This paper describes the simulation of two hybrid evolutionary algorithms (EAs) to the feedforward neural networks (NNs) used in classification problems. Besides backpropagation algorithm, simple genetic algorithm and...
详细信息
This paper describes the simulation of two hybrid evolutionary algorithms (EAs) to the feedforward neural networks (NNs) used in classification problems. Besides backpropagation algorithm, simple genetic algorithm and random search algorithm, the paper considers simple hybrid genetic algorithm and hybrid random search algorithm. The objective is to analyze the performance of hybrid genetic algorithm and hybrid random search algorithm over other discussed algorithms for the classification problem. The experiments considered a feedforward NN trained with simple hybrid genetic algorithm/hybrid random search algorithm and 39 types of network structures and artificial data sets. In most cases, the hybrid evolutionary feedforward NNs seemed to be better than the other algorithms. We found few differences in the performance of the networks trained by applying the hybrid genetic algorithms, but found ample differences in the execution time. The results suggest that the hybrid evolutionary feedforward NN might be the best algorithm on the data sets we tested. (c) 2006 Elsevier B.V. All rights reserved.
A recurrent fuzzy neural network (RFNN) controller based on real-time genetic algorithms (GAs) is developed for a linear induction motor (LIM) servo drive in this paper. First, the dynamic model of an indirect field-o...
详细信息
A recurrent fuzzy neural network (RFNN) controller based on real-time genetic algorithms (GAs) is developed for a linear induction motor (LIM) servo drive in this paper. First, the dynamic model of an indirect field-oriented LIM servo drive I'S derived. Then, an online training RFNN with a backpropagation algorithm is introduced as the tracking controller. Moreover, to guarantee the global convergence of tracking error, a real-time GA is developed to search the optimal learning rates of the RFNN online. The GA-based RFNN control system is proposed to control the mover of the LIM for periodic motion. The theoretical analyses for the proposed GA-based RFNN controller are described in detail. Finally, simulated and experimental results show that the proposed controller provides high-performance dynamic characteristics and is robust with regard to plant parameter variations and external load disturbance.
An analog neural network (NN) was developed for real-time surface recognition by using two photoelectrical signals issued from a phase-shift rangefinder. The NN architecture consists of a multilayer perceptron (MLP) w...
详细信息
An analog neural network (NN) was developed for real-time surface recognition by using two photoelectrical signals issued from a phase-shift rangefinder. The NN architecture consists of a multilayer perceptron (MLP) with two inputs, three neurons in the hidden layer, and one output. The NN output is compared with threshold voltages in order to classify the tested surfaces. In this type of application, analog NN implementation has many advantages, especially the small silicon area used, a low-power consumption, and no analog-to-digital conversions. This recognition system has been successfully tested for four types of surfaces (a plastic surface, a glossy paper, a painted wall, and a porous surface), at a remote distance between the rangefinder and the target varying from 0.5 in up to 1.25 in and with a laser beam incidence angle varying between -pi/6 and pi/6. This paper presents the NN training and the experimental tests of surface discrimination.
The relationship between the physical properties of metal is often very complex in nature with its chemistry and several other rolling parameters in operation. Non-linear regression models play a very important role i...
详细信息
The relationship between the physical properties of metal is often very complex in nature with its chemistry and several other rolling parameters in operation. Non-linear regression models play a very important role in modelling the underlying mechanism, provided it is known. Artificial neural networks provide a wide class of general-purpose and flexible non-linear regression models. The most commonly used neural networks, called multi-layered perceptrons, can vary the complexity of the model from a simple parametric model to a highly flexible nonparametric model. In this particular work, an industry-based data set is used for learning and optimizing the neural network architecture using some well-known algorithms for prediction under neural-net systems. The outcome of the analysis is compared with the results achieved through empirical statistical modelling from its prediction error level and the knowledge of materials science.
The problem of automatic modulation classification is to identify the modulation type of a received signal from the signal parameters. Modulation classification has both military and civilian applications and has been...
详细信息
ISBN:
(纸本)9780819466891
The problem of automatic modulation classification is to identify the modulation type of a received signal from the signal parameters. Modulation classification has both military and civilian applications and has been the subject of intensive research for more than two decades. In this paper we use a hierarchical neural network in which the first network identifies the modulation class while a second set of networks identify the constellation size (order) of that modulation class. The set of features we use include normalized standard deviations of amplitude, phase and frequency, as well as the fourth and sixth order cumulants of the signal samples. Identifying the constellation size of quadrature amplitude modulation (QAM) has been particularly difficult in the past. In this paper we introduce two new approaches for computing the features of a QAM signal. The first uses the concatenated inphase and quadrature components of the signal to compute the features. The second method maps the in-phase and quadrature components to the first quadrant of the constellation by calculating the absolute value of each separately. The mean of the resulting constellation points is then subtracted before calculating the features. Simulation results are presented for classification of several digital modulation schemes including FSK, PSK, ASK and QAM. Our results show that the proposed method significantly improves the classification error.
This research explores the automated detection of surface defects that fall across two different background textures in a light-emitting diode (LED) chip. Water-drop defects, commonly found on chip surface, impair the...
详细信息
ISBN:
(纸本)9783540723929
This research explores the automated detection of surface defects that fall across two different background textures in a light-emitting diode (LED) chip. Water-drop defects, commonly found on chip surface, impair the appearance of LEDs as well as their functionality and security. Automated inspection of a water-drop defect is difficult because the defect has a semi-opaque appearance and a low intensity contrast with the rough exterior of the LED chip. Moreover, the blemish may fall across two different background textures, which further increases the difficulties of defect detection. We first use the one-level Haar wavelet transform to decompose a chip image and extract four wavelet characteristics. Then, the Multi-Layer Perceptron (MLP) neural network with back-propagation (BPN) algorithm is applied to integrate the multiple wavelet characteristics. Finally, the wavelet-based neural network approach judges the existence of water-drop defects. Experimental results show that the proposed method achieves an above 96.8% detection rate and a below 4.8% false alarm rate.
The relationship between the physical properties of metal is often very complex in nature with its chemistry and several other rolling parameters in operation. Non-linear regression models play a very important role i...
详细信息
The relationship between the physical properties of metal is often very complex in nature with its chemistry and several other rolling parameters in operation. Non-linear regression models play a very important role in modelling the underlying mechanism, provided it is known. Artificial neural networks provide a wide class of general-purpose and flexible non-linear regression models. The most commonly used neural networks, called multi-layered perceptrons, can vary the complexity of the model from a simple parametric model to a highly flexible nonparametric model. In this particular work, an industry-based data set is used for learning and optimizing the neural network architecture using some well-known algorithms for prediction under neural-net systems. The outcome of the analysis is compared with the results achieved through empirical statistical modelling from its prediction error level and the knowledge of materials science.
This study aims to incorporate Artificial Neural Network (ANN) for measuring the effectiveness of the TV broadcast advertisements (toothpaste) by discovering important factors that influence the advertisement effectiv...
详细信息
This study aims to incorporate Artificial Neural Network (ANN) for measuring the effectiveness of the TV broadcast advertisements (toothpaste) by discovering important factors that influence the advertisement effectiveness. The information about the effects of each of these factors has been studied and it is used for measuring the advertisement effectiveness. Fifty attributes are examined to derive values from thirteen factors. These thirteen factors are used as input to ANN model. The data collected from 837 respondents are used for training and testing the ANN. The backpropagation algorithm is used for adjusting the weights in the ANN. Experimental results show that the ANN model achieves 99% accuracy for measuring the advertisement effectiveness. (c) 2005 Elsevier Ltd. All rights reserved.
In this paper, I present a method to determine and predict precisely the GPS satellite orbit by using a neural network. The neural network used in this paper is based on the BP (backpropagation) learning algorithm. Th...
详细信息
In this paper, I present a method to determine and predict precisely the GPS satellite orbit by using a neural network. The neural network used in this paper is based on the BP (backpropagation) learning algorithm. The BP algorithm is particularly attractive because it is H-infinity optimal. It is a robust algorithm in the sense that small disturbances and modeling errors lead to small estimation errors (For a non-robust algorithm, such as the classical maximum likelihood and least square methods, it is possible that small disturbances and modeling errors may result in large estimation errors). This is certainly the case for the estimation of the GPS satellite orbit because the satellite orbital model usually contains small disturbances and perturbations that are difficult to model. Currently, the simulation result shows that we can use the well-trained network to predict about six days' data and the orbital will can be within a meter. The result is compared with the classical polynomial interpolation method. It is believed that, if we extend the training time, the prediction period can be much longer.
暂无评论