多偏移遥感图像的BP神经网络亚像元定位

提出了一种借助多偏移遥感图像来改进基于BP神经网络(BPNN)的亚像元定位新方法.不同于原BPNN方法使用单幅低空间分辨率观测图像,新方法利用多幅带有亚像元偏移的低空间分辨图像来确定亚像元属于各类的概率,然后根据概率值和地物覆盖比例确定亚像元类别,以降低BPNN定位模型中的不确定性和误差.实验表明,提出方法在视觉和定量评价上,均能获得更高精度的亚像元定位结果,验证了提出方法的有效性.     

神经网络优化ITO废靶酸浸回收铟(英文)

以提高铟浸出率为目标,通过工艺实验结合神经网络研究ITO废靶酸浸的优化工艺。首先,固定其他工艺因素,进行单因素如残酸度、氧化剂加入量、酸浸温度及时间对铟浸出率影响的实验研究。结果表明,增大残酸度可提高铟浸出率;氧化剂的加入可明显提高铟浸出率,但增加到一定程度后浸出率提高不明显;升高温度可明显提高浸出率,但继续升高则会降低铟浸出率;延长浸出时间也可提高铟浸出率。通过反向传播算法的人工神经网络(BPNN)研究多因素的综合作用对铟浸出率的影响规律,预测值与实验值相差很小,表明所建立的BP模型铟浸出率能比较准确地预测。最终,通过BPNN预测以及实验验证,获得高达99.5%浸出率的工艺参数:残酸度50~60 g/L、氧化剂含量10%、浸出温度70°C和浸出时间2 h。     

基于BP神经网络的热风炉群煤气消耗量预测

热风炉是副产煤气消耗大户,热风炉群的煤气消耗数据无规律性、波动剧烈,预测难度较大.针对热风炉群煤气消耗量难以直接预测的问题,提出一种基于BP神经网络的热风炉群煤气消耗量预测方法.该方法将热风炉群的煤气消耗数据分解为单座热风炉的煤气消耗数据,利用单座热风炉周期性煤气消耗特性,将利用BP网络模型预测出的各座热风炉煤气消耗数...     

The Relation of Kinetics of Coupling of Developer Soluble Colour Couplers to Structure and Photographic Activity

The ratio of Coupling Rate Constant: Deamination Rale Constant (k₂, k₁) for a considerable number of magenta, yellow and cyan developer soluble colour couplers has been determined by competition between deamination and coupling. The determinations were made in a medium approximating to the colour developer using 5-diethylamino-2-aminotoluene as the developing agent. pK measurements showed that all the colour couplers were fully ionised in this medium. The photographic data have been shown to fit the empirical equation, γ = A log t + B where γ = Contrast, t = Time of Development in minutes, A = Activity Constant and B = Inducticn Constant, k₂, k₁ and A values have been shown to be related by the equations, A=3.55 e^{–5.05×102(K₁/K₂) for oxidation by K₃Fe(CN)₆} A=3.55 e^{–5.05×102(K₁/K₂}) for oxidation by AgCl.

These equations show that the photographic activity of a colour coupler is related to its rate of coupling with oxidised developer and increases with increase in k₂/k₁. Further they also show that there is a limiting value for A of 3·55. The results also show a broad indication that log k₂/k₁, is proportional to the pK of the colour couplers. This indicates that electron releasing substituents would increase the coupling rate and hence the activity of the colour coupler.

Of 21 colour couplers studied approximately one-third did not obey the above equations. Experiments have shown that the anomalies arise from azomethine dye destruction in the film due to attack by (a) Na₂SO₃, and NH₂OH under alkaline conditions and (b) excess pyrazolone colour coupler in the case of magenta colour couplers.     

基于同源信息的BP神经网络在故障诊断中的应用

阐述了同源信息融合和BP神经网络（BPNN）关键技术,提出了基于同源信息融合的BP神经网络诊断方法;以旋转机械为例进行了MATLAB编程,得到了误差小和速度快的结论。     

基于深度信念网络的语音服务文本分类

在线人工语音服务已经在各种商业活动中展开,为了提供更好的客户服务就必须对语音服务质量进行有效的评估。目的就是将人工语音服务利用语音识别技术转化为文本,再进行有效的分类评估。常用文本分类模型有朴素贝叶斯、KNN、BP神经网络、支持向量机等模型,这些模型比较依赖于语音文本预处理后的特征表示,并且容易出现维数灾难、局部最优、训练时间长问题。而深度信念网络模型（DBN）可以从文本预处理后的特征表示中学习到更具有本质含义的特征表示,便于分类器分类,且避免以上模型的不足。在人工服务语音文本化后,通过深度信念网络模型转换特征表示再进行分类,最终的分类效果比上述分类模型直接利用文本的特征表示进行分类效果略微提高。     

Imputing manufacturing material in data mining

Data plays a vital role as a source of information to organizations, especially in times of information and technology. One encounters a not-so-perfect database from which data is missing, and the results obtained from such a database may provide biased or misleading solutions. Therefore, imputing missing data to a database has been regarded as one of the major steps in data mining. The present research used different methods of data mining to construct imputative models in accordance with different types of missing data. When the missing data is continuous, regression models and Neural Networks are used to build imputative models. For the categorical missing data, the logistic regression model, neural network, C5.0 and CART are employed to construct imputative models. The results showed that the regression model was found to provide the best estimate of continuous missing data; but for categorical missing data, the C5.0 model proved the best method.     

A systematic synthesis framework for extractive distillation processes

An effective extractive distillation process depends on the choice of the extractive agent. In this contribution, heuristic rules for entrainer selection and the design of entrainers through computer-aided molecular design are reviewed. The potential of the generated alternatives is then evaluated by their selectivity at infinite dilution and by the rectification body method (RBM). It is shown that a screening based on selectivity alone is not sufficient and could possibly lead to an unfavorable entrainer choice. The minimum entrainer flowrate and the minimum energy demand, calculated from the RBM, allow a more comprehensive evaluation of different entrainer alternatives. In a third step a rigorous mixed-integer optimization of the entire extractive flowsheet for the remaining entrainer candidates is executed to fix the remaining design degrees of freedom and to determine the best entrainer. Since a number of alternative entrainers have already been eliminated, only a few optimizations are necessary. These steps form a framework which facilitates the systematic generation and evaluation of entrainer alternatives. The suggested synthesis framework is illustrated with a case study where acetone and methanol are to be separated.     

Neural network approach to recognize the grid patterns in experimental mechanics

In this paper, we develop a neural network system that applies to the work of experimental mechanics. Two procedures are presented: one is the color calibration that used to filter 256 colors into three colors by adaptive-competitive learning network (ACLN). The other is the pattern recognition that is used to recognize the grid patterns and search the intersection points of grids by back propagation neural network (BPNN). In this paper, plane extrusion through a cosine-curved die is adopted to illustrate as example. The results show that the correlative points between the calculation and experiment are shown quite well agreement.