递归神经网络的模糊随机学习算法

提出了一种新的基于模糊逻辑的Alopex学习算法（FLA）。FLA算法利用模糊逻辑推理实时获得适应于学习过程的适当的算法修正值，克服了Alopex算法中修正值固定不变的缺点，使得随机学习过程在速度、精度和稳定性之间获得平衡。将该算法应用于神经网络的训练，可以无需神经网络的梯度信息和结构信息，因此可以用于具有各种结构特性的递归神经网络对动态系统的学习过程。实验结果表明了FLA算法的有效性。     

递阶神经网络建模环污水处理过程的应用

从污水处理过程的多非线性和多变量子系统的串级结构特点出发,在活性污泥过程的基础上提出了递阶神经网络建模方法。此法采用串级的方式把过程机理模型和神经网络连接起来,用神经网络对活性污泥过程中的非线性组分反应速率进行辨识,对各子过程建模误差的关系进行分析,提出了稳定性理论分析和稳定学习算法。     

基于在线神经网络算法的混合试验方法

混合试验是一种将数值模拟与物理试验相结合的新兴结构抗震试验方法,得到了相关研究者们的广泛关注。如何模拟具有强非线性的数值子结构仍是混合试验亟待解决的问题。在传统的离线神经网络基础上提出一种在线学习的神经网络算法,并应用于混合试验中来在线预测数值子结构恢复力。在线学习算法仅利用当前步的系统输入和观测样本,采用递推形式更新每一步的权值和阈值。针对两个自由度非线性结构,分别进行了基于在线学习和离线学习神经网络的混合试验数值仿真。研究表明:与离线学习神经网络算法相比,在线学习神经网络算法具有更好的自适应性,能够有效提高恢复力预测精度和计算效率;基于在线学习神经网络算法的结构混合试验方法可以提高混合试验结果精度。     

神经网络的进化学习方法研究

提出了两种用于前向神经网络的进化学习算法,实验证明它们能有效地网络权值空间中寻找全局最优解。在比较实验的基础上,得出了在神经网络的进化学习过程中变异是起主导作用的遗传算子的结论,并以此为指导配置算法的各个关键参数。通过对ＸＯＲ问题和ＩＲＩＳ模式分类问题的学习证明,这两种算法均能获得远高于传统的ＢＰ算法的性能。     

皮带秤动态称重用神经网络算法设计

针对皮带秤在使用中难以保持标称计量精度的缺点,提出将过程神经网络引入皮带秤动态称重误差的补偿中。将动态称量过程中皮带秤单位长度上的重量、皮带速度、皮带垂度变化作为模型输入,设计了应用于皮带秤动态称重误差研究的单隐层过程神经网络误差反传播学习算法,利用Matlab软件对算法模型进行训练和测试,模型经过149次学习优化达到网络精度要求,测试组误差为1%,较使用网络前的原误差明显降低,验证了算法的可行性和有效性。     

BP算法的改进与性能分析

在人工神经网络中,BP神经网络通常是指基于误差反向传播算法（BP算法）的多层前向神经网络。BP算法已成为目前应用最为广泛的神经网络学习算法,然而BP算法具有收敛速度慢、易陷入局部极小点等缺陷。为了克服常规BP算法的缺陷,MATLAB神经网络工具箱对常规BP算法进行了改进,并提供了一系列快速学习算法。实验结果表明,这些方法有效地提高了BP算法的收敛性,避免陷入局部最小点。     

动态定量称量包装系统BP神经网络PID控制算法

目的针对动态定量称量包装控制系统具有大惯性、滞后、非线性且无法建立精确数学模型等缺点,研究提高动态定量称量包装系统控制精度的方法。方法提出了一种改进型BP神经网络PID的定量称量包装控制系统,将BP神经网络与PID控制方法相结合,通过神经网络的自学习、加权系数的调整,优化PID控制器参数K_i,K_p,K_d,并将粒子群算法引入到神经网络中作为其学习算法,以有效提高BP神经网络算法的收敛速度。结果仿真和实验结果表明,改进型BP神经网络PID控制响应速度快、超调量较小,系统称量误差得到大幅度减小。结论所述控制方法可以明显提高定量称量控制过程的稳定性、精确性以及鲁棒性。     

结构损伤的遗传神经网络检测方法

将遗传算法与神经网络相结合,用遗传算法完成神经网络的学习过程,建立了结构损伤的遗传神经网络检测方法,并对遗传算法进行了改进.研究结果表明,用改进的遗传算法进化神经网络可以有效地避免BP算法有可能陷入局部极小的缺点,而且运算速度大大加快,精度提高.     

基于神经网络模型的动载荷识别

依据结构动力学理论推导了在时域中用于神经网络算法的自回归函数，相应建立了具有时延反馈的神经网络动载荷识别模型。阐明了这种网络的基本学习算法和回忆算法。数值仿真和试验件的验证试验表明该神经网络模型用于动载荷识别时具有精度高、无累积误差、抗干扰能力强等优点，并且适用于各种类型的动载荷，尤其对冲击载荷的识别更具有独特的优势。该模型在动标学习过程中要求信息量小，试验成本低，是一种非常值得在工程中推广应用的新型动载荷识别方法。     

基于区间值描述的自适应直觉模糊神经网络

针对模糊集合在语义描述上存在的不足,为提高自适应模糊神经网络的紧凑性,提出了自适应直觉模糊神经网络。首先,推导了适合神经网络计算的直觉模糊规则。接着,给出了自适应直觉模糊神经网络的结构和各层的含义,并给出了网络学习算法和收敛性分析。最后,通过典型实例仿真试验,表明提出的自适应模糊神经网络结构更为紧凑,学习算法的泛化性能更佳。     

Improved probabilistic neural network algorithm for chemical sensor array pattern recognition

An improved probabilistic neural network (IPNN) algorithm for use in chemical sensor array pattern recognition applications is described. The IPNN is based on a modified probabilistic neural network (PNN) with three innovations designed to reduce the computational and memory requirements, to speed training, and to decrease the false alarm rate. The utility of this new approach is illustrated with the use of four data sets extracted from simulated and laboratory-collected surface acoustic wave sensor array data. A competitive learning strategy, based on a learning vector quantization neural network, is shown to reduce the storage and computation requirements. The IPNN hidden layer requires only a fraction of the storage space of a conventional PNN. A simple distance-based calculation is reported to approximate the optimal kernel width of a PNN. This calculation is found to decrease the training time and requires no user input. A general procedure for selecting the optimal rejection threshold for a PNN-based algorithm using Monte Carlo simulations is also demonstrated. This outlier rejection strategy is implemented for an IPNN classifier and found to reject ambiguous patterns, thereby decreasing the potential for false alarms.     

应用概率神经网络诊断自行火炮发动机的故障

目的　研究概率神经网络模型 ,并应用于故障诊断 .方法　对基于概率统计思想和 Bayes分类规则的概率神经网络模型、网络结构、算法及其特点进行分析 ,利用其进行故障诊断 ,并提出一种优化估计平滑因子的方法 .结果　概率神经网络可很好地诊断自行火炮发动机进行中油路和气路的故障 .结论　概率神经网络在模式识别和故障诊断领域中可取得良好地应用效果     

Active control strategy of structures based on lattice type probabilistic neural network

A new neuro-control scheme for active control of structures having a basic structure similar to the Probabilistic Neural Network (PNN) is proposed. It utilizes the lattice pattern of state vector as the training data of PNN, and thus it is called the Lattice Probabilistic Neural Network (LPNN). Comparing the two schemes, PNN takes much time to obtain a control force in the application because it uses all the training patterns. This may delay the control action inevitably. However, in LPNN, the control force is calculated by using only the adjacent information of LPNN input, making the response of LPNN greatly faster than that of PNN. To investigate the general control capability of the proposed algorithm, one-story and three-story buildings under California, El Centro, and Northridge earthquakes are used as test models. Control results of the LPNN are compared with those of the conventional PNN, and these show that the structural responses have been suppressed effectively by the proposed algorithm.     

滚动轴承的MSE和PNN故障诊断方法

针对滚动轴承不同运行状态振动信号具有不同复杂性的特点,提出一种新的基于多尺度熵(multiscale entropy, MSE)和概率神经网络(probabilistic neural networks, PNN)的滚动轴承故障诊断方法。该方法首先利用MSE方法对滚动轴承振动信号进行特征提取,并将其作为PNN神经网络的输入,再利用PNN自动识别轴承故障类型及故障程度。实验数据包括不同故障类型和不同故障程度样本,结果表明,相比于小波包分解和PNN结合的诊断方法,提出的方法具有更高的诊断精度,能有效实现滚动轴承故障类型及程度的诊断。     

基于过程神经网络算法的航天器非平稳随机振动时频分析

采用时变参数模型对航天器某时段非平稳随机振动信号(NSRVS)进行建模分析,利用过程神经元网络(PNN)求解模型的时变参数并以此确定信号的时变自功率谱密度.计算结果表明:由PNN估计的NSRVS时变参数与自相关Levinson法估计的该参数基本一致,但前者建模物理意义明确,和传统的方法相比避免了计算信号的自相关矩阵,减少了存储空间,提高了频率分辨率和计算速度.     

Probabilistic neural-network-based protection of power transformer

An optimal probabilistic neural network (PNN) as a core classifier for fault detection and status indication of a power transformer has been presented. In this scheme, various operating conditions of a transformer are distinguished using signatures of the differential currents. The proposed differential protection scheme is implemented through two different structures of PNN, that is, one having one output and the other having five outputs. The developed algorithm is found to be stable against external fault, magnetising inrush, sympathetic inrush and over-excitation conditions for which relay operation is not required. For the test data of fault, it is found to operate successfully. The performance of proposed PNN and classical artificial neural network (ANN) has been compared. For evaluation of the developed algorithm, relaying signals for various operating conditions of a transformer are obtained by modelling the transformer in PSCAD/EMTDC. The algorithms are implemented using MATLAB. The results show the capability of PNN in terms of classification accuracy and speed in comparison to classical ANNs.     

Effect of calcination conditions on phase formation and particle size of lead nickel niobate powders synthesized by using Ni4Nb2O9 precursor

An approach to synthesize lead nickel niobate, Pb(Ni_1/3Nb_2/3)O₃ or PNN, powders with a modified two-stage mixed oxide synthetic route has been developed. Novel intermediate phase of nickel diniobate (Ni₄Nb₂O₉) was employed as a B-site precursor, with the formation of the PNN phase investigated as a function of calcination conditions by TG-DTA and XRD techniques. Morphology, particle size and chemical composition have been determined via a combination of SEM and EDX techniques. It has been found that the unreacted PbO and Pb_1.45Nb₂O_6.26 phases tend to form together with PNN, depending on calcination conditions. It is seen that optimization of calcination conditions can lead to a 100% yield of PNN in a cubic phase.     

用概率社会网络进行结构损伤位置识别

在不计测量误差情况下,神经网络能够成功地识别损伤位置及其程度,但在测量噪声影响下,神经网络的损伤识别效果则比较差,考虑到基于多变量模式分类的概率神经网络具有处理受噪声污染的测试数据的能力,本文将可能的损伤位置作为模式类,利用概率神经网络的分类能力来识别结构的损,地对两个算例,一个六层框架和一个两层框架进行数值模拟分析,并将概率神经网络与BP网络进行了比较,结果表明,概率神经网络具有更好的识别效果,是一种很有潜力的结构损伤位置识别方法。     

基于多源数据融合的掘进机截割岩壁硬度识别方法

针对煤矿井下掘进机截割岩壁硬度识别难度大的问题,利用其悬臂振动信号、升降油缸和回转油缸压力信号、截割电机电流信号,提出了一种基于多源数据融合的截割岩壁硬度识别方法。该方法首先对各类信号进行小波包分解,单支重构各频带信号并组建时频矩阵,通过奇异值分解得到包含时频信息的若干特征奇异值,以构造特征向量;再利用LDA算法实现数据特征级融合,得到类可分性更好的低维特征。为解决概率神经网络(PNN)平滑参数无法确定和网络结构复杂的问题,提出了基于差分进化算法(DE)和QR分解的PNN优化方法,并通过优化PNN对低维特征进行硬度识别。实验结果表明:所提出的特征量提取和模式识别方法是有效的,与目前常用的其它模式识别算法相比,优化PNN在掘进机三种工况下均有更高的硬度识别准确率。     

Effects of PNN/PZT ratios on phase structure,electric properties and relaxation behavior of PZN–PNN–PZT ceramics

Pb(Zn_1/3Ni_2/3)_c(Ni_1/3Nb_2/3)_a(Zr_xTi_y)_bO₃ (PZN–PNN–PZT, the ratios of PNN/PZT a/b were 0.88, 1 and 1.136) piezoelectric ceramics were prepared by a traditional solid-state reaction method. The effects of PNN/PZT ratio on phase structure, microstructure and electric properties as well as the relaxation behaviors of PZN–PNN–PZT ceramics were investigated. The XRD patterns showed that all ceramics samples had a pure perovskite phase structure. Meanwhile, it was found that the phase structure undergoes a tetragonal, tetragonal-rhombohedral to rhombohedral transition as ratios of PNN/PZT increased. With the increasing of a/b from 0.88 to 1.136, the dielectric constant and diffusive phase coefficient decreases, it was indicated that relaxation behaviors also decreased. When ceramics with a/b was 1.136, the dielectric relaxation γ reached the minimum and electrical properties were poor. The electric properties of ceramics with a/b was 1.00 have an excellent properties, it was indicated that ceramics reached an optimization at the MPB structure.