神经网络

Under fully considering the time relativity characteristic of double trends time series of the predicted data source (cigarette saleroom of some Yunnan tobacco company), based on CBP neural network, we set up CCBP predict model.

在充分考虑了趋势预测目标数据源(云南某烟草公司卷烟销售)的二重趋势时间序列的时间相关性的特点下，以神经网络中CBP神经网络为基础建立CCBP网络预测模型。

At the end of this thesis, the problem how to make an Expert System based on the model of Artificial Neural Networks was discussed, and some advice of model ameliorating was given.

最后，本文还对将所建立的人工神经网络模型进一步完善成为神经网络专家系统的问题进行了探讨，并提出了网络及算法进一步改进的方案。

At the end of this thesis, the problem how to makean Expert System based on the model of Artificial Neural Networks was discussed,and some advice of model ameliorating was given.

最后，本文还对将所建立的人工神经网络模型进一步完善成为神经网络专家系统的问题进行了探讨，并提出了网络及算法进一步改进的方案。

Tocopherol has very high physiological activity. Soybean deodorized sludge contains plentiful natural tocopherol. For extracting tocopherol from soybean deodorized sludge, solubility of α tocopherol in supercritical CO 2 is measured. The solubility data are correlated by RBF neural networks model and chrastil model. Relative error of chrastil model is 25.36%. By means of learning and training on RBF neural networks model, the average error of training set is 0.23% and that of testing set is 6.48%....

生育酚有很高的生理活性，油脂生产中得到的脱臭馏出物含有丰富的天然生育酚作为萃取生育酚的基础，对甲酯化油脂脱臭馏出物中α-生育酚在超临界CO2 中的溶解度进行了测试，并用Chrastil分子缔合模型和RBF神经网络模型对溶解度数据进行了拟合 Chrastil分子缔合模型的相对误差为 2 5.36%对于RBF神经网络模型，经过网络学习和训练，训练集平均误差仅为 0 。2 3%，测试集误差为 6.4 8%，效果比较理想

2Based on the nonlinear theory, we found the reservoir seismic nonlinearprediction and evaluation method technology, it is constituted by three nonlinearmethods and the technologys of the fracture prediction, the seismic inversion and thereservoir synthesis prediction evaluation: The seismic nonlinear prediction ofreservoir fracture is one new method which is composed with the phase spacereconstruction, the nonlinear parameters pick-up technology and the syntheticprediction evaluation method. The reservoir seismic high resolution nonlinearinversion is a new seismic inversion way which the BP algorithm is embeded in theauto-adapted genetic algorithms interior to have the predominances of neural networktechnology and genetic algorithms, it adopts the new embedded GA-BP mixalgorithms and the nonlinear mapping technology, and realizes the inversionautomatically, obtains the high resolution seismic inversion profile. The reservoirseismic nonlinear synthesis prediction and evaluation is a new method which iscombined organically of genetic algorithms and adaptive neural fuzzy inferencesystem, it will optimize the new seismic attribute space which are processed to take the input, uses the new adaptive mix algorithm which GDand LSE(least-square estimation) mix algorithms of ANFIS network insert to the GAinterior and taboo search algorithms is added to the intercrossed operation place,the simulation of the evaluation parameters is used quantitative evaluation guide lineto the reservoir quality and oil-gas distribution.

2基于非线性理论，创建了储层地震非线性预测与评价方法技术，它是由裂缝预测、地震反演和储层综合预测与评价等三大非线性方法与技术组成：储层裂缝地震非线性预测是由相空间重建、非线性参数提取与预测技术及综合评价方法组成的一种新型裂缝预测方法；储层地震高分辨率非线性反演是将BP算法嵌入自适应遗传算法内部所构成的集遗传算法和神经网络技术优势于一体的新的地震反演方法，它采用嵌入式新的混合算法及非线性映射技术，自动实现反演，获得高分辨率地震反演剖面；储层地震非线性综合预测与评价是由遗传算法与自适应神经网络—模糊推理系统有机地相结合而产生的储层预测与评价的新方法，它将优化处理所形成的新地震属性参数空间作为输入，采用将ANFIS网络中的混合算法嵌入到GA算法内部与禁忌搜索算法加在交叉操作处产生新的自适应混合算法，将综合评价参数作为储层品质和含油气性的定量评价指标。

But because of some bugs, they cannot be applied more widely. Base on these, a new control method called ""self-organized fuzzy control base on dynamic structure neural network ""is proposed in this article. It syncretizes the strongpoint of Fuzzy control and Neural Net control, and attain optimization control with the adjusting the number of the net's layer.

针对这种时变大时滞被控对象的特点，目前主要采用Smith、Dahlin和IMC控制，这些控制方法在某方面可能得到一定应用，但由于其本身存在某些缺陷，因此不可能得到广泛应用，笔者在此基础上提出了一种基于动态结构的神经网络自组织模糊控制，它融合了神经网络和模糊控制的优点，并能通过在线调整网络的层数以达到最优控制的效果。

On the basis of network training, neural network computing models of maximum fluctuation of water surface in ship lift chamber with tumble gate operating (NNCM-MFWS-SLC-TGO) were built through the principle of artificial neural networks.

运用人工神经网络原理，在通过大量学习样本对网络进行训练的基础上，建立了启闭时升船机承船厢内水面最大升降值的神经网络计算模型(NNCM—MFWS-SLC-TGO)。

The deviation compensation algorithm put forward here is analyzed in theory and the effect of this compensation algorithm is testified based on experimental results that the learning speed of this network can be sped up markedly and all attitude magnetic heading error can be greatly reduced. RBF neural network is quite effective and superior to BP neural network.

在分析算法原理的基础上进行了实验仿真，结果表明：采用RBF神经网络在明显提高网络收敛速度的基础上，大大减小了全姿态磁航向误差，校正效果优于BP神经网络。

Improving vertical resolution for density curves is by means of restructuring the response from large and small spacing sondes and borehole effect correction factor and coherence factor are used. Form the principle of GR logging, calculate volumes in investigating areas taken by borehole, formation, up and low shoulder beds, along with distance between center of gravity of the four unit bodies and detector. Based on the above, the weighting function of each unit body to GR signal contributions are established, thus realizing processing of GR logs with high resolution. The resistivity logging model uses "resolution matching" method to improve the resolution of the logs. After the process, PNN model is made to recognize the oil-gas-water bed. Compared with BPNN, the net work is featured swiftness, accuracy and ease of practical application.

提高密度测井曲线垂直分辨率是通过对重新组合长、短源距探测器响应的算法来提高补偿密度曲线的垂直分辨率，并加入井眼影响校正因子和相关性因子；自然伽马测井模型从基本原理出发，扣除井的影响，计算井筒、地层及其上下围岩所占探测球域体积及其重心到探测器的距离，以此建立各单元体对伽马测井信号贡献的权函数，通过权函数逐层计算不同厚度地层条件下各的单元体权因子，实现自然伽马曲线高分辨率处理；电阻率校正模型应用&分配率匹配&模型，应用低分辨率曲线提高为高分辨率曲线：曲线校正后，利用常规测井资料并且结合校正后的密度、伽马、电阻率等三个对判断油水层有利的指标，通过概率神经网络模型识别储层流体性质，较以前的模型，不仅极大的提高识别正确率，同时与普通神经网络相比较，提高了网络的运行速度和实用性。

On the basis of the algorithm of BP neural network,the network's excitation function has been improved, and an ANN-based Seismic Response recognising system for pile foundation bridge pier has been developed according to the characters of the seismic response of the pier.

在原BP神经网络基本算法的基础上，按研究桩基桥墩地震响应的特点，改进了网络的激励函数，开发了适用于桩基桥墩地震响应的神经网络辨识系统。

We have no common name for a mime of Sophron or Xenarchus and a Socratic Conversation; and we should still be without one even if the imitation in the two instances were in trimeters or elegiacs or some other kind of verse--though it is the way with people to tack on 'poet' to the name of a metre, and talk of elegiac-poets and epic-poets, thinking that they call them poets not by reason of the imitative nature of their work, but indiscriminately by reason of the metre they write in.

索夫农 、森那库斯和苏格拉底式的对话采用的模仿没有一个公共的名称；三音步诗、挽歌体或其他类型的诗的模仿也没有——人们把&诗人&这一名词和格律名称结合到一起，称之为挽歌体诗人或者史诗诗人，他们被称为诗人，似乎只是因为遵守格律写作，而非他们作品的模仿本质。

The relationship between communicative competence and grammar teaching should be that of the ends and the means.

交际能力和语法的关系应该是目标与途径的关系。