基于模糊调整的径向基函数网络图像恢复算法

为了解决传统图像恢复中存在的建模难的问题,提出了一种基于RBF神经网络的图像恢复算法,该算法利用RBF神经网络的非线性映射能力和适应性,通过记录退化过程的逆过程来恢复图像。首先改进RBF网络中心参数的确定过程,提出基于模糊调整的中心参数学习算法,然后用模糊调整后的网络进行图像恢复。仿真结果表明,改进的RBF网络可对典型退化图像进行令人满意的恢复。     

遗传优化的径向基函数船舶模糊控制器

研究径向基函数模糊神经网络在船舶控制器设计中的应用 ,设计了一个新型的径向基函数模糊神经网络控制器用以适应船舶在时变和不确定环境下的控制性能要求 .控制器设计的主导思想是在传统的径向基函数神经网络中增加一个模糊隐层 ,并采用遗传算法对控制器参数进行优化 .与传统方法相比 ,控制器模糊规则库的设计过程所需的先验知识更少 .最后采用Matlab 6 .1的Simulink工具以船舶运动模型为对象进行了船舶控制的仿真试验 ,结果证明了其有效性     

Optimizing radial basis function neural network based on rough sets and affinity propagation clustering algorithm

A novel method based on rough sets (RS) and the affinity propagation (AP) clustering algorithm is developed to optimize a radial basis function neural network (RBFNN). First, attribute reduction (AR) based on RS theory, as a preprocessor of RBFNN, is presented to eliminate noise and redundant attributes of datasets while determining the number of neurons in the input layer of RBFNN. Second, an AP clustering algorithm is proposed to search for the centers and their widths without a priori knowledge about the number of clusters. These parameters are transferred to the RBF units of RBFNN as the centers and widths of the RBF function. Then the weights connecting the hidden layer and output layer are evaluated and adjusted using the least square method (LSM) according to the output of the RBF units and desired output. Experimental results show that the proposed method has a more powerful generalization capability than conventional methods for an RBFNN.     

基于QPSO算法的RBF神经网络参数优化仿真研究

针对粒子群优化(PSO)算法搜索空间有限,容易陷入局部最优点的缺陷,提出一种以量子粒子群优化(QPSO)算法为基础的RBF神经网络训练算法,将RBF神经网络的参数组成一个多维向量,作为算法中的粒子进行进化,由此在可行解空间范围内搜索最优解。实例仿真表明,该学习算法相比于传统的学习算法计算简单,收敛速度快,并由于其算法模型的自身特性比基于PSO的学习算法具有更好的全局收敛性能。     

基于径向基函数的多帧图像超分辨重建算法

神经网络具有强大的非线性学习能力,基于神经网络的多帧超分辨重建方法获得了初步研究,但这些方法一般只能应用于帧间具有标准位移的控制成像情形,难以推广应用到其他实际情况。为了将神经网络强大的学习能力应用到非控制成像多帧超分辨重建中,以获得更好的超分辨效果,提出了一种利用径向基函数(RBF)神经网络进行解模糊的算法,并将其与多帧非均匀插值结合起来,形成了一种新的两步超分辨算法。仿真实验结果表明,该算法的结构相似度为0.55～0.7。该算法不但扩展了RBF神经网络的应用范围,还获得了更好的超分辨性能。     

基于径向基函数网络的浮游植物活体三维荧光光谱分类

将小波变换与神经网络相结合,对浮游植物活体的三维荧光光谱进行分类．首先利用小波变换对数据进行压缩,然后利用径向基函数（Radial Basis Function,RBF）神经网络对光谱曲线进行逼近,从而进行物种的识别,平均识别率高达95．8％．结果表明,该方法较传统的统计方法更方便、准确率更高．     

基于径向基函数网络的FBG传感光谱的重构

针对光纤Bragg光栅(FBG)解调系统中由于传感器部分脱粘、数据采集的量化误差引起的系统性能的非理想性,FBG反射回的光谱会发生失真。如果直接利用峰值检测法进行波长检测,存在较大的检测误差。于是便提出了用径向基函数网络(RBFN)拟合光纤光栅反射后的失真光谱的方法,此法与峰值检测法、高斯曲线拟合法相比,能够更大程度地减小波长检测的误差,提高解调的精度,并且,能够对多个光栅光谱进行重构。     

基于动态径向基神经网络的人脸识别算法研究

提出了一个通用而且有效的方法来设计RBF神经网络分类器用于人脸识别。为了避免过拟合和减少计算量,用主元分析法和Fisher线性判别技术来降低维数,以提取人脸特征;利用一个混合的学习算法来训练RBF神经网络,使梯度下降法的搜索空间大大减少;采用一种基于训练样本类别信息的新的聚类算法,所有同类的数据可被聚集在一起,尽量减少不同类数据混杂在一起,同时选取结构尽可能紧凑的RBF神经网络分类器。在ORL数据库上进行了仿真,实验结果表明,该算法具有高效性和有效性。     

基于空间分解的参数优化径向基函数近似模型构造方法

为了进一步提高径向基函数(RBF)近似模型的精度,对其近似精度影响因素进行了深入研究.深入分析了计算机舍入误差对RBF近似精度的影响,指出矩阵条件数和形状参数同为影响RBF模型近似精度的两个重要因素.结合灵敏度分析对设计空间进行了分解,改善了矩阵条件数,增加了设计自由度,在传统基于形状参数优化的RBF近似模型的基础上,提出了基于空间分解的参数优化RBF近似模型构造方法.数值实验结果表明,在两个测试算例中,所提方法较传统基于形状参数优化的RBF近似模型构造方法的均方根误差(RMSE)分别减小了51.3%、58.0%,具有更高的近似精度.     

基于增强径向函数神经网络的错误定位方法

结合径向基函数神经网络与正交实验设计理论,提出了一种增强径向基函数神经网络错误定位算法.根据选择的测试用例执行得到源程序的语句覆盖信息和执行结果;通过神经网络计算出每条语句的可疑度值,并通过正交实验设计方法自适应调整神经网络中的参数值;最后按照可疑度值由高到低的顺序逐条检查程序的可疑语句进行错误定位.通过实验对所提出方法与径向基函数神经网络算法以及反向传播神经网络算法进行比较分析,结果表明,基于增强径向基函数神经网络算法具有更精确的错误定位效果和更显著的定位效率.     

Online identification using radial basis function neural network coupled with KPCA

In this paper, we are going to propose an online radial basis function (RBF) neural network algorithm without any preprocessing step. Then a kernel principal component analysis (KPCA) is coupled with the proposed online RBF neural network algorithm. Indeed, the KPCA method is used as a preprocessing step to reduce the feature dimension which fed to the RBF neural network. Reducing memory requirements of the models makes RBF neural network training efficient and fast. These two proposed algorithms are applied, with success, for identification of a mobile robot position. The simulation results present that the used sigmoid function as a kernel, compared to other kernel functions, which gives an excellent model and a minimum mean square error.     

Consideration effect of uncertainty in power system reliability indices using radial basis function network and fuzzy logic theory

Reliability assessment of composite power systems is a critical and important part of power investigations especially in the market-driven environments. Therefore, the reliability indices as criteria for the comparison of the reliability of the power systems should be evaluated precisely and carefully. Because of the nonlinear behavior of the systems as the effect of different parameters like weather conditions, load pattern changes and some others, reliability indices always contain much uncertainty. In this paper a neuro-fuzzy based method is proposed to reduce the degree of the uncertainty in the reliability indices and therefore to evaluate the reliability of the composite power systems precisely. Fuzzy logic theory makes it possible to make use of the human experts knowledge in the reliability evaluations. Also by the use of RBFNN and its powerful characteristic to learn any nonlinear mapping between two states it would be possible to evaluate the reliability indices for every short time interval needed so that reliability evaluation in real time would be achievable and feasible.In this paper the RBFNN is trained by the training patterns that are achieved by the use of fuzzy logic theory, then the results are examined on a standard Reliability Test System (RTS-96).     

基于改进型径向基函数网络的功放非线性建模

针对功率放大器(PA)的非线性建模,提出了改进型径向基函数神经网络(RBFNN)模型。首先,在该模型的输入端加入延迟交叉项和输出反馈项,利用正交最小二乘法提取模型的权值以及隐含层的中心;然后,采用15MHz带宽的宽带码分多址（WCDMA）三载波信号对Doherty功放进行测试,其归一化均方误差(NMSE)可以达到-45dB;最后,通过逆F类功放对模型的普遍适用性进行验证。仿真结果表明,该模型能够更加真实地拟合功率放大器的特性。     

Fuzzy radial basis function network: a parallel design

The fuzzy radial basis function (FRBF) network comprises an integration of the principles of a radial basis function (RBF) network and the fuzzy c-means (FCM) algorithm. A programmable parallel architecture design is proposed for the FRBF, both for FCM clustering at the hidden layer and the weight training at the output layer of the network. The behavior of the system is described in terms of processor utilization. The performance of the parallel design is quantitatively evaluated.     

A variable fidelity information fusion method based on radial basis function

Radial basis function (RBF) model has been widely used in complex engineering design process to replace the computational-intensive simulation models. This paper proposes a variable-fidelity metamodeling (VFM) approach based on RBF, in which different levels fidelity information can be integrated and fully exploited. In the proposed VFM approach, a RBF metamodel is constructed for the low-fidelity (LF) model as a start. Then by taking the constructed LF metamodel as a prior-knowledge and mapping the output space of the LF metamodel to that of the studied high-fidelity (HF) model, a variable fidelity (VF) metamodel is created to approximate the relationships between the design variables and corresponding output responses. A numerical illustrative example is adopted to make a detailed comparison between the VFM approach developed in this research and three existing scaling function based VFM approaches, considering different sample sizes and sample noises. Results illustrate that the proposed VFM approach outperforms the scaling function based VFM approaches both in global and local accuracy. Then the proposed VFM approach is applied to two engineering problems, modeling aerodynamic data for a three-dimensional aircraft and the prediction of weld bead profile in laser welding, to illustrate its ability in support of complex engineering design.     

Control of chaotic dynamical systems using radial basis function network approximators

This paper presents a general control method based on radial basis function networks (RBFNs) for chaotic dynamical systems. For many chaotic systems that can be decomposed into a sum of a linear and a nonlinear part, under some mild conditions the RBFN can be used to well approximate the nonlinear part of the system dynamics. The resulting system is then dominated by the linear part, with some small or weak residual nonlinearities due to the RBFN approximation errors. Thus, a simple linear state-feedback controller can be devised, to drive the system response to a desirable set-point. In addition to some theoretical analysis, computer simulations on two representative continuous-time chaotic systems (the Duffing and the Lorenz systems) are presented to demonstrate the effectiveness of the proposed method.     

基于径向基函数网络的H.264全零块检测算法

针对目前全零块检测算法准确率不高的问题,提出了一种基于径向基函数(RBF)神经网络(NN)的全零块检测算法。通过分析H.264的编码特点,选取了绝对误差和(SAD)、变换绝对差值和(SATD)、编码块类型、率失真优化(RDO)代价、量化系数(QP)、参考块的全零块情况6个特征,考虑了哈达玛变换(HT)中应该使用SATD的情况,采用最小二乘法得到QP与RBF网络宽度参数的关系,根据参考块是否为零,设计了两个分类器来区分全零块与非全零块。在保证图像质量和编码率不变的前提下,平均能提高编码速度50%以上,实验结果表明,利用RBF神经网络很好地提高了全零块检测准确率和编码效率。     

基于径向基函数神经网络的网络流量识别模型

提出了一种基于径向基函数神经网络的网络流量识别方法。根据实际网络中的流量数据,建立了一个基于RBF神经网络的流量识别模型。先介绍了RBF神经网络的结构设计及学习算法,针对RBF神经网络在隐节点过多的情况下算法过于复杂的缺点,采用了优化的算法计算隐含层节点。仿真实验证明,该模型具有较好的准确率、低复杂度、高识别效果和良好的自适应性。     

The improved radial basis function neural network and its application

Compared with other feed-forward neural networks, radial basis function neural networks (RBFNN) have many advantages which make them more suitable for nonlinear system modeling, and they have recently received considerable attention. In this paper, a RBFNN is employed to model strongly nonlinear systems. First, the problems of nonlinear system modeling are analyzed, and then the structure of the RBFNN as well as the training algorithm are improved to solve these problems. Finally, an industrial high-purity distillation column, which is a strongly nonlinear system, is successfully modeled with the improved RBFNN. Owing to the complexities of a nonlinear system, it is necessary to use a real-time model correction method to modify the parameters of the RBFNN model in real time. One efficient method is proposed in this paper. The idea is to employ the Givens transformation to modify the parameters of the RBFNN-based model. This work was presented, in part, at the International Symposium on Artificial Life and Robotics, Oita, Japan, February 18–20, 1996