泡沫分离技术在扇贝育苗上的应用研究

本文从泡沫分离后有机物颗粒的大小及含量入手，进行泡沫分离后有机物及胶体物质含量和物理特性的分析，研究了泡沫分离后的水质变化和对幼虫生长发育的影响，取得比较好的效果，可明显改善育苗水质，从而提高幼虫成活率和育苗成功率．     

Evaluation of the ability of an artificial neural network model to simulate the input-output responses of a large karstic aquifer: the La Rochefoucauld aquifer (Charente,France)

The ability of artificial neural networks (ANN) to model the rainfall-discharge relationships of karstic aquifers has been studied in the La Rochefoucauld karst system, south-west France, which supplies water to the city of Angoulême. A neural networks model was developed based on MLP (multi-layer perceptron) networks and the Levenberg-Marquardt optimization algorithm. Raw rainfall data were used without transformation into effective rainfall. This allowed for the elimination of certain non-verifiable simplifying assumptions and their subsequent introduction into the modeling procedure. The raw rainfall and discharge data were divided into three groups for the training, the validation and the prediction test of the ANN model. The training and validation phases led to a very satisfactory calibration of the ANN model. The attempt to predict discharges showed that the ANN model is able to simulate the karstic aquifer discharges. The shape of the simulated hydrographs was found to be similar to that of the actual hydrographs. These encouraging results make it possible to consider interesting and new prospects for the modeling of karstic aquifers, which are highly non-linear systems.

---

Resumen Se ha estudiado la capacidad de las redes artificiales neurales (ANN) para modelizar las relaciones de lluvia-descarga de acuíferos kársticos en el sistema kárstico La Rochefocauld, al suroeste de Francia, el cual abastece de agua a la ciudad de Angoulême. Se desarrolló un modelo de redes neurales en base a redes MLP (Perceptron Multi-Capas) y el algoritmo de optimización Levenberg-Marquardt. Se utilizaron datos de lluvia sin la transformación hacia lluvia efectiva. Esto permitió la eliminación de ciertos supuestos simplificadores no verificables y su subsiguiente introducción en el procedimiento de modelizado. Los datos brutos de descarga y lluvia se dividieron en 3 grupos para la preparación, validación y la prueba de predicción del modelo ANN. Las fases de preparación y validación llevaron a una calibración muy satisfactoria del modelo ANN. El intento por predecir descargas mostró que el modelo ANN es capaz de simular las descargas del acuífero kárstico. Se encontró que la forma de los hidrogramas sintéticos es similar a la de los hidrogramas reales. Estos resultados alentadores hacen posible considerar prospectos nuevos e interesantes para el modelizado de acuíferos kársticos los cuales son sistemas altamente no-lineares.

---

Résumé L’aptitude des réseaux de neurones artificiels (RNA) à modéliser les relations pluie-débit des aquifères karstiques a été évaluée sur le karst de La Rochefoucauld (Sud-Ouest de la France), qui fournit l’eau potable à la capitale régionale Angoulême. Un modèle RNA a été développé à cet effet, basé sur les réseaux PMC (Perceptron Multicouche) et l’algorithme d’optimisation de Levenberg-Marquardt. Les données de pluie utilisées concernent la pluie brute, sans transformation en pluie efficace, ce qui permet de s’affranchir de certaines hypothèses simplificatrices non vérifiables. Les données de pluie brute et de débit ont été réparties en 3 groupes pour l’apprentissage, la validation et le test de prédiction du RNA. Les phases d’apprentissage et de validation ont permis d’aboutir à une calibration très satisfaisante du modèle RNA. La tentative de prédiction a montré que le RNA est apte à simuler les débits de l’aquifère karstique à partir de la pluie brute. La forme des hydrogrammes simulés est semblable à celle des hydrogrammes réels. Les résultats obtenus sont très encourageants et permettent d’envisager des perspectives intéressantes et nouvelles de modélisation des aquifères karstiques, qui sont des systèmes hautement non-linéaires.

     

Risk degree of debris flow applying neural networks

A number of methods for prediction of debris flows have been studied. However, the successful prediction ratios of debris flows cannot always maintain a stable and reliable level. The objective of this study is to present a stable and reliable analytical model for risk degree predictions of debris flows. This study proposes an Artificial Neural Networks (ANN) model that was constructed by seven significant factors using back-propagation (BP) algorithm. These seven factors include (1) length of creek, (2) average slope, (3) effective watershed area, (4) shape coefficient, (5) median size of soil grain, (6) effective cumulative rainfall, and (7) effective rainfall intensity. A total of 171 potential cases of debris flows collected in eastern Taiwan were fed into the ANN model for training and testing. The average ratio of successful prediction reaching 99.12% demonstrates that the presented ANN model with seven significant factors can provide a highly stable and reliable result for the prediction of debris flows in hazard mitigation and guarding systems.     

Creation of an artificial carbonate sand

This paper deals with the efficacy of creating artificial carbonate sand from crushed chalk to model the natural carbonate sand. Direct drained shear using 100 mm shear box and compressibility tests using 150 mm Rowe cell have been performed on artificial carbonate sand in order to determine the shear stress-strain and compressibility characteristics of the artificial material. These are compared with data from natural carbonate sands given in the literature. The results obtained demonstrate that shear stress-strain and compressibility characteristics of the artificial material closely mirror that of natural carbonate soils. The artificial carbonate sand shows generally similar characteristics, which are in the range of the natural carbonate material.     

A neural network approach for attenuation relationships: An application using strong ground motion data from Turkey

This paper presents an application of neural network approach for the prediction of peak ground acceleration (PGA) using the strong motion data from Turkey, as a soft computing technique to remove uncertainties in attenuation equations. A training algorithm based on the Fletcher–Reeves conjugate gradient back-propagation was developed and employed for three sample sets of strong ground motion. The input variables in the constructed artificial neural network (ANN) model were the magnitude, the source-to-site distance and the site conditions, and the output was the PGA. The generalization capability of ANN algorithms was tested with the same training data. To demonstrate the authenticity of this approach, the network predictions were compared with the ones from regressions for the corresponding attenuation equations. The results indicated that the fitting between the predicted PGA values by the networks and the observed ones yielded high correlation coefficients (R²). In addition, comparisons of the correlations by the ANN and the regression method showed that the ANN approach performed better than the regression. Even though the developed ANN models suffered from optimal configuration about the generalization capability, they can be conservatively used to well understand the influence of input parameters for the PGA predictions.     

人工神经网络在徐州岩溶地面塌陷评价中的应用

岩溶地面塌陷的影响因素很多,发展过程也复杂。在众多的对岩溶地面塌陷的评价方法中,神经网络具有自学习、自适应与高度非线性映射的特点,是一种非常有效的评价手段。在徐州岩溶石地面塌陷的评价中,成功地运用了人工神经网络技术,它具有的强大非线性映射能力,能够建立评价因子和评价对象之间的关系,正确选取评价因子,避免主观判断取值,从而得出可靠的预测模型和岩溶塌陷危险性分区图。     

单多普勒雷达在山区火箭人工增雨中的应用

利用单多普勒雷达资料产品对2003年8月3日和8月9日鄂西山区两次火箭增雨的作业条件和效果进行分析。分析表明:依据多普勒雷达资料,利用山区迎风坡地形对云系的抬升作用,适时开展火箭作业进行人工催化,可以达到显著的增雨效果。     

石漠化的成因机理与防治对策

从石漠化的研究现状出发，依据我国西南岩溶石山地区石漠化特征，着重分析其主要控制介质为岩性、构造、地貌、气候、岩溶、地下水、植被等自然介质因子与森林砍伐、高坡开垦等人为介质因子．石漠化是以岩溶山区脆弱的岩溶生态环境为基础，以内、外地质作用和人类恶性生态地质作用为驱动力，以资源退化和恶化为本质，基岩不断裸露，生态环境不断恶化，周而复始的恶性循环过程．提出对石漠化的研究重点是成固机理研究，对石漠化的防治应着重于对石漠化的驱动力及其主异因子的防治，阐述了治山必须先治水等防治对策与建议．     

人工神经网络在遥感数字图像分类处理中的应用

近年来,随着人工神经网络系统理论的发展,神经网络技术日益成为遥感数字图像分类处理的有效手段,并有逐步取代最大似然法的趋势。本文重点讨论国外遥感图像分类处理研究中应用效果显着或应用前景广阔的一些神经网络方法。最后讨论了有关上述方法在遥感数据分类处理中实际应用的一些原则性问题。     

伸臂式结构地震剪力图和弯矩图的理论包络线

本文以人工工振型作为理论工具、推演出伸臂式结构地震剪力地震弯矩的主、次上界的简练公式。不难理解,上界地震剪力图和弯矩图就分别相当于无数地震剪力图的弯矩图的包络线。文中还推演出主、次上界与等效静力解之间的一个十分简明的理论关系。