Second , a self - study fast bp algorithm is purposed in the paper considering that cold - rolling mill ' s tension control is a fast time - critical system . two design methods of controller were purposed too in this paper . the first is based on robust control 其次，针对冷连轧机为一快速性、实时性系统，本文提出了基于回归神经网络自适应快速bp算法，通过改进的bp算法，提高了系统的实时性。
Abstract : artifical intelligence methods are implemented to simulate thebehaviors of axially and laterally loaded piles using the field observation tests data obtain ed f rom the drilled shafts and driven piles . the optimal neural network model is deve loped using only simple input data of spt - n values and piles ' geometrical featu r es etc . . the analysis for r . c piles of some projects is performed adopting the bp n n and grnn models respectively , and the obtained predicated results are compared w ith the data from conventional design method . it demonstrated the obvious advanta ges of neural networks in the design of pile foundations over the traditional me thods . this paper has an important practical significance and a referential worth iness in the design of pile foundations 文摘:根据钻孔桩和打击桩的原型试验观测的数据,运用人工智能方法对横向承载桩和轴向承载桩的工作特性进行模拟,并利用标准贯入试验( spt - n )值和桩的几何特性等简单的输入数据,开发出相应的优化神经网络模型;然后,运用反向传播神经网络模型和广义回归神经网络模型分别对某工程的钢筋混凝土桩进行分析,并将求得的预测结果与常规设计法的结果进行比较,结果表明神经网络方法比传统方法有明显的优越性,在实际工程设计中具有重要的参考价值和现实意义。
Ann methods are feasible for the verification measurements in nuclear safeguards . experimental data sets have been used to study the performance of neural networks involving radial basis function neural network and generalized regression neural network ( grnn ) . the optimization of the parameter spreads have been given and the analysis error of grnn no more than 0 . 2 % 分析结果表明，使用泛化能力较高的混合训练集训练神经网络，网络给出的富集度值与标准样品的标称值之间的相对差异小于13 % ；使用泛化能力相对较弱的分组训练集训练神经网络，网络给出的分析结果的不确定度小于2 % ；使用分组训练集和广义回归神经网络，网络给出的分析结果的不确定度小于0
Abstract : on the basis of the experimental data of microstructure and strength for gray cast iron with high carbon equivalent , the adapted fuzzy neural network model of relationship between microstructure and strength for predicting the strength of gray cast iron has been developed by using adaptive neural - fuzzy inference method . comparing with the models based on multiple statistic analysis , fuzzy regression or generalized regression neural network , it shows better learning precision and generalization 文摘:以高碳当量灰铸铁组织-强度实验数据为基础，用自适应模糊推理方法，建立了灰铸铁强度自适应模糊神经网络预测模型，与多元线性回归、模糊回归和广义回归神经网络模型相比，该模型学习精度高且具有较好的泛化性。
1 . this thesis puts forward a new model ? the model of crop ' s water rising and vaporizing regression nerve networks ( nn ) . this model adds a deviation cell based on existing bp ( nn ) the effect of the deviation cell is to amend training errors of the nn , to increase convergence speed of the nn , and to reach expectant objective 本文主要完成了以下几个方面的内容： 1 、提出了一个新型的模型? ?作物蒸腾蒸发回归神经网络决策模型，该模型以已有的bp神经网络为基础，增加了一个偏差单元，该偏差单元的作用是修正网络的训练误差、提高网络的收敛速度，从而达到预期的目的。
This paper discuss a modeling and predicting means for nonlinear systems proceeding from nonlinear systems modeling and predicting theory , whch is based on drnn model . this means overcomes the fact that ar model is used only in linear systems , at the same time it connects itself with approximation theory symbolic statistics and conjugate gradient algorithm , and formulate a system of large watercrafts motion modeling and predicting which is based on drnn model , and simulate it 本论文从非线性系统建模与预报的理论及应用观点出发，系统地阐述了一类适用于非线性系统的建模预报方法? ?基于drnn模型的建模预报方法，克服了ar模型仅局限于线性的情况，同时结合逼近论、数理统计等知识，运用共轭梯度算法，提出并建立了基于对角回归神经网络的大型舰船运动建模预报系统，并进行了仿真。
The bic method generalized from ar model was adopted to determine the number of input neurons in grnn prediction model . the grnn was applied to single - step and multi - step ahead prediction of the vibration time series of a rotating machine , and its performance was compared with that of 3 - layers perceptrons network with error back propagation training algorithm ( bpnn ) . it is indicated that the grnn is more appropriate for prediction of time series than the bpnn , and the performance of grnn is qualified even with sparse sample data 研究了基于广义回归神经网络( grnn )的大型旋转机械振动状态预测，提出了应用bic准则确定grnn预测模型输入神经元数目的方法，将grnn用于大型机组振动峰?峰值时间序列的预测，与采用误差反向传播学习算法的三层前馈感知器网络( bpnn )的预测结果对比表明， grnn的预测性能优于bpnn ，而且，即使样本数据稀少，也能获得满意的预测结果。
