人工神经网络在白云岩类型识别中的应用

利用地球化学数据,运用人工神经网络方法对美国密苏里州东南Ｂｏｎｎｅｔｅｒｒｅ组（寒武纪）滨海相的白云岩进行了分类、识别,判别率达１００％,结果表明,该方法性能良好,可望成为岩石分类、判别的有效手段。     

On the tidal variation of the geopotential

In this paper analytical expressions are derived for the temporal variations ofJ ₂ andJ ₂₂ due to the tides of the solid Earth, taking into account only the deformation of the mantle, and employing a procedure already used by the authors in their Hamiltonian theory of the Earth's rotation, which obtain the necessary parameters in a direct way by integration of those provided by a selected model of Earth interior.Numerical tables giving the periodic variation of coefficients are given, as well as a new prediction for UT1. For J ₂ and J ₂₂ the amplitudes reach such a magnitude that both two variations should not be ignored in studies involving the analysis of highly precise satellite tracking data. Moreover, the possibility of improving our knowledge of the value of those harmonic coefficients in only a more exact digit appears as to be strongly dependent on the limitations in the theoretical modeling of the variations of the inertia tensor due to solid tides.     

南极长城站地区现代环境地球化学特征

本文通过环境背景值、风化壳地球化学、对流层(大气气溶胶)地球化学和人为地球化学异常,初步探讨了南极长城站地区的现代环境地球化学特征。分析表明:环境要素固有的地球化学性质、区域环境条件和自然环境演变之间具有深刻的内在联系。     

西安南郊丰水年秋季土壤水分研究

 通过对西安南郊丰水年秋季土壤含水量的测定，研究了地下0~6 m之间土壤含水量的变化与土壤干层的恢复问题。研究结果表明，丰水年西安南郊人工林下2~3 m的土壤含水量在20％左右，远远高于正常年份的土壤含水量，而且大于表层和深层的土壤含水量。分析得出，土壤干层发育较弱的地区在降水丰富的年份可以得到一定程度的恢复；西安地区基本适合进行人工造林；通过人工措施，增强降水入渗和土壤含水量，对树木成活和长期的生长应当具有明显的作用。     

Overcoming flood and desiccation made by implementing artificial recharge toward absorption reservoir

Flood and desiccation are perceived as two most critical and influential disasters which contradict between causes and consequences. Flood occurs when the water surface could no longer flow the whole water flow, thus, the water flooded. Contradicted to that, desiccation occurs when the water flow contains a low volume of water deposit, thus, the water requirement exceeds the available potential. That condition was caused by land utilization as the consequences due to the increased land requirement for housing or industrial needs. An attempt to overcome the flood, nowadays, is implemented mostly in a structural way, through building canals, implementing rivers normalization, building gateways or building flood control pump which are more directed toward the flood direction in order to increase the surface flow in an immediate maner to the sea. However, the effort to overcome the flood its self, could be more effective if followed by an effort to increase the soil ability to absorb natural recharge or artificial recharge or by refilling the water into the earth surface. Absorption reservoir used as one of technology alternatives （artificial recharge） could also be used to support the attempt to overcome flood and desiccation. Absorption reservoir is a dam which was designed according to the basic principles, such as the bottom surface of the reservoir that has a high permeability surface; the surface of the reservoir water is higher （higher aquifer） along with a high permeability, considering the availability of water source that has been absorbed and its quality; considering the aquifer category on the water absorption dam;     

人工神经网络在天津市区地面沉降预测中的应用

在分析天津市区地面沉降特点的基础上，结合人工神经网络原理，选择1961-1980年的天津市区降水量、地下水开采量、前年沉降量、固结度作为训练样本的输入量，以这20年的地面沉降量作为输出量，用贝叶斯正则化算法训练BP网络，得到沉降的仿真模型。并把1981-1993年的资料用来进行预测检验，结果表明这是一种比较理想的地面沉降预测方法。最后在不同的降水量保证率下，预测了到2010年天津市区地面沉降的情况。     

反扩散格式在二维锋生数值模拟中的应用

把无粘流中能成功捕捉接触间断的反扩散差分格式，用在了准地转二维理想锋生的数值模拟中，并将计算结果与Cullen的计算结果和半地转方程的精确解进行了比较。用反扩散格式模拟的锋面较Cullen的结果更加显，而且更接近半地转方程的精确解。     

吐鲁番山区人工增水作业效果的统计分析

应用统计检验和回归分析两种数理统计方法，对吐鲁番1997－2000年山区人工增水作业效果进行了分析。结果得出作业影响区内作业年份5－6月平均流量显著增加。     

ARTIFICIAL NEURAL NETWORK CORRECTION OF ROUGH ERRORS OF OBSERVATIONS

In the context of tower measured radiation datasets.following the correction principle meetinga diagnostic equation in data quality control and in terms of a technique for model construction ondata and ANN(artificial neural network)retrieval for BP correction of radiation measurementswith rough errors available,a BP model is presented.Evidence suggests that the developed modelworks well and is superior to a convenient multivariate linear regression model,indicating its wideapplications.     

Structural damage detection using the optimal weights of the approximating artificial neural networks

This work presents a novel neural network‐based approach to detect structural damage. The proposed approach comprises two steps. The first step, system identification, involves using neural system identification networks (NSINs) to identify the undamaged and damaged states of a structural system. The partial derivatives of the outputs with respect to the inputs of the NSIN, which identifies the system in a certain undamaged or damaged state, have a negligible variation with different system errors. This loosely defined unique property enables these partial derivatives to quantitatively indicate system damage from the model parameters. The second step, structural damage detection, involves using the neural damage detection network (NDDN) to detect the location and extent of the structural damage. The input to the NDDN is taken as the aforementioned partial derivatives of NSIN, and the output of the NDDN identifies the damage level for each member in the structure. Moreover, SDOF and MDOF examples are presented to demonstrate the feasibility of using the proposed method for damage detection of linear structures. Copyright © 2001 John Wiley & Sons, Ltd.