RBF神经网络在传感器故障诊断中的应用研究

提出一种基于RBF神经网络的传感器在线故障诊断方法,用某气囊隔振系统隔振装置中大量的传感器数据进行了仿真实验,验证了方法的可行性.通过对每个传感器建立单独的神经网络预测模型,实现多传感器的故障诊断.     

径向基函数神经网络在集装箱吞吐量预测中的应用

针对一类具有时间序列特性的数据,构造一种基于径向基函数神经网络的预测模型,并将该模型应用于上海港集装箱吴吐量的预测。     

基于RBF神经网络的雷达网抗干扰效果评估

通过分析雷达网抗干扰效果的评估指标集,建立了一种基于RBF神经网络的雷达网抗干扰效果评估模型,并通过试验数据样本进行学习训练RBF神经网络模型。最后,利用训练好的RBF神经网络模型对雷达网的抗干扰效果进行评估,结果表明基于RBF神经网络的雷达网抗干扰效果模型具有一定实用性和可行性。     

基于RBF神经网络的舰船维修费预测

鉴于单个预测模型在舰船维修费预测中存在信息源不够广泛的不足,提出用组合预测法进行维修费预测。构建了基于径向基函数(RBF-Radial Basis Function)神经网络的舰船维修费组合预测模型,并用实例进行了验证。结果证明了此方法的有效性。     

RBF神经网络模型在边坡监测中的应用

在matlab环境下建立了边坡位移预测的RBF神经网络模型.利用已有的监测数据训练神经网络并进行测试,将训练好的网络模型用于预测边坡位移的变化值.最后将该预测方法用于斯里兰卡某边坡监测工程,预测结果与实测的监测数据相比误差较小,从而合理安排监测频次,提高了监测效率.     

基于高维优化的RBF神经网络螺旋桨性能预测

螺旋桨式水下推进器推力的大小决定着水下机器人的水动性能.为快速准确预测螺旋桨的敞水性能,建立一种RBF神经网络螺旋桨敞水性能估计器模型.利用几种型号螺旋桨敞水仿真值作为训练样本,对网络规模进行调整.在此基础上,广义逆动态地调整各网络间的连接权重,进行网络参数的优化.通过不断地迭代优化达到学习精度要求,最终得到一种高维优...     

Nuclear power plant fault diagnosis based on genetic-RBF neural network

1 Introduction1 Nuclear power plant is a kind of complicated large-scale system, and the fault mechanism and reason are very complex either. Generally, the fault has no obvious relation with omen. Besides, there are also large quantities of uncertainties.…     

基于RBF的混沌序列产生的新方法

介绍了利用RBF（Radial Basis Function）网络产生混沌序列的方法，即利用RBF网络跟踪混沌序列的模型，分析RBF神经网络对混沌序列的学习能力。最后利用最大Lyapunov指数判别法检验其序列，结果表明其序列具有较好的保密性能。     

基于RBF神经网络的港口货物吞吐量预测

将人工神经网络理论引入港口货物吞吐量远景观测，并依据，1978－1998年北方某大港主要要经济腹地的相关经济变量的历史数据建立RBF神经网络预测模型，最后利用模型对未来15该港货物吞吐量进行了预测。     

基于小波分析与径向基函数神经网络的气缸压力重构

介绍了RBF神经网络的原理及其此法重构气缸压力的实现步骤,并根据柴油机缸盖振动的激励机理,对径向基函数神经网络的参数进行了有效的设置。通过用不同工况下的样本进行训练,建立了柴油机缸盖振动信号与气缸压力之间的非线性映射关系。为了提高重构精度,更为准确地提取压力在缸盖处的响应,提出了利用小波分析对振动信号进行降噪处理。试验表明应用RBF网络法重构气缸压力简单有效,重构压力曲线与实测曲线拟合度良好。     

基于VMD和RBF的舰船管路泄漏识别和定位

针对工作环境恶劣、维护保养不便的舰船管路难以迅速定位泄漏点并对其进行损害管制,提出了一种基于VMD和RBF的舰船管路泄漏识别和定位方法。首先,对管路泄漏产生的空化现象、湍流和流体与管路的摩擦进行分析,研究影响泄漏产生激励的因素；然后,提出一种基于变分模态分解(VMD)与径向基函数(RBF)神经网络的管路泄漏识别和定位方法,通过VMD得到有效分量的中心频率和能量值分别构造特征向量,输入RBF神经网络以达到泄漏识别和定位目的；最后,模拟舰船环境,搭建泄漏管路试验平台,分析泄漏管路不同工况下的振动信号,并对RBF神经网络的诊断准确率进行验证。实测舰船管路故障信号分析表明,泄漏识别的准确率为90%,泄漏定位的准确率为87.5%。     

A method of multi-channel reference signals acquiring in broadband ANC

In a flank array on an unmanned underwater vehicle （UUV）, self-generated noise which has broadband and colored spectrum property in frequency and spatial domain is the main factor affecting the performance of weak signal detection, so the technique of adaptive noise cancellation （ANC） as well as physical denoising and active noise cancellation are often used in practice. Because ANC is based on correlations, improvements in performance come from better correlation between reference signals and primary signals. Taking full advantage of the characteristics of flank arrays and the characteristics of information obtained from hydrophones, a new method for reference signal acquisition for adaptive noise cancellation is proposed, in which the multi-channel reference signals are obtained by accurate delaying for a given direction of arrival （DOA） and differencing between adjacent outputs of array elements. The validity of the proposed method was verified through system modeling simulations and lake experiments which showed good performance with little additional computational burden.     

基于径向基函数网络的水声信号预测模型研究

随着舰船隐身降噪技术的不断发展,舰船辐射噪声信号检测变得愈加困难.针对这种情况,将RBF神经网络引入到水声信号的预测中,建立水声信号的全局预测模型.通过对Henon和Lorenz映射仿真,验证了RBF预测模型的有效性.运用全局预测模型对两种实际水声信号进行预测.实验证明,该预测模型学习速度快,所需样本点少,效果较准确,在水声信号检测中具有良好的发展前景.     

一种船用燃气轮机强跟踪扩展卡尔曼滤波器健康参数估计方法（英文）

Monitoring and evaluating the health parameters of marine gas turbine engine help in developing predictive control techniques and maintenance schedules. Because the health parameters are unmeasurable, researchers estimate them only based on the available measurement parameters. Kalman filter-based approaches are the most commonly used estimation approaches; however, the conventional Kalman filter-based approaches have a poor robustness to the model uncertainty, and their ability to track the mutation condition is influenced by historical data. Therefore, in this paper, an improved Kalman filter-based algorithm called the strong tracking extended Kalman filter(STEKF) approach is proposed to estimate the gas turbine health parameters. The analytical expressions of Jacobian matrixes are deduced by non-equilibrium point analytical linearization to address the problem of the conventional approaches. The proposed approach was used to estimate the health parameters of a two-shaft marine gas turbine engine in the simulation environment and was compared with the extended Kalman filter(EKF) and the unscented Kalman filter(UKF). The results show that the STEKF approach not only has a computation cost similar to that of the EKF approach but also outperforms the EKF approach when the health parameters change abruptly and the noise mean value is not zero.     

基于模糊RBF神经网络的辐射源识别

针对模糊识别系统的不足,为了提高辐射源识别系统的识别正确率,构建了基于模糊RBF神经网络的辐射源识别系统,提出了一种等价型模糊RBF神经网络的结构和学习算法,采用五层神经网络结构来实现模糊系统的模糊化和规则推理,神经网络的所有节点和参数对应了模糊系统的隶属函数和推理过程.在仿真实验中,分别采用模糊识别系统、并联型模糊RBF神经网络、结构等价型模糊RBF神经网络进行辐射源识别,给出了三种算法在相同噪声环境下的仿真结果,表明等价型模糊RBF效神经网络有较高的正确识别率,具有更强的抗干扰能力,但运算量相对较大.     

对称Alpha稳定分布噪声下的Duffing振子检测方法

Duffing振子是利用系统对与策动力同频的小信号敏感而对噪声免疫实现微弱信号检测,特定分布下的噪声激励Duffing振子系统不会发生相变是应用该方法的前提条件。文中主要研究了服从Alpha稳定分布的噪声激励Duffing振子产生相变的鲁棒性问题,研究结果表明Duffing振子相变在Alpha稳定分布源的激励下为小概率事件。为消除小概率相变的影响,利用多支路并行检测及多数判决准则对常规的Duffing振子检测方法进行改进,即将待测信号分段截短周期延拓后送入多个并行Duffing振子检测单元,若检测单元多数发生相变,必然是由于弱目标信号而非噪声激励所致,即可判定检测信号中包含目标小信号。将该方法应用于水下目标回波信号的检测中,实测数据处理结果验证了该方法是有效的。     

基于高阶谱的舰船噪声特征提取与实验

利用高阶谱对舰船辐射噪声进行了分析,着重研究了采用双谱分析的方法提取舰船噪声特征,给出了维谱特征提取的算法,提取了9维的特征向量并利用神经网络进行了训练与识别.实验表明,该方法对舰船辐射噪声信号具有很好的分类效果.     

基于卡尔曼滤波的水声信号线谱增强（英文）

To detect weak underwater acoustic signals radiated by submarines and other underwater equipment, an effective line spectrum enhancement algorithm based on Kalman filter and FFT processing is proposed. The proposed algorithm first determines the frequency components of the weak underwater signal and then filters the signal to enhance the line spectrum, thereby improving the signal-to-noise ratio(SNR). This paper discussed two cases: one is a simulated signal consisting of a dual-frequency sinusoidal periodic signal and Gaussian white noise, and the signal is received after passing through a Rayleigh fading channel;the other is a ship signal recorded from the South China Sea. The results show that the line spectrum of the underwater acoustic signal could be effectively enhanced in both cases, and the filtered waveform is smoother. The analysis of simulated signals and ship signal reflects the effectiveness of the proposed algorithm.