Papers by Keyword: Probabilistic Neural Network

Abstract: In order to implement the recognition automation of rock section pore images, a method combined K-means clustering with probabilistic neural network is proposed and applied to rock thin section images. Firstly, K-means clustering is used as segmentation algorithm, the rock images are divided into two types and extracted enough features and it is shown good classification recognition effect on testing dataset. Secondly, 100 pieces of rock image section are used as validation dataset, including 20 groups, each group has 5 images and 200 data samplings. Experiments show that the probabilistic neural network can be used as rock texture classifier, the average correct classification rate is around 95.12%, which can meet the practical application needs.

Abstract: This paper presents development of an automatic fault diagnosis system in the nuclear power plants to minimize possible nuclear disasters. Probabilistic Neural Network (PNN) overcame the shortcomings of entrapment in local optimum, slow convergence rate which was in BP algorithm. PNN is fit to diagnose the fault of nuclear power plant and has auto-adaptability. The method not only makes the original neural network smaller in terms of computation and realization, but also improves diagnosis speed and accuracy of nuclear power plant.

Abstract: During the study of English sentence pronunciation evaluation system, we found that sentence pronunciation emotion and intonation evaluation are very important. Probabilistic neural network has been used to study English sentence pronunciation emotion, and DTW (Dynamic Time Warping) algorithm has been used in the intonation analysis. The probability neural network basic principle is introduced in this paper. An emotion recognition algorithm based on MFCC（Mel　Frequency　Cepstrum　Coefficient）is present. The keynote and energy of the sentences are used to analyse the accuracy of the tones.　The experimental results of the proposed method effectiveness are given.

Abstract: The network model of probabilistic neural network and its method of pattern classification and discrimination are first introduced in this paper. Then probabilistic neural network and three usually used back propagation neural networks are established through MATLAB7.0. The pattern classification of dots on a two-dimensional plane is taken as an example. Probabilistic neural network and improved back propagation neural networks are used to classify these dots respectively. Their classification results are compared with each other. The simulation results show that compared with back propagation neural networks, probabilistic neural network has simpler learning rules, faster training speed and it needs fewer training samples; the pattern classification method based on probabilistic neural network is very effective, and it is superior to the one based on back propagation neural networks in classifying speed, accuracy as well as generalization ability.

Abstract: Using feature selection and neural networks to experiment the data, then we bring a warning model of user complaints. It is the core that using the known information of network index sample to analyze and discriminate. First, the training samples need to be extract, because there are too many features in training data will have an adverse impact on machine learning classification algorithm. Using extraction method to explore the feature subset with feature, feature subset is a set of feature vectors, then the feature vectors are input into the probability neural network prediction, find out the best features quantum set. This model can be achieved using the MATLAB software, and it is operational, and it can be extended to the network quality assessment and monitoring practice.

Abstract: During long-time charge-discharge cycling, the capacity of the vanadium redox flow battery (VRFB) will reduce gradually. To recognize the capacity loss condition in the time of the operation process, a method for classification of the capacity loss degree based on Probabilistic Neural Network (PNN) is presented. The network inputs are the value of the voltage per second and the average power of the cell stack in any two minutes of the circulation. The network will give out three classes in form of three numbers to classify the capacity loss degree into different levels. The network is trained and validated by experimental data and the results show that the network is suitable for the classification problem of VRFB capacity loss and the method is useful to determine whether the capacity is sufficient and when to restore the cell capacity in real time.

Abstract: In fault diagnosis of three-phase induction motors, traditional methods usually fail because of the complex system of three-phase induction motors. Short circuit is a very common stator fault in all the faults of three-phase induction motors. Probabilistic neural network is a kind of artificial neural network which is widely used due to its fast training and simple structure. In this paper, the diagnosis method based on probabilistic neural network is proposed to deal with stator short circuits. First, the principle and structure of probabilistic neural network is studied in this paper. Second, the method of fault setting and fault feature extraction of three-phase induction motors is proposed on the basis of the fault diagnosis of stator short circuits. Then the establishment of the diagnosis model based on probabilistic neural network is illustrated with details. At last, training and simulation tests are done for the model. And simulation results show that this method is very practical with its high accuracy and fast speed.

Abstract: Probabilistic neural network compared with the traditional BP neural network structure is simpler and it is faster to be identificated, so it is widely used in the field of pattern recognition. This paper is mainly focused on similar gesture recognition research, propose an probabilistic neural network gesture recognition algorithm. The simulation results show that the improved probabilistic neural network algorithm on the recognition rate and training time is better than the traditional BP network.

Abstract: Fuyu oil layer in Changyuan of Daqing is a strategic area for Daqing oil field to increase and stabilize production at present, but its reservoir is structured mostly by riverway sand bodies. The sand bodies are thin and change with complexity, so they are difficult to identify. Through PNN probability neural network, and combined with the seismic multi-attribute and earthquake frequency division data, wave impedance inversions are implemented, and the maximum correlation coefficient is 0.517 if a single attribute is described and it is raised to 0.773 by linear weighting, and also 9 independent attributes are used for PNN network training so that a more complex nonlinear relationship is obtained. Then, the correlation coefficient continues to increase until 0.869, so that the spatial distribution characteristics of the sandstones in the study area are accurately described. This indicates that the PNN neural network inversion based on the integration of seismic multi-attribute with frequency division RGB can play an obvious effect in the areas of thin and changing sand bodies, and can meet the needs of the oil field exploration and development.

Abstract: For a typical ECoG-based brain-computer interface system that the subjects task is to imagine movements of either the left small finger or the tongue, a feature extraction algorithm using wavelet variance was proposed. Firstly the wavelet transform was discussed, and the definition and significance of wavelet variance were bring out and taken as feature, 6 channels with most distinctive features were selected from 64 channels for analysis; consequently the EEG data were decomposed using db4 wavelet, the wavelet coefficients variances containing Mu rhythm and Beta rhythm were taken out as features based on ERD/ERS phenomenon, and the features were classified by probabilistic neural network under a optimal spread with an algorithm of cross validation. The result of off-line showed high average classification accuracies of 89.21% and 88% for training and test data were achieved, the wavelet variance has characteristics of more simple and effective and it is suitable for feature extraction in BCI research.