人工神经网络在电积锌工艺中的应用

本文提出了一种基于人工神经网络分析电积锌工艺的方法 ,通过仿真计算 ,表明这种方法切实可行     

人工神经网络在材料设计中的应用

在实验数据的基础上,利用人工神经网络建立高Ｃｏ－ Ｎｉ 二次硬化钢的力学性能与合金成分及热处理温度对应关系的模型． 首次提出将五个材料力学性能指标及部分合金成分作为网络的输入,其它合金成分和热处理温度作为网络的输出,根据要求的力学性能设计材料的合金成分含量及热处理条件,获得了满意的结果,为高性能材料设计提供了一定的理论辅助手段．     

An artificial neural network model for virtual Superpave asphalt mixture design

This study presents an artificial neural network (ANN) model to predict the asphalt mixture volumetrics at Superpave gyration levels. The input data-set needed by the algorithm is composed of gradation of the mix, bulk specific gravity of aggregates, low- and high-performance grade of the binder, binder content of the mix and the target number of gyrations (i.e. N_ini, N_des and N_max). The proposed ANN model uses a three-layer scaled conjugate gradient back-propagation (feed-forward) network. The ANN was trained using data obtained from numerous roads with a total of 1817 different mix designs. Results revealed that the ANN was able to predict V_a within V_a (measured) ± 1.0% range 85–93% of the time and within V_a (measured) ± 0.5% range 60–70% of the time. Currently with the developed ANN model, Superpave mix design can take approximately between 1.5 and 4.5 days, which corresponds to 3–6 days of savings.     

基于神经网络趋势分析

文章在分析研究了国内外现状的基础上 ,利用神经网络的非线性处理特性 ,提出了通过神经网络预测常见机械零件剩余寿命的方法 ,用实例验证了其有效性     

基于人工神经网络的氮化锆薄膜颜色预测模型

建立了氮化锆薄膜制备工艺参数与薄膜色度参数之间的人工神经网络预测模型，结果表明，预测结果与实测结果吻合，最大色差在5．45以内。利用所建立的模型研究了单个参数对薄膜颜色的影响规律，及多参数间交互作用与薄膜颜色的关系。并且利用神经网络根据加工要求反向预测工艺参数，从而实现了对加工参数的优化选择。     

一种改进的人工神经网络学习算法及其在超声检测中的应用

本文用多层感知器 (MLP)与误差反向传播算法 (errorback propagationalgorithm)构造训练人工神经网络 ,提出了新的误差反向传播改进算法。试验结果表明 ,改进的BP算法收敛速度较之常规BP算法明显加快 ,因而在工业现场的超声检测领域有广阔应用前景。     

基于人工神经网络的NiZn铁氧体结构不敏感性能的预测模型

为了系统研究配方对铁氧体电磁性能的影响,制备了一系列Mn2 、Ge4 和Si4 替代的NiZn铁氧体材料,建立了铁氧体配方与结构不敏感性能之间的人工神经网络预测模型.利用所建立的模型研究了ZnO对NiZn铁氧体3个结构不敏感性能居里温度、磁饱和强度及介电常数的影响规律,以及多个组分的交互作用.结果表明:模型的预测结果与实验结果吻合良好,二者的相对误差较小.ZnO含量的增加会导致铁氧体居里温度下降,但会提高饱和磁化强度和介电常数.NiO和ZnO的交互作用对铁氧体的结构不敏感性能影响明显.利用模型得到的铁氧体性能-成分等值线图对寻找最佳配方有较高参考价值.     

一类神经网络对连续函数的逼近

借助卷积逼近的工具研究前向神经网络对连续函数的逼近,构造了具有nd个神经元的一类神经网络,并证得用它逼近[0,1]d上的连续函数f(X)时,偏差是O{ω{f,n-d1+2}+n-d1+2‖f‖∞}.其中ω(f,δ)表示f(X)在[0,1]d上的连续模,‖f‖∞表示|f(X)|的极大值.     

模拟退火优化BP神经网络在航向控制系统中的应用

目前，广泛运用于神经网络中的误差反向传播算法（BP算法）训练时间较长，且易陷入局部最优．为了克服BP算法的固有缺陷，文中提出了在BP算法中加入模拟退火算法权因子．在航向控制系统中进行了仿真，数据显示该算法比单纯BP算法更能优化控制器性能参数和全局搜索能力，收敛速度更快，精度提高比较明显。     

A hybrid artificial neural network method with uniform design for structural optimization

This paper presents a new hybrid artificial neural network (ANN) method for structural optimization. The method involves the selection of training datasets for establishing an ANN model by uniform design method, approximation of the objective or constraint functions by the trained ANN model and yields solutions of structural optimization problems using the sequential quadratic programming method (SQP). In the proposed method, the use of the uniform design method can improve the quality of the selected training datasets, leading to a better performance of the ANN model. As a result, the ANN dramatically reduces the number of required trained datasets, and shows a good ability to approximate the objective or constraint functions and then provides an accurate estimation of the optimum solution. It is shown through three numerical examples that the proposed method provides accurate and computationally efficient estimates of the solutions of structural optimization problems.     

Radar signal design problem with neural network processing

Binary and ternary sequences with peaky autocorrelation, measured in terms of high discrimination and merit factor have been searched earlier, using optimization techniques. It is shown that the use of neural network processing of the return signal is much more advantageous. It opens up a new signal design problem, which is solved by an optimization technique called Hamming scan, for both binary and ternary sequences.     

基于改进神经网络的干点软测量

针对传统的神经网络收敛判断以模型的拟合精度为指标造成训练时间过长和过拟合等缺点,提出了一种改进神经网络(M-ANN).M-ANN将样本分成训练样本和校验样本,并提出了过拟合判据参数.通过训练样本采用误差反传算法对网络进行训练,训练过程中以模型对校验样本的预测性能为指标,通过过拟合判据参数的计算自适应地在获得具有最佳预测性能模型时终止网络训练.同时,针对影响初馏塔塔顶石脑油干点的因素众多且呈高度非线性的特征,应用M-ANN建立初顶石脑油干点软测量模型,获得模型的预测相对误差平方和均值比传统神经网络模型降低了27.5%.     

基于BP神经网络的吸运风机故障诊断

吸运风机是农业现代化生产机械联合收割机的重要组成设备.针对目前吸运风机经常出现的故障,收集了吸运风机故障征兆和其对应的故障类型.将故障样本数据和模糊神经网络相结合,并根据BP神经网络确定网络的输入和输出向量,对风机进行故障诊断,诊断结果与实际情况比较吻合.运用MATLAB实现神经网络故障诊断仿真,仿真结果表明诊断误差较小,输出向量与实际故障矩阵结果接近.     

基于人工神经网络的Mg-Zn-Zr-Ce合金热压缩变形研究

为研究含稀土元素铈的镁合金中高温流变行为,利用热模拟试验机对Mg-6Zn-0.5Zr-1.5Ce合金在变形温度523～673 K、应变速率0.001～1 s-1范围内进行热压缩实验.基于真应力真应变实验数据构建了单隐层前馈误差反向传播人工神经网络模型,利用该模型对ZK60-1.5Ce合金的流变应力行为进行预测,并分析了变形温度、应变速率与真应变对流变应力的影响.研究表明:Ce添加可显著细化晶粒;该镁合金的流变应力随变形温度降低和应变速率升高而增加;其流变应力行为可用双曲正弦函数进行描述,依据峰值应力拟合求得该合金的表观激活能为161.13 kJ/mol;变形温度和应变速率对流变应力的影响高于真应变.所建立的人工神经网络模型可以很好地描述该镁合金的流变应力,其预测值与实验数值吻合良好.     

A neural network decision-making model for job-shop scheduling

Scheduling in a job-shop system is a challenging task. Simulation modelling is a well-known approach for evaluating the scheduling plans of a job-shop system; however, it is costly and time-consuming, and developing a model and interpreting the results requires expertise. As an alternative, we have developed a neural network (NN) model focused on detailed scheduling that provides a versatile job-shop scheduling analysis framework for management to easily evaluate different possible scheduling scenarios based on internal or external constraints. A new approach is also proposed to enhance the quality of training data for better performance. Previous NN models in scheduling focus mainly on job sequencing and simple operations flow, and may not consider the complexities of real-world operations. The proposed model’s output proved statistically equivalent to the results of the simulation model. The study was accomplished using sensitivity analysis to measure the effectiveness of the input variables of the NN model and their impact on the output, revealing that the batch size variable had a significant impact on the scheduling results in comparison with other variables.     

用人工神经网络研究合金元素对NdFeB永磁体磁性能的影响

为了系统研究合金元素对Nd-Fe-Co-Zr-B系永磁合金磁性能的影响，采用均匀设计方法设计了Nd、Co、Zr和B的4因素6水平U18（6^4）试验方案，根据试验结果，建立了合金成分与磁性能之间的人工神经网络（ANN）预测模型。利用该预测模型获得的成分-性能的二维曲线、三维曲面及等高线图，研究了单个合金元素以及多元素间的交互作用对NdFeB磁体磁性能的影响规律。结果表明：预测结果与实测结果吻合良好，预测精度高；Nd、Zr为提高矫顽力Hcj而降低剩磁Br的元素；Co、B则对提高Br有利而对提高Hcj不利；合金元素对Hcj与Br的影响呈相反的趋势；元素间交互作用对磁性能影响显著。     

MATNET: A neural network for medial axis transformation

Abstract

This paper describes a novel neural network, called MATNET, to perform the medial axis transformation which is often used to extract a stick‐figure‐like representation from a binary object for pattern analysis or recognition. The MATNET is derived from the structure of the retina, which consists of five neural layers, namely, receptors, horizontal cells, bipolar cells, ganglion cells, and response. In principle, the horizontal cell is implemented for distance computation; the bipolar cell (B‐net) and the ganglion cell (G‐net) are implemented for calculation of local minimum and local maximum, respectively. The B‐net and G‐net are concerned with the maximal neural network (Maxnet). The properties of Maxnet are also discussed. Experimental results show that the MATNET performs reasonably.     

Applying likelihood on Hopfield neural network for radar tracking

Abstract

The multiple‐target tracking (MTT) algorithm plays an important role in radar systems. Data association is the most important technique to solve the tracking problems associating dense measurements with existing tracks. A new approach applying Likelihood to measurements and existing tracks in a radar system based on Neural Network computation is investigated in this paper. The proposed algorithm will solve both the data association and the target tracking problems simultaneously. With this approach, the matching between radar measurements and existing target tracks can achieve global relevance. Computer simulation results indicate the ability of this algorithm to keep track of targets under various conditions.     

一种心音小波神经网络识别系统

设计一种心音小波神经网络识别系统,将心音特征抽取、有针对性的神经网络层次化架构和分类识别融合一体,以解决复杂条件下的心音分类识别问题。提出基于心音小波神经网络的识别模型,讨论如何构造心音小波和心音小波神经网络的方法,重点讨论在网络结构的隐含层中引入心音小波作为激活函数的算法,从而获得一种把心音的针对性学习和心音识别技术高度融合的心音小波神经网络识别系统。通过选取正常心音信号与早搏心音信号作为实验对象,验证了心音小波神经网络识别系统的有效性和实用性,并且通过与morlet和Mexican-hat小波神经网络识别系统相比较,证明心音小波神经网络识别系统在收敛性、算法速度上呈现明显的优越性。     

A neural network approach to microwave imaging

We present a neural network approach to microwave imaging for medical diagnosis. The problem is to reconstruct the complex permittivity of the biological tissues illuminated by the transverse magnetic (TM) incident waves. In order to avoid the inherent ill‐posedness of the inverse scattering problem, we introduce a stochastic process based on Markov random field and a priori knowledge. A coupled gradient neural network is proposed to deal with the mixed‐variable problem because the reconstructed dielectric permittivities are continuous complex variables and the line processes, which can preserve the edges of the reconstructed image, are binary variables. We report the numerical results of a simple human forearm model. We also point out the advantages and the limitations of this method. © 2001 John Wiley & Sons, Inc. Int J Imaging Syst Technol 11, 159–163, 2000