灰色模型分析法在高层建筑物变形监测预测中的应用

在变形监测过程中,获取监测资料以后,通过对高层建筑物观测资料进行研究分析,可以掌握地基随时间的沉降规律,预测其下一步的变形趋势,并为可能发生的变形做好必要的准备,减少因非预计沉降带来的损失。灰色模型分析法是一种常用的数据处理及预测分析方法,本文在简单介绍了灰色模型分析法的基础上,利用灰色模型分析法预测了变形监测数据,并得出了有益结论。     

IGS精密星历和钟差对天顶对流层延迟精度的影响

为研究IGS精密轨道和钟差产品对天顶对流层延迟精度的影响,文章利用位于中国北京、上海、拉萨等地的6个IGS跟踪站所提供的2013年4月7日～10日4天的数据,采用GPSTools软件进行实验,计算各跟踪站的天顶对流层延迟(ZTD),并与IGS提供的对流层延迟产品进行对比.结果表明,利用IGS精密轨道解算的ZTD与IGS提供的ZTD相当,两者偏差的平均RMS优于5mm,利用IGS超快速钟差预报部分解算的ZTD与IGS提供的ZTD存在2cm～3cm误差,平均RMS大于1cm.     

利用MorPAS对上黑龙江成矿带进行金成矿预测

上黑龙江成矿带主要由早-中侏罗世河流-湖泊-洪积相含煤碎屑岩组成,目前已发现多个中小型矿床,在该区进行成矿预测意义重大。文章利用MORPAS成矿预测系统,采用证据权重法进行了成矿预测工作,在充分研究区内已知矿床的成矿规律基础上,以1：257Y地质图为背景,物化探异常分析子系统对1：20万区域化探数据进行C—A分形确定异常下限,综合现有的找矿标志,提取有利的成矿要素,进行针对处理,制成上黑龙江成矿带MORPAS金成矿预测分析图,提供成矿靶区,为森林高覆盖区提供找矿有利单元。     

神经网络辅助的GPS/MEMS-INS组合导航算法

在GPS/MEMS-INS(Micro-electromechanical System-Inertial Navigation System)组合导航中,当载体处于恶劣环境下或者载体处于大机动运动情况下,会导致GPS失锁。此时单独工作的低精度MEMS-INS会由于其位置和速度误差随着时间的变化而迅速积累最终无法导航。针对此问题,设计了一种结构简单易于实现的神经网络辅助的GPS/MEMS-INS组合导航系统。通过模拟实验,与标准卡尔曼滤波(KF)和无迹卡尔曼滤波(UKF)框架下组合导航相比,所提出的结构简单容易实现的神经网络辅助的组合导航系统具有较高的稳定性,并且组合导航整体过程中经纬度与速度精度均提高了65%左右。     

基于LSTM神经网络的青藏高原月降水量预测

青藏高原的降水量预测不仅为该地区水资源合理规划利用提供依据,同时对中国及周边国家气候变化研究有着重要的意义。论文利用1990—2016年青藏高原降水量数据,采用长短期记忆神经网络（LSTM）对青藏高原月降水量进行预测,主要包括：① 使用青藏高原86个测站1990—2013年的月降水资料,预测各个测站2014—2016年的月降水量,并与传统的RNN、NAR、SSA和ARIMA预测模型相比,平均决定系数R²分别提高了0.07、0.15、0.13和0.36,均方根误差（RMSE）和平均绝对误差（MAE）表现更低;② 分析了降水量预测精度的空间分布特征,将各模型的R²在青藏高原地区内插值,分析R²的空间分布特征,发现所有模型降雨稀少的干旱地区和降雨多的湿润地区R²较低,在气候稳定、降水规律性明显的地区R²较高,且LSTM模型R²≥0.6的空间范围远大于传统模型;③ 分析了不同预测长度对各模型预测精度的影响,发现所有模型会随着预测长度增加而预测精度降低,但在不同的预测长度下LSTM预测的RMSE值都低于其他模型。     

人工神经网络和决策树模型在滑坡易发性分析中的性能对比

机器学习模型广泛应用于区域性滑坡易发性分析。模型的选择关系到评价结果的可信度、准确率和稳定性。现有滑坡易发性分析模型对比研究侧重模型的预测精度。模型的稳定性和数据量敏感性对机器学习模型的性能评估同样非常重要。本文以福建省南平市蔡源流域为研究区,以四川省绵阳市北川县为验证区,从预测精度、稳定性和数据量敏感性3个方面深入对比BP（Back Propagation）人工神经网络模型和CART（Classification and Regression Tree）决策树模型在滑坡易发性分析中的效果,主要结论如下:① 在逐渐增加一定数量训练样本的过程中,BP人工神经网络模型预测精度的增长率更高。在蔡源流域内,当训练样本数量增加10 000时,BP人工神经网络模型的预测精度上升5.22%,CART决策树模型的预测精度上升2.11%。② BP人工神经网络的预测精度高于CART决策树模型,且较为稳定。在100组数据集上,BP人工神经网络模型验证集预测精度的均值和验证集滑坡样本预测精度的均值分别为81.60%和84.86%,高于CART决策树模型的72.97%和76.59%。与此同时,BP人工神经网络模型对应预测精度的标准差分别是0.32%和0.37%,小于CART决策树模型的0.35%和0.67%。③ BP人工神经网络模型分析的滑坡易发区相比CART决策树模型,更接近实际滑坡的空间分布。最后,北川县的验证实验也出现了相同的现象。     

�й���½GPS����Ĺ��켰����ѧ��������Ԫ����

????????й?????????????????????????????????????????????????????????????????????????μ??й?????α? ?????GPS???????????????????????????????:??????????????????10???????й????????????????????????10??GPS?????????????????????;?????????????????????:??????????????????????????,??????????????????????????????????????,??????????????????????黡???????????????????????????,?????????????????????????????????????????????:1)?й?????????????????????????????????????????????????????????????????????????????;2)?Щ????????????????????????????????????й??????????????????     

与地震有关的水文及地球化学变化

简要回顾了几十年来对地震发生前、地震过程中和震后地下流体和地球化学变化的研究和成果，这些研究一般都是以探索地震预报可能性为目的的。论述了与地震有关的地下水文及地球化学变化的机理，这些地下流体(包括地下水和气体诸如氢、氧和惰性气体)的起源和迁移流动现象以及详细介绍了早期和近代对有关地震的地下流体和地球化学变化的观测成果。同时指出了对地下流体和地球化学作为地震前兆来观测研究的困难所在以及为了克服这些困难而应该采取的地震前兆观测研究的方向，例如多种手段和多种原理方法，开发有效的地球物理和地球化学模型以及适当的数据分析统计方法等。     

Study of Distinctive Regional Features of Surface Solar Radiation in North and East China

Solar radiation is an important energy source for plants on the earth and also a major component of the global energy balance.Variations in solar radiation incident at the earth's surface profoundly affect the human and terrestrial environment,including the climate change.To provide useful information for predicting the future climate change in China,distinctive regional features in spatial and temporal variations of the surface solar radiation (SSR) and corresponding attributions (such as cloud and aerosol) are analyzed based on SSR observations and other meteorological measurements in North and East China from 1961 to 2007.Multiple models,such as the plane-parallel radiative transfer model,empirical and statistical models,and corrclation and regrcssion analysis methods are used in the study.The results are given as follows.(1) During 1961-2007,the total SSR in North China went through a process from quickly “dimming” to slowly “dimming”,while in East China,a significant transition from “dimming” to “brightening” occurred.Although there are some differences between thc two regional variation trends,long-term variations in SSR in the two regions are basically consistent with the observation worldwide.(2) Between the 1960s and 1980s,in both North and East China,aerosols played a critical rolc in the radiation dimming.However,after 1989,different variation trends of SSR occurred in North and East China,indicating that aerosols were not the dominant factor.(3) Cloud cover contributed less to the variation of SSR in North China,but was thc major attribution in East China and played a promoting role in the reversal of SSR from dimming to brightening,especially in the “remarkable brightening” period,with its contribution as high as 70％.     

Ensemble rainfall forecasting with numerical weather prediction and radar‐based nowcasting models

The overall objective of this study is to improve the forecasting accuracy of the precipitation in the Singapore region by means of both rainfall forecasting and nowcasting. Numerical Weather Predication (NWP) and radar‐based rainfall nowcasting are two important sources for quantitative precipitation forecast. In this paper, an attempt to combine rainfall prediction from a high‐resolution mesoscale weather model and a radar‐based rainfall model was performed. Two rainfall forecasting methods were selected and examined: (i) the weather research and forecasting model (WRF); and (ii) a translation model (TM). The WRF model, at a high spatial resolution, was run over the domain of interest using the Global Forecast System data as initializing fields. Some heavy rainfall events were selected from data record and used to test the forecast capability of WRF and TM. Results obtained from TM and WRF were then combined together to form an ensemble rainfall forecasting model, by assigning weights of 0.7 and 0.3 weights to TM and WRF, respectively. This paper presented results from WRF and TM, and the resulting ensemble rainfall forecasting; comparisons with station data were conducted as well. It was shown that results from WRF are very useful as advisory of anticipated heavy rainfall events, whereas those from TM, which used information of rain cells already appearing on the radar screen, were more accurate for rainfall nowcasting as expected. The ensemble rainfall forecasting compares reasonably well with the station observation data. Copyright © 2012 John Wiley & Sons, Ltd.