Search

Abstract: Photovoltaic (PV) generation systems (PGSs) have become an attractive option among renewable energy sources because they are clean, maintenance-free and environmental friendly. For PGSs, a simple and fast maximum power point tracking (MPPT) algorithm is essential. Although the static tracking efficiency of conventional MPPT method is usually high, it drops noticeably in case of rapidly changing irradiance conditions. In this paper, two fast and accurate digital MPPT methods for fast changing environments are proposed. By using piecewise line segments or cubic equation to approximate the maximum power point (MPP) locus, two high-speed, low-complexity MPPT techniques can be developed. To make the developed system more convenient for common PGS users, neural network (NN)-based program which can be used to calculate the parameters of the emulated MPP locus is also developed and embedded into the proposed digital MPPT system. Theoretical derivation and detailed design procedure will be provided in this paper. The advantages of the proposed system include low computation requirement, fast tracking speed and high static/dynamic tracking efficiencies. To validate the effectiveness and correctness of the proposed methods, simulation and experimental results of a 230W PV system will also be provided. [Copyright &y& Elsevier]

Concurrency control (CC) algorithms guarantee the correctness and consistency criteria for concurrent execution of a set of transactions in a database. A precondition that is seen in many CC algorithms is that the writeset (WS) and readset (RS) of transactions should be known before the transaction execution. However, in real operational environments, we know the WS and RS only for a fraction of transaction set before execution. However, optional knowledge about WS and RS of transactions is one of the advantages of the proposed CC algorithm in this paper. If the WS and RS are known before the transaction execution, the proposed algorithm will use them to improve the concurrency and performance. On the other hand, the concurrency control algorithms often use a specific static or dynamic equation in making decision about granting a lock or detection of the winner transaction. The proposed algorithm in this paper uses an adaptive resonance theory (ART)-based neural network for such a decision making. In this way, a parameter called health factor (HF) is defined for transactions that is used for comparing the transactions and detecting the winner one in accessing the database objects. HF is calculated using ART2 neural network. Experimental results show that the proposed neural-based CC (NCC) algorithm increases the level of concurrency by decreasing the number of aborts. The performance of proposed algorithm is compared with strict two-phase locking (S2PL) algorithm, which has been used in most commercial database systems. Simulation results show that the performance of proposed NCC algorithm, in terms of number of aborts, is better than S2PL algorithm in different transaction rates. [ABSTRACT FROM AUTHOR]

In this paper, the method of the graphical interpretation of the single-layer network weights is introduced. It is shown that the network parameters can be converted to the image and their particular elements are the pixels. For this purpose, weight-to-pixel conversion formula is used. Moreover, new weights' modification method is proposed. The weight coefficients are computed on the basis of pixel values for which image filtration algorithms are implemented. The approach is applied to the weights of three types of the models: single-layer network, two-layer backpropagation network and the hybrid network. The performance of the models is then compared on two independent data sets. By means of the experiments, it is presented that the adjustment of the weights to new values decreases test error value compared to the error obtained for initial set of weights. [ABSTRACT FROM AUTHOR]

Multitudinous parameters involved have made the direct methanol fuel cell (DMFC) a complex “black-box,” posing challenges and difficulties in its modeling. This paper presents a neural network (NN) model with immune-based particle swarm optimization (IPSO) approach of the DMFC system, which is different from the conventional complex mathematical models. With the actual operation of DMFC taken into consideration, the polarization curves are run under a stable condition as the reference data for training the model. To reduce time cost for the training procedure and maintaining minimum modeling error, the IPSO algorithm is applied to the learning procedure of NN model. By combining the NN and the IPSO, the weight of the transfer function on the node in the hidden layer can be adjusted to minimize modeling error. The simulation results were in agreement with the experimental results, showing that the hybridization of NN model with IPSO approach can effectively demonstrate the polarization behaviors on a DMFC system. Therefore, this hybrid NN model with IPSO approach can be used as a simulation tool, which can save much money and time for reforming the conventional mathematical models with expensive experiment. Furthermore, the proposed method reveals an adaptive ability to improve the model even if the DMFC system structure is different. [ABSTRACT FROM PUBLISHER]

The dressing process is affected on the wheel surface topography and the selection of optimal dressing parameters makes increasing in the grinding performance. In this paper, a feed forward backpropagation neural network (FFBP-NN) and a simulated annealing (SA) algorithm were used for simultaneously minimize the tangential grinding force as well as minimize the surface roughness. First, the FFBP-NN was developed using the data generated based on experimental observations. In the experiments procedure, the grinding conditions were constant and only the disk dressing conditions were varied. Then, a SA algorithm was applied to the FFBP-NN for solving the optimal dressing parameters. In order to demonstrate the model performance, several initial trial values have tested. Results show that the proposed model has an acceptable performance to optimize the grinding process. [ABSTRACT FROM AUTHOR]

People counting is a very important problem in visual surveillance. An accurate and real time estimation of people in a crowded place can provide valuable information. Here video is given as input and outputs the average number of people passing over the video. The video input is separated to number of frames and some processing steps are performed on background subtraction results to estimate the number of people in a complicated scene. Foreground pixel extraction is done with ViBe (VIsual Background Extractor) algorithm. The extracted foreground image's pixels count is calculated and gives as input to the neural network. In learning phase, the people count is calculated manually with test dataset and while testing phase remaining test cases are tested by adjusting weight parameters to obtain relative to the target result. [ABSTRACT FROM AUTHOR]

In this paper, a batch gradient algorithm with adaptive momentum is considered and a convergence theorem is presented when it is used for two-layer feedforward neural networks training. Simple but necessary sufficient conditions are offered to guarantee both weak and strong convergence. Compared with existing general requirements, we do not restrict the error function to be quadratic or uniformly convex. A numerical example is supplied to illustrate the performance of the algorithm and support our theoretical finding. [ABSTRACT FROM AUTHOR]

Multiple algorithms have been developed to correct user's typing mistakes. However, an optimal solution is hardly identified among them. Moreover, these solutions rarely produce a single answer or share common results, and the answers may change with time and context. These motivated this research to synthesize some distinct word correction algorithms to produce an optimal prediction based on database updates and neural network learning. In this paper, three distinct typing correction algorithms are integrated as a pilot research with key factors such as Time Change, Context Change and User Feedback being considered. Experimental results show that the developed WLR model (i.e., word-list neural network ranking model) achieves the best results in comparison with Levenshtein distance, Metaphone. and Two-Gram prediction algorithms throughout all stages. It achieves 57.50% Ranking First Hitting Rate with samples of category one and a best Ranking First Hitting Rate of 74.69% within category four. Further work is recommended to extend the number of parameters and integrate more algorithms to achieve a higher level of accuracy. [ABSTRACT FROM AUTHOR]

The paper proposes a policy design method for system dynamics models based on recurrent neural networks. A policy maker first directly creates an arbitrary desired reference mode and run the algorithm to search for the most appropriate model(s) automatically to fit it. In the searching process, both the system structure and its parameter values evolve simultaneously. Several experiments are conducted to evaluate our approach. The results show that our approach is as good as or better than other comparable methods. Copyright © 2011 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

In this paper, a novel multiscale algorithm to simulate accumulation of trabecular bone crack density and crack length at macroscopic scale during cyclic loading is developed. The method is based on finite element analysis and neural network computation to link mesoscopic (trabecular level) and macroscopic (whole femur) scales. The finite element calculation is performed at the macroscopic level and a trained neural network incorporated into the finite element code Abaqus is employed as a numerical device to perform the local mesoscopic computation. Based on a set of mesoscale simulations of representative volume elements obtained by digital image-based modeling technique using µ-CT and voxel finite element, a neural network is trained to approximate the local finite element responses. The input data for the artificial neural network are the applied stress, the stress orientation and the cycle frequency. The output data are the averaged crack density and crack length at a given site of the bone. Copyright © 2010 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

The single viewpoint constraint is a principal optical characteristic for most catadioptric omnidirectional vision. Single viewpoint catadioptric omnidirectional vision is very useful because it allows the generation of geometrically correct perspective images from one omnidirectional image. Therefore precise calibration for single viewpoint constraint is needed during system assembling. However, in most image detection based calibration methods, the nonlinear optical distortion brought by lens is often neglected. Hence the calibration precision is poor. In this paper, a new calibration method of single viewpoint constraint for the catadioptric omni-directional vision is proposed. Firstly, an image correction algorithm is obtained by training a neural network. Then, according to characteristics of the space circular perspective projection, the corrected image of the mirror boundary is used to estimate its position and attitude relative to the camera to guide the calibration. Since the estimate is conducted based on actual imaging model rather than the simplified model, the estimate error is largely reduced, and the calibration accuracy is significantly improved. Experiments are conducted on simulated images and real images to show the accuracy and the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Abstract: The utilization of motor-in-wheel in the electric vehicle (EV) is a very important research topic. In order to simplify the control of EV, a new system is proposed by Guilin Tao, which changes the traditional configuration and control strategy of direct-wheel-driven EV. In this system two permanent-magnet-brushless dc motors (PMBDCMs) are connected directly in serial, in order to cut down the differential algorithm thoroughly in velocity control strategy. This structure can realize the differential automatically when the EV runs on the smooth road at low speed. But this structure considers the characters of the motor only, which is an ideal system. EV is a complex electromechanical system, and its work condition is also complex. The variation of the external condition and the work condition of operation can affect the operation of motor, so the state of the EV will change. In this paper, according to the change of the work condition and road condition, combined with the dynamics of EV, this problem was well studied. The neural network PID control mode was used in this paper, and by means of the Matlab/Simulink and the experimental verification, it reveals that this method can enhance the safety and stability of this system. [Copyright &y& Elsevier]

Abstract: Conditions under which it is possible to design a correct algorithm and a six-level spatial neural network reproducing the computations performed by this algorithm for recognition problems with binary data (Ω-regular problems) are found. A distinctive feature of this network is the use of diagonal activation functions in its internal layers, which significantly simplify intermediate computations in the inner and outer loops. Given an Ω-regular problem, the network sequentially computes the rows of the classification matrix for the test sample objects. The computational process (i.e., the inner loop) for each test object consists inside the elementary 3-level network (i.e., μ-block) of a single iteration determined by a single object of the training set. The proposed approach to the neural network construction does not rely on the conventional approach based on the minimization of a functional; rather, it is based on the operator theory developed by Zhuravlev for solving recognition and classification problems. [ABSTRACT FROM AUTHOR]

Collaborative filtering that relies on overall ratings has been widely accepted due to the ability to generate satisfactory recommendations. However, the most challenging difficulty of this approach is the lack of sufficient ratings or the so-called data sparsity. Moreover, sometimes these ratings alone are not sufficient to precisely understand users' specific behaviours. A user may show his/her overall preferences on an item through the overall ratings but at the same time, they may not satisfy with certain aspects of the item. This situation happened due to the emphasis on aspects that may be different among users and will effect a user's final decisions. Therefore, in this paper, we proposed a model called Neural Network model for Multi-Aspect with Strong Correlation (NNMASC) that utilize the significance of aspect's correlation to enhance the predictive accuracy of personalized recommendation. We integrate the user, item, aspects and overall ratings via embedding features by utilizing the available multi-aspect ratings from hotel reviews dataset. NNMASC adopts a feed-forward neural network with back propagation algorithm to make rating prediction. The experimental result using MAE shows that the proposed model has significantly outperformed the traditional models and the state-of-the-art approaches in terms of prediction accuracy. [ABSTRACT FROM AUTHOR]

general neural network based methodology that has the potential to replace the computationally-intensive site-specific seismic analysis of structures is proposed in this paper. The basic framework of the methodology consists of a feed forward back propagation neural network algorithm with one hidden layer to represent the seismic potential of a region and soil amplification effects. The methodology is implemented and verified with parameters corresponding to Delhi city in India. For this purpose, strong ground motions are generated at bedrock level for a chosen site in Delhi due to earthquakes considered to originate from the central seismic gap of the Himalayan belt using necessary geological as well as geotechnical data. Surface level ground motions and corresponding site-specific response spectra are obtained by using a one-dimensional equivalent linear wave propagation model. Spectral acceleration values are considered as a target parameter to verify the performance of the methodology. Numerical studies carried out to validate the proposed methodology show that the errors in predicted spectral acceleration values are within acceptable limits for design purposes. The methodology is general in the sense that it can be applied to other seismically vulnerable regions and also can be updated by including more parameters depending on the state-of-the-art in the subject. [ABSTRACT FROM AUTHOR]

A prediction scheme for Ethernet traffic data using a Multiscale-Bilinear Recurrent Neural Network with Adaptive Learning (M-BRNN-AL) is proposed and presented in this paper. The proposed predictor integrates an M-BRNN and an AL algorithm. In M-BRNN, the wavelet transform is employed to decompose the original traffic signals into several simple traffic signals. A BRNN is then used to predict each decomposed traffic signal. An AL algorithm is also applied in order to improve the learning process at each resolution level in M-BRNN-AL. Experiments and results on a set of Ethernet network traffic predictions show that the proposed scheme converges faster and archives better prediction performance than the other conventional models such as the Multi-layer Perception Type Neural Network, BRNN, and the original M-BRNN in terms of the normalized mean square error. [ABSTRACT FROM AUTHOR]

This paper deals with an indoor propagation problem where it is difficult to rigorously obtain the field strength distribution. We have developed a propagation model based on a neural network, which has advantages of deterministic (high accuracy) and empirical (short computation time) approaches. The neural network architecture, based on the multilayer perception, is used to absorb the knowledge about the given environment through training based on measurements. Such network then becomes capable to predict signal strength that includes absorption and reflection effects without additional computation and measurement efforts. The neural network model is used as a cost function in the optimization of the base station location. As optimization algorithm we have applied the particle swarm optimization (PSO) algorithm, i.e. a global optimization routine based on the movement of particles and their intelligence. Appropriate PSO parameters are discussed in the paper, and the results of PSO are compared with results obtained with two standard algorithms such as simplex optimization method and Powell's conjugate direction method. [ABSTRACT FROM AUTHOR]

The main idea of this paper is to neural network controller and intelligent agents to create an algorithm and coordination mechanism for climate parameters control to save electrical energy and keep high comfort level in the object. Interest is concentrated on the climate parameters optimization in passengers' interior of public electric transport vehicles. The article presents mathematical problem using neural network controller in mechatronic systems for climate parameters optimal control. The methods of the problem solving and structure of problem solving algorithm are given in the article. A special interest for investigations and further development is devoted to intelligent heating, ventilation and air condition systems allowing more flexible regulation of the system's compressor, fan and heater operation, and, therefore, improvement of efficiency and energy saving. In the paper there is described elaborated control system computer model neural network controller support. [ABSTRACT FROM AUTHOR]

This paper presents a new approach, known as symmetry axis based feature extraction and recognition (SAFER), for recognizing objects under translation, rotation and scaling. Unlike most previous invariant object recognition (IOR) systems, SAFER puts emphasis on both simplicity and accuracy of the recognition system. To achieve simplicity, it uses simple formulae for extracting invariant features from an object. The scheme used in feature extraction is based on the axis of symmetry and angles of concentric circles drawn around the object. SAFER divides the extracted features into a number of groups based on their similarity. To improve the recognition performance, SAFER uses a number of neural networks (NNs) instead of single NN are used for training and recognition of extracted features. The new approach, SAFER, has been tested on two of real world problems i.e., English characters with two different fonts and images of different shapes. The experimental results show that SAFER can produce good recognition performance in comparison with other algorithms. [ABSTRACT FROM AUTHOR]

This paper presents the development procedure of the feature extraction and classification module of an intelligent sorting system for electronic components. This system was designed as a prototype to recognise six types of electronic components for the needs of the educational electronics laboratories of the Kavala Institute of Technology. A list of features that describe the morphology of the outline of the components was extracted from the images. Two feature selection strategies were examined for the production of a powerful yet concise feature vector. These were correlation analysis and an implementation of support vector machines. Moreover, two types of neural classifiers were considered. The multilayer perceptron trained with the back-propagation algorithm and the radial basis function network trained with the K-means method. The best results were obtained with the combination of SVMs with MLPs, which successfully recognised 92.3% of the cases. [ABSTRACT FROM AUTHOR]

Mid-term load forecasting (MTLF) becomes an essential tool for today power systems, mainly in those countries whose power systems operate in a deregulated environment. Among different kinds of MTLF, this paper focuses on the prediction of daily peak load for one month ahead. This kind of load forecast has many applications like maintenance scheduling, mid-term hydro thermal coordination, adequacy assessment, management of limited energy units, negotiation of forward contracts, and development of cost efficient fuel purchasing strategies. However, daily peak load is a nonlinear, volatile, and nonstationary signal. Besides, lack of sufficient data usually further complicates this problem. The paper proposes a new methodology to solve it, composed of an efficient data model, preforecast mechanism and combination of neural network and evolutionary algorithm as the hybrid forecast technique. The proposed methodology is examined on the EUropean Network on Intelligent TEchnologies (EUNITE) test data and Iran’s power system. We will also compare our strategy with the other MTLF methods revealing its capability to solve this load forecast problem. [Copyright &y& Elsevier]

The neural networks are successfully applied to many applications in different domains. However, due to the results made by the neural networks are difficult to explain the decision process of neural networks is supposed as a black box. The explanation of reasoning is important to some applications such like credit approval application and medical diagnosing software. Therefore, the rule extraction algorithm is becoming more and more important in explaining the extracted rules from the neural networks. In this paper, a decompositional algorithm is analyzed and designed to extract rules from neural networks. The algorithm is simple but efficient; can reduce the extracted rules but improve the efficiency of the algorithm at the same time. Moreover, the algorithm is compared to the other two algorithms, M-of-N and Garcez, by solving the MONK’s problem. [ABSTRACT FROM AUTHOR]

In this paper, a new algorithm is proposed for resolution enhancement in hyperspectral images (HSIs). The key techniques are included: spectral unmixing and superresolution mapping, by which spatial and spectral information of HSIs is substantially fused. The proposed algorithm first represents each pixel in scene as a linear combination of landcover spectra and noise. Then, a fully constrained least squares algorithm is used to obtain the proportion of each landcover in each pixel, i.e., abundance, subjecting to two constraints: nonnegativity and sum-to-one. After that, superresolution mapping is performed on high-resolution grids according to spectral unmixing abundances of each landcover and following spatial correlation of clutters. Thus, by reasonably integrating spatial and spectral information of landcovers in HSIs, the proposed algorithm realizes resolution enhancement of the HSIs based on a back-propagation neural network. The proposed algorithm is independent from the a priori information associated with original HSIs, i.e., a main merit of the algorithm. In order to evaluate the performance of the new algorithm, numerical experiments are conducted on both simulated images and real HSIs collected by the Airborne Visible/Infrared Imaging Spectrometer. The proposed algorithm is compared with the traditional method in the experiments. The experimental results prove that the proposed algorithm effectively enhances the resolution of HSIs and indicate its applicability. [ABSTRACT FROM AUTHOR]

Abstract: Influenced by detector materials’ non-uniformity, growth and etching techniques, etc., every detector’s responsivity of infrared focal plane arrays (IRFPA) is different, which results in non-uniformity of IRFPA. And non-uniformity of IRFPA generates fixed pattern noises (FPN) that are superposed on infrared image. And it may degrade the infrared image quality, which greatly limits the application of IRFPA. Non-uniformity correction (NUC) is an important technique for IRFPA. The traditional non-uniformity correction algorithm based on neural network and its modified algorithms are analyzed in this paper. And a new improved non-uniformity correction algorithm based on neural network is proposed in this paper. In this algorithm, the desired image is estimated by using three successive images in an infrared sequence. And blurring effect caused by motion is avoided by applying implicit motion detection and edge detection. So the estimation image is closer to real image than the estimation image estimated by other algorithms, which results in fast convergence speed of correction parameters. A comparison is made to these algorithms in this paper. And experimental results show that the algorithm proposed in this paper can correct the non-uniformity of IRFPA effectively and it prevails over other algorithms based on neural network. [Copyright &y& Elsevier]

It is difficult to determine the recognition mechanism in the neurons of a neural network trained for pattern recognition due to the non-linear nature of neural networks. This paper describes a recognition mechanism of a four-layer back propagation neural network using Alopex algorithm. We have developed a small-scale, four-layered neural network model for simple character recognition, which can recognize the patterns transformed by affined conversion. Alopex algorithm is an iterative and stochastic processing method, which was proposed for optimization of a given cost function. In this case the receptive fields of the neurons in the output layers are obtained using the Alopex algorithm. [ABSTRACT FROM AUTHOR]

This paper proposes a new passive robust fault detection scheme using non-linear models that include parameter uncertainty. The non-linear model considered here is described by a group method of data handling (GMDH) neural network. The problem of passive robust fault detection using models including parameter uncertainty has been mainly addressed by checking if the measured behaviour is inside the region of possible behaviours based on the so-called forward test since it bounds the direct image of an interval function. The main contribution of this paper is to propose a new backward test, based on the inverse image of an interval function, that allows checking if there exists a parameter in the uncertain parameter set that is consistent with the measured system behaviour. This test is implemented using interval constraint satisfaction algorithms which can perform efficiently in deciding if the measured system state is consistent with the GMDH model and its associated uncertainty. Finally, this approach is tested on the servoactuator being a FDI benchmark in the European Project DAMADICS. [Copyright &y& Elsevier]

In this paper, a novel human visual system (HVS)-directed neural-network-based adaptive interpolation scheme for natural image is proposed. A fuzzy decision system built from the characteristics of the HVS is proposed to classify pixels of the input image into human perception nonsensitive class and sensitive class. Bilinear interpolation is used to interpolate the nonsensitive regions and a neural network is proposed to interpolate the sensitive regions along edge directions. High-resolution digital images along with supervised learning algorithms are used to automatically train the proposed neural network. Simulation results demonstrate that the proposed new resolution enhancement algorithm can produce a higher visual quality for the interpolated image than the conventional interpolation methods. [ABSTRACT FROM AUTHOR]

This paper presents the recognition of speech commands using a modified neural fuzzy network (NFN). By introducing associative memory (the tuner NFN) into the classification process (the classifier NFN), the network parameters could be made adaptive to changing input data. Then, the search space of the classification network could be enlarged by a single network. To train the parameters of the modified NFN, an improved genetic algorithm is proposed. As an application example, the proposed speech recognition approach is implemented in an eBook experimentally to illustrate the design and its merits. [ABSTRACT FROM AUTHOR]

This paper presents an analysis of a random access memory (RAM)-based associative memory which uses a weighted voting scheme for information retrieval. This weighted voting memory can operate in heteroassociative or autoassociative mode, can store both real-valued and binary-valued patterns and, unlike memory models, is equipped with a rejection mechanism. A theoretical analysis of the performance of the weighted voting memory is given for the case of binary and random memory sets. Performance measures are derived as a function of the model parameters: pattern size, window size, and number of patterns in the memory set. It is shown that the weighted voting model has large capacity and error correction. The results show that the weighted voting model can successfully achieve high-detection and -identification rates and, simultaneously, low-false-acceptance rates. [ABSTRACT FROM AUTHOR]

This paper presents the use of a hybrid algorithm based on the artificial neural networks (ANN) and the genetic algorithms (GA) for the identification of the parameters of the modified Lorentz function from the measured cycles. The principle of the identification is based on the minimization of a function criterion which represents the difference between the measured and simulated cycles. [ABSTRACT FROM AUTHOR]

This paper presents the design and stability analysis of a sampled-data neural-network-based control system. A continuous-time nonlinear plant and a sampled-data three-layer fully connected feedforward neural-network-based controller are connected in a closed loop to perform the control task. Stability conditions will be derived to guarantee the closed-loop system stability. Linear-matrix-inequality- and genetic-algorithm-based approaches will be employed to obtain the largest sampling period and the connection weights of the neural network subject to the considerations of the system stability and performance. An application example will be given to illustrate the design procedure and effectiveness of the proposed approach. [ABSTRACT FROM AUTHOR]

Abstract: In this paper, we present a state-space formulation of a neural-network-based hidden dynamic model of speech whose parameters are trained using an approximate EM algorithm. This efficient and effective training makes use of the output of an off-the-shelf formant tracker (for the vowel segments of the speech signal), in addition to the Mel-cepstral observations, to simplify the complex sufficient statistics that would be required in the exact EM algorithm. The trained model, consisting of the state equation for the target-directed vocal tract resonance (VTR) dynamics on all classes of speech sounds (including consonant closure and constriction) and the observation equation for mapping from the VTR to Mel-cepstral acoustic measurement, is then used to recover the unobserved VTR based on the extended Kalman filter. The results demonstrate accurate estimation of the VTR, especially during rapid consonant–vowel or vowel–consonant transitions and during consonant closure when the acoustic measurement alone provides weak or no information to infer the VTR values. The practical significance of correctly identifying the VTRs during consonantal closure or constriction is that they provide target frequency values for the VTR or formant transitions from adjacent sounds. Without such target values, the VTR transitions from vowel to consonant or from consonant to vowel are often very difficult to extract accurately by the previous formant tracking techniques. With the use of the new technique reported in this paper, the consonantal VTRs and the related transitions become more reliably identified from the speech signal. [Copyright &y& Elsevier]

This paper presents a new empirical model and a two-layer neural network approach for the determination of optimum body diameter (OBD) of air cyclones. OBD values were calculated by help of a MATLAB® algorithm for 505 different artificial scenarios given in a wide range of five main operating variables. The predicted results obtained from each proposed approach were compared with the wellknown Kalen and Zenz's model. The computational analysis showed that the empirical model and neural network outputs obviously agreed with the Kalen and Zenz's model, and all the predictions proved to be satisfactory, with a correlation coefficient of about 0.9998 and 1, respectively. The maximum diameter deviations from Kalen and Zenz's model were recorded as only ±1.3 cm and ± 0.0022 cm for the proposed model and NN outputs, respectively. In addition to proposed approaches, the pressure drop problem was controlled using a MATLAB® algorithm, and results were obtained rapidly and practically for varying data used in the cyclone design. [ABSTRACT FROM AUTHOR]

In this paper, based on genetic algorithm (GA) and steepest descent method (SDM), we proposed a hybrid algorithm for the learning of neural network to identify the chaotic systems. The chaotic systems considered in this paper are the Duffing equation. An identification scheme of neural network trained by the proposed hybrid algorithm is used to identify the Duffing equation. Simulation results show that performance of the hybrid algorithm proposed in this paper is more efficient than those of other methods. [ABSTRACT FROM AUTHOR]

This paper analyzes the sensitivity to noise in BAM (Bidirectional Associative Memory), and then proves the noise immunity of BAM relates not only to the minimum absolute value of net inputs (MAV) but also to the variance of weights associated with synapse connections. In fact, it is a positive monotonically increasing function of the quotient of MAV divided by the variance of weights. Besides, the performance of pseudo-relaxation method depends on learning parameters (λ and ξ), but the relation of them is not linear. So it is hard to find a best combination of λ and ξ which leads to the best BAM performance. And it is obvious that pseudo-relaxation is a kind of local optimization method, so it cannot guarantee to get the global optimal solution. In this paper, a novel learning algorithm EPRBAM (evolutionary psendo-relaxation learning algorithm for bidirectional association memory) employing genetic algorithm and pseudo-relaxation method is proposed to get feasible solution of BAM weight matrix. This algorithm uses the quotient as the fitness of each individual and employs pseudo-relaxation method to adjust individual solution when it does not satisfy constraining condition any more after genetic operation. Experimental results show this algorithm improves noise immunity of BAM greatly. At the same time, EPRBAM does not depend on learning parameters and can get global optimal solution. [ABSTRACT FROM AUTHOR]

The algorithm to derive the concentrations of coastal (case 2) water constituents from the Medium Resolution Imaging Spectrometer (European Space Agency satellite ENVISAT) is based on neural network (NN) technology. The NN not only transforms water leaving radiance reflectances with high efficiency into concentrations but also checks if its input is in the domain of reflectance spectra which were simulated for the training of the NN. Two NNs are trained with simulated reflectances: 1) invNN to emulate the inverse model (reflectances, geometry) → concentrations and 2) forwNN to emulate the forward model (concentrations, geometry) → reflectances. The invNN is used to obtain an estimate of the concentrations. These concentrations are fed into the forwNN, and the derived reftectances are compared with the measured reflectances, Deviations above a threshold are flagged. The paper describes a further improvement: the result obtained by invNN is used as a first guess to start a minimization procedure, which uses the forwNN iteratively to minimize the difference between the calculated reflectances and the measured ones. The procedure is very fast as it takes advantage of the Jacobian which is a byproduct of the NN calculation. [ABSTRACT FROM AUTHOR]

Principal component analysis (PCA) by neural networks is one of the most frequently used feature extracting methods. To process huge data sets, many learning algorithms based on neural networks for PCA have been proposed. However, traditional algorithms are not globally convergent. In this paper, a new PCA learning algorithm based on cascade recursive least square (CRLS)neural network is proposed. This algorithm can guarantee the network weight vector converges to an eigenvector associated with the largest eigenvalue of the input covariance matrix globally. A rigorous mathematical proof is given. Simulation results show the effectiveness of the algorithm. [ABSTRACT FROM AUTHOR]

This paper proposes an information system that classifies Web pages according a taxonomy, which is mainly used from seven search engines/directories. The proposed classifier is a four-layer generalised regression neural network (GRNN) that aims to perform the information segmentation according to information filtering techniques using content descriptor vectors. Eight categories of Web pages were used in order to evaluate the robustness of the method, while no restrictions were imposed except for the language of the content, which is English. The system can be used as an assistant and consultative tool for classification purposes as well as for estimating the population of Web pages at any given point in time. [ABSTRACT FROM AUTHOR]

In this paper, we present a new neural network (NN) for three-dimensional (3D) shape reconstruction. This NN provides an analytic mapping of an initial 3D polyhedral model into its projection depth images. Through this analytic mapping, the NN can analytically refine vertices position of the model using error back-propagation learning. This learning is based on shape-from-shading (SFS) depth maps taken from multiple views. The depth maps are obtained by Tsai–Shah SFS algorithm. They are considered as partial 3D shapes of the object to be reconstructed. The task is to reconstruct an accurate and complete representation of a given object relying only on a limited number of views and erroneous SFS depth maps. Through hierarchical reconstruction and annealing reinforcement strategies, our reconstruction system gives more exact and stable results. In addition, it corrects and smoothly fuses the erroneous SFS depth maps. The implementation of this neural network algorithm used in this paper is available at http://kumazawa-www.cs.titech.ac.jp/~fanany/MV-SPRNN/mv-sprnn.html. [Copyright &y& Elsevier]

We employ statistical dynamics to study the convergence of the Wake–Sleep (W-S) algorithm, which is a learning algorithm for neural network models having hidden units. Although there have been several reports on the effectiveness of the W-S algorithm based on experimental methods, the theoretical side is not clear even for a simple network. In this paper, we investigate the dynamic characteristics of the W-S algorithm expressed by a single factor analysis problem, which is the simplest state setting. The advantage of our approach is the ability to quantitatively evaluate the effect that the learning coefficients have on the convergence, which is difficult when using other methods. The result was that the settings of the learning coefficients, particularly in the Sleep step, had a substantial effect on the convergence of the algorithm. © 2001 Scripta Technica, Electron Comm Jpn Pt 3, 85(1): 41–49, 2002 [ABSTRACT FROM AUTHOR]

Morse code is a simple, speedy and low cost means of communication composed of a series of dots, dashes and space intervals. Each tone element (either a dot, dash or space interval) is transmitted by sending a signal for a defined length of time. This poses a challenge as the automatic recognition of Morse code is dependent upon maintaining a stable typing rate. In this paper, a suitable adaptive automatic recognition method, combining the least-mean-square (LMS) algorithm with a neural network, was applied to this problem. The method presented in this paper is divided into five modules: space recognition, tone recognition, learning process, adaptive processing and character recognition. Statistical analyses demonstrated that the proposed method elicited a better recognition rate in comparison with other methods in the literature. [ABSTRACT FROM AUTHOR]

Investigations into quantum computations have begun from the pioneering theoretical studies of Feynman, Deutsch, and others, and detailed studies have been done since the discovery of a quantum algorithm which can solve the problem of factorizing a large integer in polynomial time by Shor in 1994. Recently, the existence of nonalgorithmic quantum computations in microtubules inside a neural circuit has been debated, resulting in the proposal of the concept of the quantum neural computation theory, although detailed studies have not as yet been made. In this paper, in order to construct a new framework for describing the cohesiveness of the distribution and synthesis inherent in a neural network, a neural state is described quantum dynamically and a qubitlike neural network corresponding to the quantum circuit of quantum computations is studied. Specifically, a qubitlike neural network is constructed for a 3-bit quantum circuit, which is the minimum quantum logical gate describing all basic logical operations, and in this model we investigate how to determine circuit parameters by learning. © 2000 Scripta Technica, Electron Comm Jpn Pt 3, 83(10): 67–73, 2000 [ABSTRACT FROM AUTHOR]

The authors previously reported on a variable logic circuit using a neural network (NN), in which logic operation was controlled by an analog voltage. In that design, the synaptic weights were determined by the error back-propagation (BP) algorithm. Then, logic outputs outside the specification were produced between logic outputs with Hamming distances of two or more. This paper proposes a design of NN in which no logic outside the specification is output, together with a method of calculating the synaptic weights. Based on the correspondence between the change of the output logic and the control voltage, the necessary number of hidden-layer neurons can be defined. The computation time for the synaptic weights is reduced. By constructing an NN circuit by the proposed method, it is verified that the desired operation is realized. © 1999 Scripta Technica, Electron Comm Jpn Pt 3, 83(1): 14–20, 2000 [ABSTRACT FROM AUTHOR]

Short-term traffic prediction is a key component of Intelligent Transportation Systems. It uses historical data to construct models for reliably predicting traffic state at specific locations in road networks in the near future. Despite being a mature field, short-term traffic prediction still poses some open problems related to the choice of optimal data resolution, prediction of nonrecurring congestion, and the modelling of relevant spatiotemporal dependencies. As a step towards addressing these problems, this paper investigates the ability of Artificial Neural Networks, Random Forests, and Support Vector Regression algorithms to reliably model traffic flow at different data resolutions and respond to unexpected traffic incidents. We also explore different feature selection methods to identify and better understand the spatiotemporal attributes that most influence the reliability of these models. Experimental results indicate that data aggregation does not necessarily achieve good performance for multivariate spatiotemporal machine learning models. The models learned using high-resolution 30-second input data outperformed the corresponding baseline ARIMA models by 8%. Furthermore, feature selection based on Recursive Feature Elimination resulted in models that outperformed those based on linear correlation-based feature selection., 1. Introduction Traffic congestion results in significant monetary losses in countries around the world, with the cost of traffic congestion in 2014 estimated to be $160 billion in the US [...]

Wheel and rail wear is a significant issue in railway systems. Accurate prediction of this wear can improve economy, ride comfort, prevention of derailment and planning of maintenance interventions. Poor prediction can result in failure and consequent delay and increased costs if it is not controlled in an effective way. However, prediction of wheel and rail wear is still a great challenge for railway engineers and operators. The aim of this paper is to predict wheel wear and rail wear using an artificial neural network. Nonlinear Autoregressive models with exogenous input neural network (NARXNN) have been developed for wheel and rail wear prediction. Testing with a twin disc rig, together with measurement of wear using replica material and a profilometer have been carried out for wheel and rail wear under dry, wet and lubricated conditions and after sanding. Tests results from the twin disc rig have been used to train, validate, and test the neural network. Wheel and rail profiles plus load, speed, yaw angle, and first and second derivative of the wheel and rail profiles were used as an inputs to the neural network, while the output of neural network was the wheel wear and rail wear. Accuracy of wheel and rail wear prediction using the neural network was investigated and assessed in term of mean absolute percentage error (MAPE). The results demonstrate that the neural network can be used efficiently to predict wheel and rail wear. The methods of collecting wear data using the replica material and the profilometer have also proved effective for wheel and rail wear measurements for training and validating the neural network. The laboratory tests have aimed to validate the wear predictions for realistic wheel and rail profiles and materials but they necessarily cover only a limited set of conditions. The next steps for this work will be to test the methods for rail and wheel data from field tests. [ABSTRACT FROM AUTHOR]

Industrial asphalt foaming is a process of mixing, transferring heat and phase transition of multi-phase multi-component flow in the expansion chamber, including a series of complex problems such as difficult analysis and observation. In the context of this research, asphalt foaming quality control model using neural network was constructed to map the pertinent parameters (asphalt temperature, water content and air pressure) into the foaming properties (maximum expansion ratio and half-life time). In addition, Particle swarm optimization was adopted to avoid the local infinitesimal defect and slow constringency in the traditional BP algorithm. The prediction error of Shell 60/70 quality control model indicates that the asphalt foaming quality control model using PSO-BP neural network is effective, and this provides a novel idea for studying of the foam asphalt, which has practical significance for the design of asphalt foam equipment. Finally, a parameter optimization model with the maximization of the foaming index was proposed based on the quality control model to improve the foaming performance of the asphalt foaming equipment. With the analysis of a parameter optimization model of Shell 60/70 asphalt, the better foaming properties are the maximum expansion ratio of 12.28 times and half-life time of 11.02 s at the foaming condition of asphalt temperature of about 168 °C, the water content of about 1.5 wt% and the air pressure of about 1 bar. The method proposed in this paper is of important reference significance for the engineering with difficult analysis and observation. [ABSTRACT FROM AUTHOR]

Studies on interval type-2 TSK fuzzy logic system is a hot topic in the current academic area. Parameter identification is very important for system design. Commonly used parameter identification methods are the least-square algorithm, BP algorithm, etc., few scholars use QPSO algorithm for parameter identification. In this paper, we propose a design of interval type-2 TSK fuzzy logic system based on quantum behaved particle swarm optimization (QPSO) intelligent algorithm. Firstly, by combining the A1-C1, A2-C0, A2-C1 interval type-2 TSK fuzzy logic system with neural network, the fuzzy neural network system is designed. Then, the QPSO intelligent algorithm was used to tune the fuzzy neural network system parameters, and the designed system model is applied to predict the Nasdaq Composite Index and International Gold Prices. Both QPSO algorithm and BP algorithm have been used to train the system model. By comparing QPSO algorithm and BP algorithm, the performance index and simulation results illustrated the proposed model is effective and feasible, which can achieve a better performance. Finally, compare the performance of the four fuzzy logic systems, it can be seen the effect of A2-C1 fuzzy logic system is better than that of the other three fuzzy logic systems. [ABSTRACT FROM AUTHOR]

The traditional RSA information encryption algorithm uses one-dimensional chaotic equations to generate pseudo-random sequences that meet the encryption requirements. This encryption method is too simple and the security performance is poor. A multi-level encryption algorithm for user-related information across social networks is proposed, and a user association model across social networks is constructed to obtain user-related information across social networks. This multi-level chaotic encryption algorithm based on neural network is used to select three different chaotic mapping models based on user-related information, and a multi-level chaotic encryption algorithm is designed. According to the characteristics of error sensitivity of chaotic system, the neural network is used to inversely propagate the error. A chaotic encryption algorithm that implements multi-level encryption of user-related information across social networks is optimized. The experimental results show that the average rate for which the proposed algorithm correctly identified the user-related information across social networks was 97.6%, the highest frequency of average character distribution probability in cipher text was 0.021, and the average time for encryption was 18.45 Mbps. The average time for decryption was 21.90Mbps. [ABSTRACT FROM AUTHOR]

This paper presents a method for designing neural networks using a genetic algorithm (GA) with deterministic mutation (DM) based on learning. The GA presented in this paper has a large framework including DM, which is performed on the basis of the results from neural network learning. It can achieve better convergence properties than traditional GAs. This framework is an evolutional system based on mutual interaction between DM and traditional genetic operators including stochastic mutation. It is also a model of transcription and reverse transcription in DNA. We show that the present method is better than conventional GAs with respect to convergence in learning. © 1998 Scripta Technica. Syst. Comp Jpn, 29(3): 10–17, 1998 [ABSTRACT FROM AUTHOR]

Kohomen's model as the self-organizing model for the neural network can be considered as a kind of adaptive vector quantization algorithm. Numerous reports have been presented on the application of the model to practical problems. Although some results have been presented for the theoretical properties of Kohonen's model, many properties remain to be clarified. Among various properties of Kohonen's model as an adaptive vector quantization algorithm, this paper considers the problem of how the reference vectors are placed according to the probability distribution of the input signal. Considering the limit where the number of reference vectors is increased to infinity, this problem can be discussed theoretically as the distribution of the reference vectors. Due to the effect of the "learning by neighborhood," which is the feature of Kohonen's model, the property of the Kohonen model differs quantitatively from the property of the ordinary vector quantization algorithm. This paper discusses quantitatively the properties of Kohonen's model using the average learning equation. [ABSTRACT FROM AUTHOR]