一种基于改进神经网络的系统辨识方法

该文在分析神经网络辨识技术特点及现状的基础上,将BP神经网络结构和遗传算法相结合,设计了一种适用于非线性系统的辨识器模型。该辨识器模型首先建立初始的BP神经网络结构,再利用遗传算法对BP神经网络的权值和阈值进行优化,从而优化BP神经网络,通过迭代最终建立辨识器模型。最后,通过一个三阶非线性多输入单输出系统的仿真实验证明了所设计的辨识器具有辨识时间短、辨识精度高的特点,为神经网络辨识技术的研究提供了新的思路和方法。     

基于BP神经网络辨识模型的PEMFC系统建模

为分析燃料电池系统特性,采用BP神经网络结构辨识质子交换膜燃料电池系统模型,模型输入为系统实际输入,模型输出为电堆输出电压和电堆工作温度。由于PEMFC系统是一个时变非线性系统,采用一种串-并联前向神经网络辨识结构模型,将模型前几个时刻输出作为模型输入,使得静态网络结构具有动态特性。BP网络模型通过PEMFC系统所得到的实验数据进辨识。训练完成后BP网络模型输出与实际系统输出基本一致,结果表明BP网络模型能够有效反映质子交换膜燃料电池系统输出电压和电堆温度特性。     

一种新的模糊聚类神经网络及其在语音信号系统辨识中的应用

主要解决语音信号模型的系统辨识问题.针对过去的模糊聚类算法进行系统辨识时逼近性能不理想的问题,提出了一种新的模糊聚类神经网络(FCNN).该方法以模糊系统模型为基础,将每个状态看作一个模糊系统,用连续的若干序列作为系统的输入,利用改进的模糊聚类辨识算法构成一种新型的模糊聚类神经网络,对系统的输出进行预测.通过语音信号系统辨识的实验,验证了本网络的有效性.     

带动量批处理梯度下降法对模型的动态辨识

针对生产过程中参数容易受外界影响而改变，传统的系统辨识方法难以得到精确的数学模型的实际情况，介绍一种用改进的BP神经网络辨识对象模型的方法。采用串一并联型的辨识结构；针对BP算法收敛速度慢，容易陷入局部最小的缺点，提出了带动量的批处理梯度下降的方法；为了更有效地辨识对象的动态过程模型，其输入／输出加上按拍延迟线。用MATLAB对该改进的BP神经网络辨识方法进行了设计、仿真和性能分析，结果表明：该方法具有良好的辨识能力。     

基于神经网络的机械传动方案决策系统的建模与仿真

针对机械传动方案决策系统难以建立准确数学模型的特点,提出了利用BP及BP改进神经网络的非线性映射能力,进行机械传动方案决策系统模型辨识建模的新方法.介于网络的输入输出数据难以精确描述的特点,运用模糊理论对其进行了处理.通过仿真实验验证了利用BP及BP改进网络建立其决策模型的可行性,表明了BP改进算法收敛速度更快,模型更准确,从而为机械系统传动方案的决策提供了一种更为有效的推理模型.     

盾构机密封舱压力控制模型参数辨识

为了解决盾构机密封舱压力控制模型参数辨识问题,提出了基于神经网络的盾构机密封舱压力控制模型参数辨识方法。根据系统输出（密封舱压力）与系统输入（螺旋输送机转速）之间的关系,建立了串并联神经网络辨识器。数值仿真结果表明,该辨识方法对于观测噪声具有良好的鲁棒性。实验台实验结果验证了所提出的模型参数辨识方法的有效性,模型预测的密封舱压力能够较精确拟合密封舱压力观测值。     

改进神经网络模型在光伏发电预测中的应用

针对传统的BP神经网络模型在光伏发电功率预测中,存在着的预测精度不高、收敛速度较慢的弊端,本文提出一种改进型的BP神经网络模型,运用增加动量项以及自适应选取最佳隐含层的方法来改进传统BP模型的缺陷.文中首先进行了各类气象因素对于光伏发电功率输出影响的相关性分析,提取出最能影响光伏发电功率的6个气象因素,作为网络模型的输入,然后建立了改进型的BP网络模型,结合光伏功率输出的历史数据,来进行发电数据的直接预测;最后根据不同气候类型下的预测结果,分析并验证了采用该模型进行功率预测的可行性和有效性.     

一种多输入单输出Hammerstein系统的集成辨识方法

针对多输入单输出(MISO)Hammerstein系统提出了一种稳态与动态辨识相结合的集成辨识方法.该方法利用稳态信息获取稳态模型的强一致性估计,并通过稳态模型以神经网络获得其非线性逼近函数,再利用动态信息辨识获取多输入单输出(MISO)Hammerstein系统的线性子系统未知参数的一致性估计.仿真结果表明了该方法的有效性和实用性.     

基于神经网络的PID自整定控制系统

文章介绍了一种应用神经网络技术建立的PID自整定控制系统,给出了系统结构,详细分析了BP神经网络和RBF神经网络的结构和学习算法。该系统采用3层BP神经网络,其输出为PID控制器的参数;通过变结构的RBF神经网络辨识控制对象,将得到的输出对输入的梯度信息提供给BP神经网络,BP神经网络根据该信息优化PID控制器参数。仿真结果表明,该系统对于参数扰动较大的非线性系统,其收敛速度快、动态响应能力强、稳定性好,且具有较强的鲁棒性和适应性。     

BP神经网络合理隐结点数确定的改进方法

合理选择隐含层结点个数是BP神经网络构造中的关键问题,对网络的适应能力、学习速率都有重要的影响.在此提出一种确定隐结点个数的改进方法.该方法基于隐含层神经元输出之间的线性相关关系与线性无关关系,对神经网络隐结点个数进行削减,缩减网络规模.以零件工艺过程中的加工参数作为BP神经网络的输入,加工完成的零件尺寸作为BP神经网络的输出建立模型,把该方法应用于此神经网络模型中,其训练结果证明了该方法的有效性.     

"弹性"BP神经网络在识别带有噪声字母中的应用

字符识别是模式识别中的一个典型应用,通过训练网络可以教会计算机如何识别字符,这在票据处理方面可以大大地提高效率.该文中所建立的神经网络具有Sigmoid型可微函数的三层BP神经网络,它可以以任意精度逼近任何连续函数,实现输入和输出之间的任意非线性映射.文中分析了BP神经网络的“弹性”学习算法,利用五位二进制数来识别的输出26种状态.建立的一个三层的BP神经网络能对带有噪声的26个英文大写字母进行识别.利用MATLAB编写仿真程序对BP神经网络进行训练,仿真结果表明训练的BP神经网络可以对给定的带有噪声的字母正确地识别.     

Correlation analysis based MIMO neuro-fuzzy Hammerstein model with noises

A novel identification algorithm for neuro-fuzzy based MIMO Hammerstein system with noises by using the correlation analysis method is presented in this paper. A special test signal that contains independent separable signals and uniformly random multi-step signal is adopted to identify the MIMO Hammerstein system, resulting in the identification problem of the linear model separated from that of nonlinear part. As a result, it can circumvent the problem of initialization and convergence of the model parameters encountered by the existing iterative algorithms used for identification of MIMO Hammerstein model. Moreover, least square method based parameter identification algorithms of dynamic linear part and static nonlinear part are proposed to avoid the influence of noise. Examples are used to illustrate the effectiveness of the proposed method.     

Fast and improved backpropagation learning of multi-layer artificial neural network using adaptive activation function

In the conventional backpropagation (BP) learning algorithm used for the training of the connecting weights of the artificial neural network (ANN), a fixed slope−based sigmoidal activation function is used. This limitation leads to slower training of the network because only the weights of different layers are adjusted using the conventional BP algorithm. To accelerate the rate of convergence during the training phase of the ANN, in addition to updates of weights, the slope of the sigmoid function associated with artificial neuron can also be adjusted by using a newly developed learning rule. To achieve this objective, in this paper, new BP learning rules for slope adjustment of the activation function associated with the neurons have been derived. The combined rules both for connecting weights and slopes of sigmoid functions are then applied to the ANN structure to achieve faster training. In addition, two benchmark problems: classification and nonlinear system identification are solved using the trained ANN. The results of simulation-based experiments demonstrate that, in general, the proposed new BP learning rules for slope and weight adjustments of ANN provide superior convergence performance during the training phase as well as improved performance in terms of root mean square error and mean absolute deviation for classification and nonlinear system identification problems.     

On-line system identification of complex systems using Chebyshev neural networks

This paper proposes a computationally efficient artificial neural network (ANN) model for system identification of unknown dynamic nonlinear discrete time systems. A single layer functional link ANN is used for the model where the need of hidden layer is eliminated by expanding the input pattern by Chebyshev polynomials. Thus, creation of nonlinear decision boundaries in the multidimensional input space and approximation of complex nonlinear systems becomes easier. These models are linear in their parameters and nonlinear in the inputs. The recursive least squares method with forgetting factor is used as on-line learning algorithm for parameter updation. The good behaviour of the identification method is tested on Box and Jenkins Gas furnace benchmark identification problem, single input single output (SISO) and multi input multi output (MIMO) discrete time plants. Stability of the identification scheme is also addressed.     

A RECURSIVE 4SID FROM THE INPUT-OUTPUT POINT OF VIEW

In this paper, recursive algorithms of subspace state-space system identification (4SID) for multiple-input, multiple-output (MIMO), finite dimensional, linear time-invariant (FDLTI) systems are proposed. These algorithms are derived based on the Matrix Inversion Lemma. The investigation of our algorithms clarifies that a series of 4SID is the extension of the classical least square method to identification for multivariable systems, and also that our algorithms are the direct extension of the recursive least square algorithm to such ones. For PO-MOESP (the ordinary MOESP scheme with instrumental variables constructed from Past input and Output measurements), we show the mechanism of how the effect of the process and measurement noises is eliminated asymptotically by a projection related to the input and regressor matrices. “MOESP” is an abbreviation for “the MIMO output-error state-space model identification.”     

基于遗传算法优化的BP网络速度观测器的研究

根据三相异步电机的数学模型,提出了一种基于智能算法优化的速度观测器,以实现无速度传感器在直接转矩控制系统中的速度闭环控制。在通过BP神经网络训练得到的DTC系统的速度观测器的基础上,针对BP神经网络寻优参数多、易陷于局部极值以及初始设置对训练结果影响大等的不足,采用遗传算法对其进行优化设计。由仿真结果可知,用遗传算法优化后的BP神经网络较单纯的BP神经网络速度观测器具有更高的精度。     

Remote sensing of algal blooms by aircraft and satellite in Lake Erie and Utah Lake

During late summer, when the surface waters of Lake Erie reach their maximum temperature, an algal bloom is likely to develop. Such phenomena, which characterize eutrophic conditions, have been noticed on other shallow lakes using the Earth Resources Technology Satellite (ERTS-1). The concentration of the algae into long streamers provides additional information on surface circulations. To augment the ERTS Multispectral Scanner Subsystem (MSS) data of Lake Erie an aircraft was used to obtain correlative thermal-IR and additional multiband photographs. The algal bloom is highly absorptive in the visible wavelengths but reverses contrast with the surrounding water in the near-IR bands. The absorption of shortwave energy heats the dark brown algal mass, providing a hot surface target for the thermal-IR scanner. A large bloom of Aphanizomenon flos-aquae observed in Utah Lake together with recent bloom history in Lake Erie is used to verify the Great Lakes bloom.     

优化的线谱特征在水下目标识别中的应用

水下目标舰船辐射噪声中有很强的线谱成份，这些成份对分类识别是非常有价值的，因此利用线谱特征射舰船目标进行识别是水声领域研究的重要内容。采用最小二乘曲线拟合方法提取了舰船辐射噪声的连续谱。在此基础上主要研究了线谱特征的提取方法，获得了线谱的4维特征，对所提取的4维线谱特征进行了优化处理。基于这些优化的线谱特征采用BP神经网络对海上实录的三类水下目标辐射噪声进行了分类识别，实验结果证明了方法的可行性和有效性，识别率达到90％以上，分类效果很好。     

基于BP人工神经网络的信息过滤技术研究

首先介绍了信息过滤中向量空间模型的基本工作原理。针对目前信息过滤技术过滤精度低、速度慢的缺点,利用BP人工神经网络网络稳定、学习能力强的特点,重点研究了BP人工神经网络特征输入数据的取得、网络拓扑结构的选择、BP人工神经网络的训练过程、神经网络产生结果数据的处理方式,在此基础上提出了一种基于BP人工神经网络的信息过滤模型,实验证明,此模型过滤效率有了显著提高。     

AGCL,a FORTRAN IV program for agglomerative,nonhierarchical Q-mode classification of large data sets

This paper describes a FORTRAN IV computer program for an agglomerative, nonhierarchical classification scheme for large data sets. The program will accomodate 1000 items and 10 components. An example of geochemical data from Lake Erie is given.