Artificial wavelet neural network and its application in neuro-fuzzy models

In the proposed work, two types of artificial neural networks are proposed by using well-known advantages and valuable features of wavelets and sigmoidal activation functions. Two neurons are derived by adding and multiplying the outputs of the wavelet and the sigmoidal activation functions. These neurons in a feed-forward single hidden layer network result summation wavelet neural network (SWNN) and multiplication wavelet neural network (MWNN). An algorithm is introduced for structure determination of the proposed networks. Approximation properties of SWNN and MWNN have been evaluated with different wavelet functions. The above networks in the consequent part of the neuro-fuzzy model result summation wavelet neuro-fuzzy (SWNF) and multiplication wavelet neuro-fuzzy (MWNF) models. Different types of wavelet function are tested with the proposed networks and fuzzy models on four different dynamical examples. Convergence of the learning process is also guaranteed by adaptive learning rate and performing stability analysis using Lyapunov function.     

一种网络流量预测的小波神经网络模型

结合小波变换和人工神经网络的优势,建立一种网络流量预测的小波神经网络模型。首先对流量时间序列进行小波分解,得到小波变换尺度系数序列和小波系数序列,以系数序列和原来的流量时间序列分别作为模型的输入和输出,构造人工神经网络并且加以训练。用实际网络流量对该模型进行验证,结果表明,该模型具有较高的预测效果。     

小波神经网络模型的改进方法

为了改善小波神经网络(WNN)在处理复杂非线性问题的性能,针对量子粒子群优化(QPSO)算法易早熟、后期多样性差、搜索精度不高的缺点,提出一种同时引入加权系数、引入Cauchy随机数、改进收缩扩张系数和引入自然选择的改进量子粒子群优化算法,将其代替梯度下降法,训练小波基系数和网络权值,再将优化后的参数组合输入小波神经网络,以实现算法的耦合。通过对3个UCI标准数据集的仿真实验表明,与WNN、PSO-WNN、QPSO-WNN算法相比,改进的量子粒子群小波神经网络(MQPSO-WNN)算法的运行时间减少了11%~43%,而计算相对误差较之降低了8%~57%。因此,改进的量子粒子群小波神经网络模型能够更迅速、更精确地逼近最优值。     

Efficient text independent speaker recognition with wavelet feature selection based multilayered neural network using supervised learning algorithm

A wavelet packet feature selection derived by using multilayered neural network for speaker identification is described. The concept of a multilayered neural network is without using a gradient method. First, the outputs of each hidden unit are algebraically determined by an error backpropagation method. Then, the weight parameters are determined by using an exponentially weighted least squares method. Our results have shown that this feature selection introduced better performance than the other methods with respect to the percentages of recognition.     

改进的HMM和小波神经网络的抗噪语音识别

通过MFFC计算出的语音特征系数,由于语音信号的动态性,帧之间有重叠,噪声的影响,使特征系数不能完全反映出语音的信息。提出一种隐马尔可夫模型（HMM）和小波神经网络（WNN）混合模型的抗噪语音识别方法。该方法对MFCC特征系数利用小波神经网络进行训练,得到新的MFCC特征系数。实验结果表明,在噪声环境下,该混合模型比单纯HMM具有更强的噪声鲁棒性,明显改善了语音识别系统的性能。     

A self-evolving functional-linked wavelet neural network for control applications

The structure of a neural network is determined by time-consuming trial-and-error tuning procedure in advance for the reason that it is difficult to consider the balance between the neuron number and the desired performance. To attack this problem, a self-evolving functional-linked wavelet neural network (SFWNN) is proposed. Without the need for preliminary knowledge, a self-evolving approach demonstrates that the properties of generating and pruning the hidden neurons automatically. Then, an adaptive self-evolving functional-linked wavelet neural control (ASFWNC) system which is composed of a neural controller and a supervisory compensator is proposed. The neural controller uses a SFWNN to online estimate an ideal controller and the supervisory compensator is designed to eliminate the effect of the approximation error introduced by the neural controller upon the system stability in the Lyapunov sense. To investigate the capabilities of the proposed ASFWNC approach, it is applied to a chaotic system and a DC motor. The simulation and experimental results show that favorable control performance can be achieved by the proposed ASFWNC scheme.     

A new watermarking approach based on probabilistic neural network in wavelet domain

A novel scheme of digital image watermarking based on the combination of dual-tree wavelet transform (DTCWT) and probabilistic neural network is proposed in this paper. Firstly, the original image is decomposed by DTCWT, and then the watermark bits are added to the selected coefficients blocks. Because of the learning and adaptive capabilities of neural networks, the trained neural networks can recover the watermark from the watermarked images. Experimental results show that the proposed scheme has good performance against several attacks.     

基于小波和神经网络相结合的股票价格模型

针对股票价格构成的时间序列具有随机性与偶然性,传统的单一模型很难满足建模要求的问题,提出一种基于小波和神经网络相结合的股票预测模型.将股票价格进行小波分解成尺度不同的分层数据,分别利用Elman神经网络预测各层数据,将各层的预测结果使用BP神经网络合成最终预测结果.通过实际的股票价格对该模型进行验证,结果表明,该组合模型具有较高的预测效果,可以提高股票价格预测的准确率.     

An expert system based on wavelet decomposition and neural network for modeling Chua’s circuit

This paper presents an expert system based on wavelet decomposition and neural network for modeling and simulation of Chua’s circuit which is used for chaos studies. The problems which arise in modeling Chua’s circuit by neural networks are high structural complexity and slow and difficult training. With this proposed method a new solutions is produced to solve these problems. Wavelet decomposition is used for new useful feature extracting from input signal and neural network is used for modeling. Test results of proposed wavelet decomposition and neural network model are compared with test results of neural network model. Desired performance is provided by this new model. Test results showed that the suggested method can be used efficiently for modeling nonlinear dynamical systems.     

A multi-resolution wavelet neural network approach for fouling resistance forecasting of a plate heat exchanger

A new framework regarding wavelet neural network, termed a multi-resolution wavelet neural network (MRWNN), is composed based on the theory of multi-resolution wavelet analysis and orthogonal multi-scale spaces. The hidden layer of the network is divided into two parts, neurons with the Meyer scaling activation function and the Meyer wavelet activation function which is orthogonal to the scaling function. Neurons with the scaling function approximate the contour of the aimed function for its lentitude, and neurons with the wavelet function approximate the details of the aimed function for its sensitive trend. Hidden neurons are mapped to different resolution spaces by redefining the network frame depending on the multi-resolution wavelet analysis theory. By incorporating the Gradient Descent Algorithm, the network can be optimized with less interaction within hidden neurons, and thus, it will acquire a further error convergence state when all the correspondent parameters are adjusted in different resolution spaces. When applied to fouling forecasting of a plate heat exchanger, the MRWNN achieved better performance than other neural networks (NNs) when applied to simulations, proving that the MRWNN is effective in nonlinear function approximations.     

Energy and entropy-based feature extraction for locating fault on transmission lines by using neural network and wavelet packet decomposition

The aim of this paper is to estimate the fault location on transmission lines quickly and accurately. The faulty current and voltage signals obtained from a simulation are decomposed by wavelet packet transform (WPT). The extracted features are applied to artificial neural network (ANN) for estimating fault location. As data sets increase in size, their analysis become more complicated and time consuming. The energy and entropy criterion are applied to wavelet packet coefficients to decrease the size of feature vectors. The test results of ANN demonstrate that the applying of energy criterion to current signals after WPT is a very powerful and reliable method for reducing data sets in size and hence estimating fault locations on transmission lines quickly and accurately.     

A Neural Network Model Based on Logical Operations

On the basis of analyzing some neural network storage capacity problems a network model comprising a new encoding and recalling scheme is presented.By using some logical operations which operate on the binary pattern strings during information processing procedure the model can reach a high storage capacity for a certain size of network framework.     

Temporary short circuit detection in induction motor winding using combination of wavelet transform and neural network

Monitoring system for induction motor is widely developed to detect the incipient fault. Such system is desirable to detect the fault at the running condition to avoid the motor stop running suddenly. In this paper, a new method for detection system is proposed that emphasizes the fault occurrences as temporary short circuit in induction motor winding. The investigation of fault detection is focused on the transient phenomena during starting and ending points of temporary short circuit. The proposed system utilizes the wavelet transform for processing the motor current signal. Energy level of high frequency signal from wavelet transform is used as the input variable of neural network which works as detection system. Three types of neural networks are developed and evaluated including feed forward neural network (FFNN), Elman neural network (ELMNN) and radial basis functions neural network (RBFNN). The results show that ELMNN is the most simply and accurate system that can recognize all of unseen data test. Laboratory based experimental setup is performed to provide real-time measurement data for this research.     

Correlation analysis method based SISO neuro-fuzzy Wiener model

A novel identification algorithm for neuro-fuzzy based single-input-single-output (SISO) Wiener model with colored noises is presented in this paper. The separable signal is adopted to identify the Wiener model, leading to the identification problem of the linear part separated from nonlinear counterpart. Then, the correlation analysis method can be employed for identification of linear part. Moreover, in the presence of random signal, the least square method based parameters estimation algorithm of static nonlinear part are proposed to avoid the impact of colored noise. As a result, proposed method can circumvent the problem of initialization and convergence of the model parameters encountered by the existing iterative algorithms used for identification of Wiener model. Examples are used to verify the effectiveness of the proposed method.     

Control based on double neural networks-PI for parallel mechanism

In this paper, a control system based on double neural networks-PI for 3-RPS parallel mechanism is presented to aim for the nonlinear modeling and controlling. The control system is composed of three linear controllers, one neural network controller (NNC) for compensating the nonlinear modeling and one neural network identification (NNI) for the controlling model. Simulation results have shown that the response time, movement accuracy and resistance to load disturbance of the 3-RPS parallel mechanism system can be improved using the double neural networks-PI.     

Transformation parameter estimation using parallel output based neural network

One of the challenging tasks in image registration is to estimate transformation parameters automatically and efficiently. In this paper, we propose a task decomposition based parallel trained neural network to estimate transformation parameters as well as order of transformations. This parameter estimation problem can be divided into several subproblems like rotation, translation and scaling estimation. Each subproblem or module consists of decomposed input datasets, as well as a part of the output vector. Each module is trained in parallel for some specific and fixed input–output vector pattern. Feature vectors are used as input dataset of the proposed neural network. 2D PCA (two dimensional principal component analysis) feature extraction technique is used to build feature vector. This modular technique requires effectively less computation time in comparison to non-modular network. Moreover, this technique can robustly estimate different transformational parameters. The added advantage of this technique is that it can identify order of the transformation. Experimental results justify the effectiveness of the proposed technique.     

Flood simulation using parallel genetic algorithm integrated wavelet neural networks

The conventional means of flood simulation and prediction using conceptual hydrological model or artificial neural network (ANN) has provided promising results in recent years. However, it is usually difficult to obtain ideal flood reproducing due to the structure of hydrological model. Back propagation (BP) algorithm of ANN may also reach local optimum when training nodal weights. To improve the mapping capability of neural networks, wavelet function was adopted (WANN) to strengthen the non-linear simulation accuracy and generality. In addition, genetic algorithm is integrated with WANN (GAWANN) to avoid reaching local optimum. Meanwhile, Message Passing Interface (MPI) subroutines are introduced for distributed implement considering the time consumption during nodal weights training. The GAWANN was applied in the flood simulation and prediction in arid area. The test results of 4 independent cases were compared to reveal the relations between historical rainfall and runoff under different time lags. The simulation was also carried out with Xinanjiang model to demonstrate the capability of GAWANN. The numerical experiments in this paper indicated that the parallel GAWANN has strong capability of rain-runoff mapping as well as computational efficiency and is suitable for applications of flood simulation in arid areas.     

基于Gamma分布和小波的城域网流量仿真模型

针对网络仿真的需要,在对实际网络流量进行拟合的基础上,设计了基于Gamma分布和小波方法的流量仿真模型。对拟合效果的评估显示,该模型较好地刻画了网络流量的自相似特征,为网络流量仿真提供了一种有效的方法。     

基于神经网络的需求预测模型

零售业的销售过程中积累了大量数据,如何从这些海量数据中提取知识、建立有效的需求预测模型,为零售商提供市场和趋势分析、降低库存成本是零售行业亟待解决的问题。在传统的零售业需求预测模型——Holt-Winter模型中应用神经网络方法,使得需求预测不依赖于数学模型的精度,预测模型中的季节性影响因子等参数能够根据预测误差作相应调整,避免了传统算法中误差的累积,大大提高了预测精度。利用Excel内嵌的VBA实现了该算法,使需求预测能够根据用户需要实现,并提供可视化的结果。     

基于Java的发动机紧致型小波神经网络故障诊断

鉴于发动机是一种复杂的机电液一体化设备,其故障现象和原因之间存在复杂的非线性关联。本文结合小波变换的良好时频域特性和神经网络良好的非线性映射的优势,将MexicanHat小波基作为神经网络的传递函数,组建紧致型小波神经网络,用于发动机的故障诊断;本文以小波神经网络为算法基础,应用具有跨平台、可移植优点的Java语言和SQL Server2005数据库,开发出发动机智能故障诊断软件。