Neural Network (NN)

An adaptive neural network (NN) controller is developed for an underwater remotely operated vehicle (ROV). Radial basis neural network and multilayer neural network are used in the closed-loop to approximate the nonlinear dynamics of the vehicle. The low cost ROV was designed and fabricated by department Mechanical Engineering of Guilan University, is considered. This technique does not require the iterative off-line training process to identify the plant parameters. The stability of the presented control systems are guaranteed on the basis of the Lyapunov theory. Finally the performances of the vehicle with NN controllers are compared with a PD controller. The significant improvements are observed in tracking performance of the ROV in all controllable degrees of freedom.

In this paper an efficient classification system using hierarchal multi_agent's technology based on neural network is introduced, where each agent implements as a neural network (trained using back propagation learning algorithm). The system consist of two layers of agents, the top layer contain one agent works as control agent, its responsibility is to select the right agent from the agents in the lower layer to classify the related pattern depending on data's features. Two techniques were used (regularization and earlier stopping criteria) to select the best one for each data set depending. The system provides a degree of generalization with the ability to improve it if there is a need for more generalization. The developed system was tested using different standard datasets obtained from the (University of California, Irvine) UCI Machine Learning Repository for the empirical analysis of machine learning algorithms. These are (User Knowledge Level, iris, and Banknote Authentication Dataset).

An accurate monitoring of power system behavior is a hot-topic for modern grid operation. Low-frequency oscillations (LFO), such as inter-area electromechanical oscillations, are detrimental phenomena impairing the development of the grid itself and also the integration of renewable sources. An interesting countermeasure to prevent the occurrence of such oscillations is to continuously identify their characteristic electromechanical mode parameters, possibly realizing an online monitoring system. In this paper an attempt to develop an online modal parameters identification system is done using machine learning techniques. An approach based on the development of a proper artificial neural network exploiting the frequency measurements coming from actual PMU devices is presented. The specifically developed offline training stage is fully detailed. The output results from the dynamic mode decomposition method are considered as reference in order to validate the machine learning approach. Some results are presented in order to validate the effectiveness of the proposed approach on data coming from recordings of real grid events. The main key points affecting the performance of the proposed technique are discussed by means of proper validation scenarios. This contribution is the first step of a more extended project whose final aim is the development of an artificial neural networks (ANN) architecture able to predict the system behavior (in a given time span) in terms of LFO modal parameters, and to classify the contingencies/disturbances based on an online training that has memory of the passed training samples.

Raman Spectroscopy enables in-depth study into the molecular structure of solid, liquid and gasses from its scattering spectrum. As such, the spectrum could offer a biochemical fingerprint to identify unknown molecules. Surface Enhanced Raman Spectroscopy (SERS) amplifies the weak Raman signal by 10 +3 to 10 +7 times, revolutionary making the method appealing to the research community. SERS has been proven useful for disease detection from a medium such as a cell, serum, urine, plasma, saliva, tears. The spectra displayed are noisy and complicated by the presence of other molecules, besides the targeted one. Moreover, the difference between the infected and controlled samples is far too minute for detection by the naked human eyes. Hence, signal processing techniques are found crucial to single out fingerprint of the target molecule from biological spectra. Our work here examines signal processing techniques attempted on SERS spectra for disease detection, such as Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), Artificial Neural Network (ANN), Support Vector Machine (SVM) and Logistic Regression Analysis (LRA). It is found that PCA-LDA is the most popular (45%), ensued by PCA-ANN (33%) and SVM (22%). PCA-SVM yields the highest in accuracy (99.9%), followed by PCA-ANN (98%) and LRA (97%). PCA-LDA and SVM score the highest in both sensitivity-specificity.

In this paper, frequency oscillation has been present as it is an important
issue in power systems, especially when it is joined to the new trend to use alternative energies as an energy source, as well as to be a big risk for the inter-connection of the modern electric power networks. Wind farm acts as alternative energy and connected to two buses on the Iraqi power system. Because the low-frequency oscillation monitoring needs be accurate and fast, the main objective is to propose a novel online monitoring system consisting of phasor measurement unit's (PMU) with artificial intelligence neural network (PMU-NN). The location of the phasor measurement units has been optimized using (graph-theoretic procedure algorithm) and the function for the artificial intelligence (NN) is radial basis function (RBFNN). The data information from phasor measurement units is the inputs to the artificial intelligence system then predictions are made Information on low-frequency oscillation (target). The MATLAB toolboxes (PSAT & NN) used to obtain results. Finally, from the results, the validity of the proposed (PMU-NN) system has been proven and tested on the Iraqi power grid (24 bus) in several cases and several places on the network and the comparison was made with the analysis model.

A position control of a discussion of DC servomotors is addressed in this article via a novel adaptive PID with sliding mode control approach. In this works to introduce the exact type of power control by sliding mode control for position and speed control of DC servo motor. The paper contributes union and examination of DC servo motor with sliding mode controller and a regular PID controller. This procedure has done through the demonstrating and reproductions are done and their performing is evaluated as in consistent and in addition in a transient state. Thusly, the DC servo motor position drive is dubious to parameter grouping and load disturbing effect, a liberal control approach in light of sliding mode. The proposed approach has the additionally favored standpoint that, for outside unsettling influence, it just requires a bound to exist, without having to know the extent of this bound. The proposed controller is connected to control a model of unverifiable acceptance servomotor subject to huge unsettling influences and a model of DC servomotor with obscure parameters and vulnerability in load condition. In this article, the helpfulness of the anticipated Procedure is approved by performing simulations utilizing MATLAB device. The reproduction comes about an exhibit that the part of the sliding mode-based course of action is supplementary strong than robust than fixed gain PID controller.

In this work, a neuro-fuzzy (NF) simulation study was conducted in order to screen candidate reservoirs for enhanced oil recovery (EOR) projects in Angolan oilfields. First, a knowledge pattern is extracted by combining both the searching potential of fuzzy-logic (FL) and the learning capability of neural network (NN) to make a priori decisions. The extracted knowledge pattern is validated against rock and fluid data trained from successful EOR projects around the world. Then, data from Block K offshore Angolan oilfields are then mined and analysed using box-plot technique for the investigation of the degree of suitability for EOR projects. The trained and validated model is then tested on the Angolan field data (Block K) where EOR application is yet to be fully established. The results from the NF simulation technique applied in this investigation show that polymer, hydrocarbon gas, and combustion are the suitable EOR techniques.

Development and applications of neural network (NN)-based approaches for representing potential energy surfaces (PES) of bound and reactive molecular systems are reviewed. Specifically, it is shown that when the density of ab initio points is low, NNs-based potentials with multibody or multimode structure are advantageous for representing high-dimensional PESs. Importantly, with an appropriate choice of the neuron activation function, PESs in the sum-of-products form are naturally obtained, thus addressing a bottleneck problem in quantum dynamics. The use of NN committees is also analyzed and it is shown that while they are able to reduce the fitting error, the reduction is limited by the nonrandom nature of the fitting error. The approaches described here are expected to be directly applicable in other areas of science and engineering where a functional form needs to be constructed in an unbiased way from sparse data. V

Diagnosis of heart disease is complex. ECG plays an important role inthe analysis and diagnosis of cardiac disease. Normally ECG signals are affected by different noises and analysis of those signal is also a tedious process.The main objective of the paper is to de-noise and analyze the ECG signal using DWT(Discrete Wavelet Transform) technique. We use the Daubechies wavelet analysis (level 5)technique in DWT for better performance. Then we obtain the decomposed wave.By analyzing the parameters of the decomposed ECG signal we find the disease present in the heart of the patient by applying the decomposed parameters to ANN(Artificial Neural Network). Thus, we can identifythe cardiac disease. This method is very effective because the accuracy of the result is high. Also, the training and testing of the network take very less time. Hence, it is more effective for users.

There is an increase in cardio logical patients all over the world due to change in modern life style. It forces the medical researchers to search for smart devices that can diagnosis and predict the onset of cardiac problem before it is too late. This motivates the authors to predict Arrhythmia that can help both the patients and the medical practitioners for better healthcare services. The proposed method uses the frequency domain information which can represent the ECG signals of Arrhythmia patients better. Features representing the MIT-BIH Arrhythmia are extracted using the efficient Short Time Fourier Transform and the Wavelet transform. A comparison of these features is made with that of normal human being using Neural Network based classifier. Wavelet based features has shown an improvement of Accuracy over that of STFT features in classifying Arrhythmia as our results reveal. A Mean Square Error (MSE) of with wavelet transform has validated our results.

Electrocardiogram (ECG) is the most important and widely used method to study the diseases related to heart. The detailed study of ECG graph helps the medical practitioner to diagnose the condition of the heart. Based on the information provided by ECG graph, an appropriate treatment can be given to the patient. The patient with the medical history of heart alignments should maintain a record of ECG papers for timely analysis and diagnosis of the diseases, which requires large storage space and extensive manual effort. The visual technique of analyzing the ECG signals is tedious and time consuming. In order to overcome this problem, an automatic system which involves digital signal integration and analysis tool is developed using MATLAB. This provides an effective strategy for analog to digital conversion of legated paper biomedical map, equipped with a plottingter. The conversion of electrocardiography (ECG) information from charts into digital ECG signals is designed using MATLAB....

The paper presents a new system for ECG (ElectroCardioGraphy) signal recognition using different neural classifiers and a binary decision tree to provide one more processing stage to give the final recognition result. As the base classifiers, the three classical neural models, i.e., the MLP (Multi Layer Perceptron), modified TSK (Takagi-Sugeno-Kang) and the SVM (Support Vector Machine), will be applied. The coefficients in ECG signal decomposition using Hermite basis functions and the peak-to-peak periods of the ECG signals will be used as features for the classifiers. Numerical experiments will be performed for the recognition of different types of arrhythmia in the ECG signals taken from the MIT-BIH (Massachusetts Institute of Technology and Boston’s Beth Israel Hospital) Arrhythmia Database. The results will be compared with individual base classifiers’ performances and with other integration methods to show the high quality of the proposed solution

This paper deals with the design and implementation of a neural network-based selftuning controller. The structure of the controller is based on using a neural network, or a set of them, as a self-tuner for a controller. The intention of this approach is to take advantage of the ability to learn of the neural networks and to use them in place of an identifier in the conventional self-tuner scheme. The work is divided into two main parts. The first one is dedicated to the design of the self-controller. And the second is an application of the algorithm on a nonlinear system: an overhead crane. Some simulations were carried out to verify the efficiency of the self-tuner and then a real-time implementation on a scale prototype was performed.

In this method, radial basis function network RBFNN is an artificial intelligent which is used to identify and classify the communication system performance. RBFNN is one type of neural network which has activation functions. It consists of three-layer input layer, hidden layer and output linear combination. One of the main problems of communication system is that it causes slow response for sending signal via the transmission devices. Therefore, the artificial intelligent by RBFNN is used to optimize the transmission signal. The input signal is trained and testing by neurons with weight and this lead to provide linear output. The simulation results have the optimization specifics over the traditional communication transmission devices.

The classical Proportional-Integral (PI) control for Direct Current (DC) motor causes slow response of actual speed with high overshoot and undershoot which leads to sluggishness of the system. To minimize the problem of PI controller, intelligent technique based on hybrid neural network sliding mode control NN-SMC is suggested. The benefits of SMC are that it is simple, and tough to parameter deviations as compared with other controllers. In this paper, the neural network NN is used to minimize the error between reference speed and actual speed. In addition, the SMC aim is to control and optimize the voltage that is supplies the DC motor which guarantees the robust performance of the speed controller under disturbances. The proposed method for the speed control is first calculated and executed to DC motor by using MATLAB SIMULINK. The results of the suggested NN-SMC are compared with the traditional PI controller. The results obviously show the supremacy of NN-SMC over PI controller.

In public and private places, people panic when unpredicted events occur. The modelling and analysis of the crowd surveillance videos is an issue that exists due to the unconstrained behaviors observed. The use of computer aided techniques to address this issue has been proposed by researchers. Similarly, we proposed a Graph theoretic approach based Crowd Behavior Analysis and Classification System (GCBACS) model in this paper. The crowd personnel behavior is observed and graph theoretic approach is used for analysis and classification of the activity observed. Streak Flows is considered to observe crowd behaviors. The novelty of GCBACS is that considers inter personnel activities for analysis and classification. The GCBACS is evaluated using varied surveillance videos in the experimental study. The results presented prove that the GCBACS is able to achieve accurate classification results in dense and sparse crowd scenarios. The GCBACS is compared with the Spatio-Temporal Viscous Fluid Field method. The results prove that the GCBACS outperforms the Spatio-Temporal Viscous Fluid Field method.

Nowadays image processing occupies a dynamic role in every area. Image processing refers to progress the images and mine attributes from images for various process. Image segmentation is one of the processes of image processing which is used to segregate an image into different homogenous segments. It is an essential process of pixel clustering, partitions raw image into non-overlapping regions. Image segmentation plays an important role in almost all image processing applications. This paper focuses to provide an overview of the literature in image segmentation with various techniques implemented in them, their merits and demerits etc. Comparison based on accuracy was also done to prove the efficiency of the various techniques for image segmentation. The comparison results show the best image segmentation method among them.

A continuous video contains two important components such as tracks of the person in the video, and localization of the actions that are performed by the actors. The analysis of the activity is used for solving both the tracking, and recognition problems. In this paper, we have deployed an efficient activity analysis framework for determining the activity of the human in the video. Initially, the input video is obtained from the ULCA, and VIRAT datasets, then the video file is converted into multiple video frames named as frames.The information regarding each framesare obtained and further the frames are resized for preventing the memory from dumping.The noise present in each frame is filtered using the Gaussian filter. The Graph Cut technique is used for extracting the shape of the object from the background. The tracking of the video file is performed using the Bounding Box technique. The features from the resultant image are extracted using the Local Binary Pattern (LBP). Based on the features obtained the frames a pattern is generated. These feature values are grouped into activity segments using the Neural Network (NN) classifier. To validate the performance of the proposed NN classifier it is validated with the existing Support Vector Machine (SVM) classifier for the metrics such as accuracy, precision, and recall. The experimental results proved that the proposed NN classifier produced optimal results than the existing SVM.

A continuous video contains two important components such as tracks of the person in the video, and localization of the actions that are performed by the actors. The analysis of the activity is used for solving both the tracking, and recognition problems. In this paper, we have deployed an efficient activity analysis framework for determining the activity of the human in the video. Initially, the input video is obtained from the ULCA, and VIRAT datasets, then the video file is converted into multiple video frames named as frames.The information regarding each framesare obtained and further the frames are resized for preventing the memory from dumping.The noise present in each frame is filtered using the Gaussian filter. The Graph Cut technique is used for extracting the shape of the object from the background. The tracking of the video file is performed using the Bounding Box technique. The features from the resultant image are extracted using the Local Binary Pattern (LBP). Based on the features obtained the frames a pattern is generated. These feature values are grouped into activity segments using the Neural Network (NN) classifier. To validate the performance of the proposed NN classifier it is validated with the existing Support Vector Machine (SVM) classifier for the metrics such as accuracy, precision, and recall. The experimental results proved that the proposed NN classifier produced optimal results than the existing SVM

Accurate non-invasive Electrocardiogram (ECG) analysis has a
significant emerging role in automated cardiac state diagnosis. Conventional
ECG analysis techniques such as linear and second-order spectra fail to retain
Fourier phase relationship and suppress random variations in non-linear, nonstationary
and non-Gaussian ECG signals. This may provide misguided results.
A highly accurate algorithm utilising statistics of fourth-order spectra
(trispectrum) is introduced to capture clinically significant variations in ECG
that can cater to these limitations. Five temporal interval and three trispectral
entropy features are extracted from individual beat of ECG signals loaded from
MIT-BIH arrhythmia database and quantified using box plots. A three-layer
feedforward neural network classifier with 50 hidden layer neurons is
configured using the Levenberg–Marquardt algorithm to yield an average
accuracy of 95.10% while classifying six cardiac states. Significant
enhancement in the performance with reduced computational complexity has
been observed using a low-dimensional hybrid ECG feature set and a simple
classifier showing the effectiveness of the proposed man–machine interface for
automated cardiac state diagnosis.

This paper focuses on designing an adaptive radial basis function neural network (RBFNN) control method for a class of nonlinear systems with unknown parameters and bounded disturbances. The problems raised by the unknown functions and external disturbances in the nonlinear system are overcome by RBFNN, combined with the single parameter direct adaptive control method. The novel adaptive control method is designed to reduce the amount of computations effectively. The uniform ultimate boundedness of the closed-loop system is guaranteed by the proposed controller. A coupled motor drives (CMD) system, which satisfies the structure of nonlinear system, is taken for simulation to confirm the effectiveness of the method. Simulations show that the developed adaptive controller has favorable performance on tracking desired signal and verify the stability of the closed-loop system.

In public and private places, people panic when unpredicted events occur. The modelling and analysis of the crowd surveillance videos is an issue that exists due to the unconstrained behaviors observed. The use of computer aided techniques to address this issue has been proposed by researchers. Similarly, we proposed a Graph theoretic approach based Crowd Behavior Analysis and Classification System (GCBACS) model in this paper. The crowd personnel behavior is observed and graph theoretic approach is used for analysis and classification of the activity observed. Streak Flows is considered to observe crowd behaviors. The novelty of GCBACS is that considers inter personnel activities for analysis and classification. The GCBACS is evaluated using varied surveillance videos in the experimental study. The results presented prove that the GCBACS is able to achieve accurate classification results in dense and sparse crowd scenarios. The GCBACS is compared with the Spatio-Temporal Viscous Fluid Field method. The results prove that the GCBACS outperforms the Spatio-Temporal Viscous Fluid Field method.

In public and private places, people panic when unpredicted events occur. The modelling and analysis of the crowd surveillance videos is an issue that exists due to the unconstrained behaviors observed. The use of computer aided techniques to address this issue has been proposed by researchers. Similarly, we proposed a Graph theoretic approach based Crowd Behavior Analysis and Classification System (GCBACS) model in this paper. The crowd personnel behavior is observed and graph theoretic approach is used for analysis and classification of the activity observed. Streak Flows is considered to observe crowd behaviors. The novelty of GCBACS is that considers inter personnel activities for analysis and classification. The GCBACS is evaluated using varied surveillance videos in the experimental study. The results presented prove that the GCBACS is able to achieve accurate classification results in dense and sparse crowd scenarios. The GCBACS is compared with the Spatio-Temporal Viscous Fluid Field method. The results prove that the GCBACS outperforms the Spatio-Temporal Viscous Fluid Field method.

The ECG is nothing but the recording of the
heart’s electrical activity.ECG Feature Extraction plays a
significant role in diagnosing most of the cardiac diseases.
One cardiac cycle in an ECG signal consists of the P-QRS-T
waves. This feature extraction scheme determines the
amplitudes and intervals in the ECG signal for subsequent
analysis. The amplitudes and intervals value of P-QRS-T
segment determines the functioning of heart of every human.
Recently, numerous research and techniques have been
developed for analyzing the ECG signal. The proposed
schemes were mostly based on Fuzzy Logic Methods,
Artificial Neural Networks (ANN), Genetic Algorithm(GA),
Support Vector Machines (SVM), and other Signal Analysis
techniques. All these techniques and algorithms have their
advantages and limitations. This proposed paper discusses
various techniques and transformations proposed earlier in
literature for extracting feature from an ECG signal. In
addition this paper also provides a comparative study of
various methods proposed by researchers in extracting the
feature from ECG signal.