generator using FACTS controllers

This paper introduces an adaptive terminal voltage regulation of a standalone self excited induction generator using flexible AC transmission systems (FACTS). FACTS in the form of static VAR compensator (SVC) is introduced. SVC is used to regulate the terminal voltage of self excited induction generator-based (SEIG) wind generator working with variable speed and load. Since, the wind generator operates over a wide range of operating conditions; the terminal voltage of the induction generator is not regulated. This terminal voltage is controlled by adapting the value of the excitation capacitance of SVC using artificial neural networks (ANN) and adaptive neuro-fuzzy inference systems (ANFIS). Changing the excitation capacitance is performed by controlling the firing angle of SVC under different operating conditions to control the terminal voltage. A mathematical model of SEIG with variable load and speed along with SVC is developed and simulated in MATLAB to get the excitation capacitance required for voltage control. Simulation results are used to train both ANN and ANFIS for on-line prediction of the suitable firing angles required to control the terminal voltage of the system under these operating conditions. Results signify the supremacy of ANFIS over ANN in terms of performance measures.

Extraction of wave parameter for classification of partial discharge using Pearson's correlation method

Partial discharge (PD) behaviour in electrical apparatus leads to the identification of the degradation of insulating materials. The major problem here is identifying the different types of PD discharges that affect the insulation system. In general the characteristics of PD pulse will be in the form of internal, corona and surface discharges (both in air and oil conditions). Thus the classification of PD patterns aims to recognise the discharges from unknown source. In this paper a new enhanced technique is used to identify the PD discharges namely Pearson's correlation coefficient techniques. Starting with PD data on different families of specimen, extract φ-q-n PD patterns (3D patterns) and apply to the enhanced techniques, then segmented the 3D patterns in to number of samples. Thereafter, by evaluating statistical operators like skewness, kurtosis and Pearson's correlated values from the PD signals and recognise the different PD patterns using artificial neural networks and adaptive neuro fuzzy interference system.

A committee machine based soft sensor as an alternative to multiphase flow meter for oil flow rate prediction of the wells

Multiphase flow meters (MFMs) which measure three-phase oil-gas-water flow rates, are being utilized to make available, quick and accurate well test data in different oil production applications, like in remote or unmanned locations, topside and subsea applications. Data acquisition and production monitoring of the wells are done discretely by conventional MFMs, due to radioactive sources and unmanned location due to wells far distance. This study presents the development of committee machine based soft sensor (CMSS), an alternative way to the conventional MFMs. The proposed CMSS combines the results of linear and nonlinear auto-regression exogenous input (ARX/NARX), artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) for overall oil flow rate prediction of the wells, on the basis of available temperature and pressure measurements of lines. Each ensemble member has a weight factor which is derived in two ways including simple averaging and weighted averaging. In the weighted averaging method, the optimal combination of the weights is obtained by a novel optimization based method called imperialist competitive algorithm (ICA). Experiments on data set of 31 wells in one of the northern Persian Gulf oil fields of Iran proved the effectiveness of the proposed ICA optimized CMSS with an improved accuracy over the individual experts.

Emergence of a Snake-Like Structure in Mobile Distributed Agents: An Exploratory Agent-Based Modeling Approach

The body structure of snakes is composed of numerous natural components thereby making it resilient, flexible, adaptive, and dynamic. In contrast, current computer animations as well as physical implementations of snake-like autonomous structures are typically designed to use either a single or a relatively smaller number of components. As a result, not only these artificial structures are constrained by the dimensions of the constituent components but often also require relatively more computationally intensive algorithms to model and animate. Still, these animations often lack life-like resilience and adaptation. This paper presents a solution to the problem of modeling snake-like structures by proposing an agent-based, self-organizing algorithm resulting in an emergent and surprisingly resilient dynamic structure involving a minimal of interagent communication. Extensive simulation experiments demonstrate the effectiveness as well as resilience of the proposed approach. The ideas originating from the proposed algorithm can not only be used for developing self-organizing animations but can also have practical applications such as in the form of complex, autonomous, evolvable robots with self-organizing, mobile components with minimal individual computational capabilities. The work also demonstrates the utility of exploratory agent-based modeling (EABM) in the engineering of artificial life-like complex adaptive systems.

Extraction of Wave Parameter for Classification of Partial Discharge using Pearson’s Correlation Method

Partial discharge (PD) behaviour in electrical apparatus leads to the identification of the degradation of insulating materials. The major problem here is identifying the different types of PD discharges that affect the insulation system. In general the characteristics of PD pulse will be in the form of internal, corona and surface discharges (both in air and oil conditions). Thus the classification of PD patterns aims to recognise the discharges from unknown source. In this paper a new enhanced technique is used to identify the PD discharges namely Pearson’s correlation coefficient techniques. Starting with PD data on different families of specimen, extract Φ-q-n PD patterns (3D patterns) and apply to the enhanced techniques, then segmented the 3D patterns in to number of samples. Thereafter, by evaluating statistical operators like skewness, kurtosis and Pearson’ss correlated values from the PD signals and recognise the different PD patterns using artificial neural networks and adaptive neuro fuzzy interference system.

Combining Personality Traits with Traditional Risk Factors for Coronary Stenosis: An Artificial Neural Networks Solution in Patients with Computed Tomography Detected Coronary Artery Disease

Background. Coronary artery disease (CAD) is a complex, multifactorial disease in which personality seems to play a role but with no definition in combination with other risk factors. Objective. To explore the nonlinear and simultaneous pathways between traditional and personality traits risk factors and coronary stenosis by Artificial Neural Networks (ANN) data mining analysis. Method. Seventy-five subjects were examined for traditional cardiac risk factors and personality traits. Analyses were based on a new data mining method using a particular artificial adaptive system, the autocontractive map (AutoCM). Results. Several traditional Cardiovascular Risk Factors (CRF) present significant relations with coronary artery plaque (CAP) presence or severity. Moreover, anger turns out to be the main factor of personality for CAP in connection with numbers of traditional risk factors. Hidden connection map showed that anger, hostility, and the Type D personality subscale social inhibition are the core factors related to the traditional cardiovascular risk factors (CRF) specifically by hypertension. Discussion. This study shows a nonlinear and simultaneous pathway between traditional risk factors and personality traits associated with coronary stenosis in CAD patients without history of cardiovascular disease. In particular, anger seems to be the main personality factor for CAP in addition to traditional risk factors.

THERMO-HYDRAULIC BEHAVIOR MODELING OF PASSIVE HEAT TRANSFER ENHANCEMENT TECHNIQUES USING A SOFT COMPUTING APPROACH

The main aim of this research is to demonstrate effectiveness of soft computing techniques in thermo-hydraulic behavior modeling of passive heat transfer enhancement (HTE) techniques. An artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS), two effective modeling methods, have been used to model Nusselt numbers and friction factors of wire coil and twisted tape inserts in various flow regimes. The experimental data sets were utilized for training and validation of these models, and their results were compared with the corresponding correlations. The mean relative error (MRE) between the predicted results and experimental data of ANN and ANFIS models were found to be less than 3% and 1.5% for thermo-hydraulic behavior modeling of wire coil and twisted tape inserts, respectively. Depending on model complexity, performance of both ANN and ANFIS models was found to be superior to that of the corresponding power-law regressions. Hence, application of the soft computing approach to predict the performance of thermal systems in engineering applications is recommended.

Artificial Neural Networks, and Evolutionary Algorithms as a systems biology approach to a data-base on fetal growth restriction

One of the specific aims of systems biology is to model and discover properties of cells, tissues and organisms functioning. A systems biology approach was undertaken to investigate possibly the entire system of intra-uterine growth we had available, to assess the variables of interest, discriminate those which were effectively related with appropriate or restricted intrauterine growth, and achieve an understanding of the systems in these two conditions. The Artificial Adaptive Systems, which include Artificial Neural Networks and Evolutionary Algorithms lead us to the first analyses. These analyses identified the importance of the biochemical variables IL-6, IGF-II and IGFBP-2 protein concentrations in placental lysates, and offered a new insight into placental markers of fetal growth within the IGF and cytokine systems, confirmed they had relationships and offered a critical assessment of studies previously performed.

Macrocell path loss prediction using artificial intelligence techniques

The prediction of propagation loss is a practical non-linear function approximation problem which linear regression or auto-regression models are limited in their ability to handle. However, some computational Intelligence techniques such as artificial neural networks (ANNs) and adaptive neuro-fuzzy inference systems (ANFISs) have been shown to have great ability to handle non-linear function approximation and prediction problems. In this study, the multiple layer perceptron neural network (MLP-NN), radial basis function neural network (RBF-NN) and an ANFIS network were trained using actual signal strength measurement taken at certain suburban areas of Bauchi metropolis, Nigeria. The trained networks were then used to predict propagation losses at the stated areas under differing conditions. The predictions were compared with the prediction accuracy of the popular Hata model. It was observed that ANFIS model gave a better fit in all cases having higher R2 values in each case and on average is more robust than MLP and RBF models as it generalises better to a different data.