基于遗传神经网络的网格资源预测模型

研究网格资源预测问题,网格资源具有非线性、混沌变化特点,传统BP神经网络具有局部极小、收敛速度慢等缺陷,预测精度较低。为提高了网格资源预测精度,提出一种基于遗传神经网络的网格资源预测模型。利用遗传算法对BP神经网络的权值和阈值进行优化,然后采用BP神经网络对网格资源建立预测模型,最后采用网格资源时间序列进行有效性仿真。仿真结果表明,遗传神经网络有效地解决了传统BP神经网络的不足,提高了网格资源的预测精度,降低了预测误差,十分适合于非线性、混沌的网格资源时间序列预测。     

原发性高血压病靶器官损害的预测仿真研究

研究原发性高血压病靶器官损害准确预测问题,由于原发性高血压病靶器官损害的影响因子之间存在着高度冗余和非线性,导致传统方法预测准确率低,为提高预测精度,构建了一种改进的基因表达式编程算法的原发性高血压病靶器官损害预测模型,首先设计了适合靶器官损害预测的适应度函数,然后为克服早熟现象提出了以平均适应度形式自适应的改变重组算子概率和变异算子概率,通过数据训练得到靶器官损害的2年预测模型。采用对山东中医药大学第二附属医院原发性高血压病历史数据进行验证性仿真,结果表明,方法的预测精度更高,在原发性高血压病靶器官损害的预测中有着广泛的应用前景。     

遗传优化的GNNM在瓦斯涌出量预测中的应用

研究矿井瓦斯涌出量准确预测一直是煤矿安全生产中重点关注的问题。煤层瓦斯爆炸因受开发环境、矿层深度、天气等因素的影响,造成与瓦斯涌出量增大而引起的。针对传统预测模型在矿井瓦斯涌出量预测中存在建模困难、收敛速度慢、要求历史数据量大的问题,提出了一种遗传优化的灰色神经网络预测模型。模型利用灰色系统对数据量要求低的特点,将灰色系统理论与神经网络有机结合起来,建立灰色神经网络模型。并采用遗传算法对所建立模型的权值和阈值进行优化。采用模型对矿井瓦斯涌出量进行预测,实验表明,遗传优化的灰色神经网络模型,可以简化系统建模,并能提高瓦斯涌出量预测精度,有一定的实用价值。     

组合模型在降雨量预测中的应用

研究降雨量准确预测问题,降水量的变化既受大气环流、地形、气压、气候带等各种环境因子的影响,降水量的动态特征呈现复杂非线性和各种干扰因素,预测不可能准确。传统预测模型难以对其进行准确预测,预测精度低。为提高降雨量的预测精度,提出一种组合模型的降雨量预测模型。首先采用小波分析将降雨量数据进行分解成线性和非线性部分,然后分别采用ARIMA和RBF神经网络模型对其进行预测,最后采用小波重构线性和非线性预测结果,得到降雨量最终预测结果。仿真结果表明,相对于传统预测模型,组合模型提高了降雨量预测精度,预测结果可以帮助农业、水利部门提高防治旱涝灾害的科学依据。     

对地观测卫星成像区域仿真研究

为了发挥对地观测卫星在大面积、实时、快速获取地表数据方面的优势,用户需要事先掌握对于自己感兴趣区域上空的卫星过境情况和成像能力。本文通过对卫星轨道预报模型与算法、卫星地面成像区域边界轨迹计算模型和算法的探讨与实践,得到了一些对卫星成像区域仿真有益的经验。试验证明,对于无侧视情况下,使用较旧的卫星轨道根数来计算卫星成像区域,仍能满足用户对于成像区域预报的精度要求。     

采用形态神经网络背景自适应预测的图像弱小目标检测*

提出一种有效的背景杂波预测形态神经网络模型,用于检测图像数据中的弱小目标.目标被假设为只有很小的空域扩展度,而且淹没于强背景杂波干扰中.通过形态神经网络,杂波背景被准确地估计并从输入数据中去除,只剩下残留噪声和目标信号.采用扩展输入层数据的办法修正了传统的形态开、闭运算三层前馈BP网络模型.为了跟踪包含不同子结构的复杂背景,原始图像被划分为多个子块,并在相应的子块中选择训练样本对结构元进行优化.对真实图像数据的计算机仿真表明该算法在性能上优于其他传统算法.     

Prediction of multicomponent adsorption equilibrium of gas mixtures including supercritical components

A new model to predict the adsorption equilibrium of mixtures containing supercritical components was presented. The difference in the adsorption mechanism between the supercritical components and the condensable components was accounted for in the new model. Lateral interaction in the adsorbed phase was also considered in the local isotherm. The new model was verified by the experimental data of the mixture CH₄/CO₂/C₂H₆ on activated carbon JX101 collected for 283-313 K and pressures up to . It was also verified with the experimental data published previously in literature. Considerable improvement in the prediction precision was achieved in using the new model to predict the adsorption of mixtures on activated carbon.     

Statistical prediction of fracture parameters of concrete and implications for choice of testing standard

This article shows how the fracture energy of concrete, as well as other fracture parameters such as the effective length of the fracture process zone, critical crack-tip opening displacement and the fracture toughness, can be approximately predicted from the standard compression strength, maximum aggregate size, water-cement ratio, and aggregate type (river or crushed). A database, consisting of 238 test data, is extracted from the literature and tabulated, and approximate mean prediction formulae calibrated by this very large data set are developed. A distinction is made between (a) the fracture energy, G_f, corresponding to the area under the initial tangent of the softening stress-separation curve of cohesive crack model, which governs the maximum loads of structures and is obtained by the size effect method (SEM) or related methods (Jenq-Shah two-parameter method and Karihaloo's effective crack model, ECM) and (b) the fracture energy, G_F, corresponding to the area under the complete stress-separation curve, which governs large postpeak deflections of structures and is obtained by the work-of-fracture method (WFM) proposed for concrete by Hillerborg. The coefficients of variation of the errors in the prediction formulae compared to the test data are calculated; they are 17.8% for G_f and 29.9% for G_F, the latter being 1.67 times higher than the former. Although the errors of the prediction formulae taking into account the differences among different concretes doubtless contribute significantly to the high values of these coefficients of variation, there is no reason for a bias of the statistics in favor of G_f or G_F. Thus, the statistics indicate that the fracture energy based on the measurements in the maximum load region is much less uncertain than that based on the measurement of the tail of the postpeak load-deflection curve. While both G_f and G_F are needed for accurate structural analysis, it follows that if the testing standard should measure, for the sake of simplicity, only one of these two fracture energies, then G_f is preferable.     

基于GM（1，1）和线性回归模型在印刷包衬压缩变形中的数据预测研究

运用线性回归对预测数据进行分析,剔除异常数据,用GM（1,1）模型进行预测,有效降低了数据相对误差,提高了预测数据的精度。选用印刷包衬压缩变形的压缩变形量值,用线性回归进行数据分析并剔除异常数据后用GM（1,1）进行预测,使得预测数据具有更高的准确性和适应性。实验及仿真结果表明,经过前期数据分析整理后的灰色预测模型,其预测期望值远优于单纯的回归模型和GM（1,1）模型。     

Prediction model of compressive strength development of fly-ash concrete

Based on experimental results concerning the compressive strength development of concrete containing fly ash, the authors derived an estimation equation for compressive strength development. The equation can express coefficient , which indicates the activity of fly ash as a binder, in the form of a function of age, fly-ash content, and Blaine specific surface area of fly ash.

This equation is capable of explaining the increases in the early strength due to fly ash in place of part of fine aggregate, the decreases in the early strength due to fly ash in place of part of cement, the increases in the long-term strength due to pozzolanic reaction, the relationship between the fly-ash replacement ratio and the ratio of strength increase/decrease, and the effect of fly ash's Blaine specific surface area on the strength.