The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
Power system static security assessment is one of the most important problems which relate power system secure-stable performance. Static security can be rapidly assessed using the artificial intelligence technology. This paper compares the advantages and disadvantages of Artificial Neural Network (ANN) and Support Vector Machines (SVM) and then selects the SVM algorithm. A new multi-classification...
This paper established a back propagation (BP) neural network tandem cold rolling force prediction model, and optimized by genetic particle swarm algorithm (GPSA). Genetic particle swarm algorithm has the advantage of both genetic algorithm (GA) and particle swarm algorithm (PSO) algorithm, integrates global searching ability with high convergence speed. Taking neural network weights and threshold...
After studying the disadvantage of BP neural network which has low convergent speed and trap into local minima easily, an idea of designing a new hybrid neural network model. By using Artificial Bee Colony Algorithm (ABC) to expand the updated space of weight and using the fitness functions to decide the better weight. On the basis, make the acquired better value as the weight of BP neural network...
Adaptive Neuro-Fuzzy Inference System (ANFIS) has become a popular tool in neuro-fuzzy modeling. However, since it includes many parameters needed to be set, its designing process is a complicated and time-intensive task for experimenters. To tackle this problem, in this paper we implement the Design of Experiment (DOE) technique to identify the significant parameters of ANFIS when it applies to the...
An important issue in design and implementation a neural network is that perturbations of training pattern pairs may cause some disadvantages to outputs. How the perturbations of training pattern pairs in Morphological Bidirectional Associative Memories (MBAMs) influence on the outputs is discussed in this paper. We define the outputs' max error to evaluate the robustness of the MBAMs. The related...
This paper presents three intelligent methods for condition monitoring of induction motors in real-time. A structured neural network has been designed to prognosis of instantaneous faults. The inputs of neural network are the standard deviation and mean of feature signal obtained by Hilbert transform of one phase current signal. The stator related faults have been diagnosed by designing fuzzy logic...
This paper takes a kind of ceramic glaze as an example, and builds an improved BP neural network model for optimizing the formulation on ceramic glaze. The improved BP neural network adopts Levenberg-Marquardt algorithms. The paper reviews how to build the ceramic formulation optimization model based on BP artificial neural network, including the establishment of neural network, the training, and...
A novel model of fuzzy clustering neural network is discussed, which synthesizes unsupervised fuzzy competitive learning algorithm and self-organized competitive network. Based on this model, an algorithm of abrupt video shot boundary detection is presented which is a two-stage clustering on a linear feature space. The experimental results obtained demonstrate that the algorithm is feasible and efficient.
Paper focuses on developing an algorithm using a Hamming Distance Classifier in Neural Networks to find the most optimal move to be made in the Tic-Tac-Toe problem such that the game always ends in a win or a draw. The basic step involves an eight-class Hamming network which has nine inputs corresponding to each cell of the grid and eight outputs respectively. The algorithm computes the Hamming Distance...
For the research of Chinese word segmentation, the BP algorithm model has a lot of defects such as low convergent velocity, easily falling into local minimum, low velocity and efficiency. In this paper, we proposed a new particle swarm neural network algorithm (NPSO-BP), and used it in Chinese word segmentation. The results show that the speed of the segmentation algorithm is obviously faster than...
This paper deals with a new approach for complex systems modeling and control based on neural and fuzzy clustering algorithms. It aims to derive a base of local models describing the system in the whole operating domain. The implementation of this approach requires three main steps: 1) determination of the structure of the model-base, the number of models are found out by using Rival Penalized Competitive...
Minor component analysis (MCA) is an important feature extraction technique which has been widely applied in data analysis fields. MCA neural networks generally are used to extract online minor component in term of adapting the demands of real time and decreasing computational complexity. However, the MCA learning algorithm can produce complicated dynamical behavior under some conditions, such as...
A diagnosis method basing on neural network classifier, genetic algorithm (GA) and wavelet transform is proposed for a pulse width modulation voltage source inverter. It is used to detect and identify the transistor open-circuit fault. BP neural network (BPNN) is capable of recognition. However, it has shortcomings obviously. These are just advantages of GA, which has ability of global search. Thus...
Process neural networks (PNN) can only receive time-varying continuous functions, can not receive discrete samples. To solve this problem, a training algorithm of PNN based on piecewise linear interpolation function is proposed. First the discrete data of both sample functions and weight functions are transformed to piecewise linear functions, and then the integrals of product of two linear functions...
Standard BP neural network is a most representative algorithm in the neural network model. But shortcomings exist in its process of application. For example: it's hard to reach global optima, but can easily form local minimum, Low study efficiency and slow convergence rate appear because of the excessive training, the selection of the hidden layer nodes lack of theoretical guidance, in training, there...
The resource-constrained multi-project scheduling (RCMPS) is a NP-hard problem and has been extensively used in manufacturing and engineering fields. In order to solve scheduling of RCMPS, a new algorithm was presented in this paper. The new algorithm combines the some advantages of ACOA and NN . Finally, the algorithm was tested on a case of the RCMPS and the results were presented in the paper....
Due to the disadvantages of the ant algorithm used in the combining optimization in continuous space and the demerit of BP algorithm being vulnerable in the local optimum, the dynamics model of chaos ant colony has been introduced into the optimization of weights in neural network model. Therefore, the chaos ant colony neural network can have both extensive mapping ability of neural network and rapid...
In order to give the computer the ability to play against human opponents, one could utilize the Alpha-Beta algorithm. However, this algorithm has several limitations restricting its playing capabilities. Over the years, many variants of this algorithm were developed, among them a couple that make use of neural networks: a neural network to focus the search in the game tree, and a neural network trained...
This paper investigates the fault-tolerance ability of complex-values neural networks (CVNNs) in classification applications. An analysis of the effect of weight loss at the units (neurons) level revealed that the loss of weight in complex neural networks is more critical than in real valued neural networks. A novel weight decay technique for fault tolerance of real-valued neural networks (RVNNs)...
Generalization capability is a key flag to evaluate the performance of a learning system. Neural network ensemble can greatly improve the generalization capability of a learning system by training many neural networks and composing the result of them. In this paper, based on the theory of neural network ensemble, we present a constructive algorithm to improve the generalization capability of coverage-based...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.