风暴分类识别技术在人工防雹中的应用

利用新一代多普勒天气雷达资料,在风暴跟踪识别算法的基础上,发展了风暴分类技术,以提高人工防雹作业指挥的效率。首先以SCIT算法为基础,结合风暴的结构特征,综合利用雷达、探空资料,自动提取风暴结构特征指数;其次采用基于决策树模型的风暴自动分类技术,将风暴按强度分为雷雨云、单体风暴、多单体风暴和强风暴;最后根据风暴强度、高度和位置等属性,对有可能产生冰雹的单体,结合GIS,自动对下游方向或附近作业点进行预警或输出作业参数。通过对2006—2014年期间重庆、辽宁大连和河南三门峡三地发生的较为典型的31次冰雹天气过程、182站次冰雹样本的检验来看:该方法通过对风暴按强度、垂直结构等综合属性进行分类,能有效提高冰雹识别的命中率、降低空报率,其中强风暴的命中率能达到100%,空报率仅为11.4%。能有效提高人工防雹作业的自动化程度,对防雹作业的科学决策有着重要参考作用。     

历史数据和强化学习相结合的低频轨迹数据匹配算法

针对低频(采样间隔大于1min)轨迹数据匹配算法精度不高的问题,提出了一种基于强化学习和历史轨迹的匹配算法HMDP-Q,首先通过增量匹配算法提取历史路径作为历史参考经验库;根据历史参考经验库、最短路径和可达性筛选候选路径集;再将地图匹配过程建模成马尔科夫决策过程,利用轨迹点偏离道路距离和历史轨迹构建回报函数;然后借助强化学习算法求解马尔科夫决策过程的最大回报值,即轨迹与道路的最优匹配结果;最后应用某市浮动车轨迹数据进行试验。结果表明:本文算法能有效提高轨迹数据与道路匹配精度;本算法在1min低频采样间隔下轨迹匹配准确率达到了89.2%;采样频率为16min时,该算法匹配精度也能达到61.4%;与IVVM算法相比,HMDP-Q算法匹配精度和求解效率均优于IVVM算法,16min采样频率时本文算法轨迹匹配精度提高了26%。     

An open source GIS‐based Planning Support System: Application to the land use plan of La Troncal,Ecuador

Planning Support Systems (PSS) comprise a wide variety of geo‐technological tools related to GIS and spatial modeling aimed at addressing land planning processes. This article describes the OpenRules system, a PSS based on a previous system called RULES. Among OpenRules new features are its architecture, based exclusively on free and open source software, and its applicability to all land use types, including rural and urban uses. In addition, OpenRules incorporates an unlimited number of land evaluation factors and a new objective in land use spatial allocation. OpenRules has been programmed in Java and implemented as a module of the free GIS software gvSIG, with full integration between the GIS and the decision support tools. Decision support tools include multicriteria evaluation, multiobjective linear programming and heuristic techniques, which support three basic stages of land use planning processes, namely land suitability evaluation, land use area optimization and land use spatial allocation. The application of OpenRules to the region of La Troncal, Ecuador, demonstrates its capability to generate alternative and coherent solutions through a scientific and justified procedure at low cost in terms of time and resources.     

Task-oriented information value measurement based on space-time prisms

Recent years have witnessed a large increase in the amount of information available from the Web and many other sources. Such an information deluge presents a challenge for individuals who have to identify useful information items to complete particular tasks in hand. Information value theory (IVT) from economics and artificial intelligence has provided some guidance on this issue. However, existing IVT studies often focus on monetary values, while ignoring the spatiotemporal properties which can play important roles in everyday tasks. In this paper, we propose a theoretical framework for task-oriented information value measurement. This framework integrates IVT with the space-time prism from time geography and measures the value of information based on its impact on an individual’s space-time prisms and its capability of improving task planning. We develop and formalize this framework by extending the utility function from space-time accessibility studies and elaborate it using a simplified example from time geography. We conduct a simulation on a real-world transportation network using the proposed framework. Our research could be applied to improving information display on small-screen mobile devices (e.g., smartwatches) by assigning priorities to different information items.     

RS & GIS-based landslide susceptibility mapping of the Balason River basin,Darjeeling Himalaya,using logistic regression (LR) model

The present study deals with the preparation of a landslide susceptibility map of the Balason River basin, Darjeeling Himalaya, using a logistic regression model based on Geographic Information System and Remote Sensing. The landslide inventory map was prepared with a total of 295 landslide locations extracted from various satellite images and intensive field survey. Topographical maps, satellite images, geological, geomorphological, soil, rainfall and seismic data were collected, processed and constructed into a spatial database in a GIS environment. The chosen landslide-conditioning factors were altitude, slope aspect, slope angle, slope curvature, geology, geomorphology, soil, land use/land cover, normalised differential vegetation index, drainage density, lineament number density, distance from lineament, distance to drainage, stream power index, topographic wetted index, rainfall and peak ground acceleration. The produced landslide susceptibility map satisfied the decision rules and ?2 Log likelihood, Cox &; Snell R-Square and Nagelkerke R-Square values proved that all the independent variables were statistically significant. The receiver operating characteristic curve showed that the prediction accuracy of the landslide probability map was 96.10%. The proposed LR method can be used in other hazard/disaster studies and decision-making.     

基于梯度提升算法的岩性识别方法

传统的岩性识别方法如岩屑录井、钻井取心及测井资料解释等技术,对录井质量的依赖程度较高,识别精度与效率低,泛化能力差。随着计算机技术的迅速发展,将测井资料与计算机技术相结合开展岩性研究已成为岩性识别的有效手段。本文提出了一种基于梯度提升算法XGBoost和LightGBM的岩性识别方法。以苏里格气田苏东41-33区块下碳酸盐岩储层为例进行测试验证,采用该方法结合测井资料中的声波时差、自然伽马、光电吸收截面指数、密度、深侧向电阻率和补偿中子等6种参数进行岩性识别,并与KNN （K近邻分类器）、朴素贝叶斯和支持向量机等传统算法进行对比,结果表明,3种传统算法的岩性识别准确率分别为78.45%、74.43%和78.72%,基于梯度提升算法XGBoost和LightGBM的识别准确率分别达到了98.90%和98.72%,远高于传统算法。     

一种基于机器学习算法的岩性填图方法

通过野外地质调查与机器学习方法的有机融合,提出了一种基于梯度提升决策树算法的岩性单元填图方法。研究以多龙矿集区为模型试验区,选择1∶5万勘查地球化学数据为基础预测数据,以1∶5万区域地质图为参考,进行基于梯度提升决策树算法的岩性预测填图模型试验。首先选择研究区内小范围空白区开展野外填图,建立原始数据集并初步构建岩性单元与预测数据对应关系;其次利用机器学习方法对预测数据进行多分类任务,进而开展目标填图区预测填图工作;最后通过概率选区选定概率较低目标区,开展进一步的小范围野外地质调查填图,对原始数据和知识库进行补充,迭代循环以上流程,直至预测填图达到要求。试验显示,随着迭代次数的增加,模型精度不断提高,并在7次迭代后模型准确率达到87%。该方法强调在实际应用中野外地质调查与基于机器学习预测填图的深度融合,以及野外实地工作在整个流程中的重要性和不可或缺性;同时能够充分挖掘已有数据资料的有用信息,用于辅助修正已有岩性填图内容,或根据已勘探区资料对邻近的未勘探区进行岩性分类,有效减少野外填图工作量,是对岩性填图方法、地质单元定量预测识别的有益探索,为区域地质填图工作提供了新的参考思路和辅助手段。      

Landslide susceptibility zonation method based on C5.0 decision tree and K-means cluster algorithms to improve the efficiency of risk management

Machine learning algorithms are an important measure with which to perform landslide susceptibility assessments,but most studies use GIS-based classification methods to conduct susceptibility zonation.This study presents a machine learning approach based on the C5.0 decision tree(DT)model and the K-means cluster algorithm to produce a regional landslide susceptibility map.Yanchang County,a typical landslide-prone area located in northwestern China,was taken as the area of interest to introduce the proposed application procedure.A landslide inventory containing 82 landslides was prepared and subse-quently randomly partitioned into two subsets:training data(70％landslide pixels)and validation data(30％landslide pixels).Fourteen landslide influencing factors were considered in the input dataset and were used to calculate the landslide occurrence probability based on the C5.0 decision tree model.Susceptibility zonation was implemented according to the cut-off values calculated by the K-means clus-ter algorithm.The validation results of the model performance analysis showed that the AUC(area under the receiver operating characteristic(ROC)curve)of the proposed model was the highest,reaching 0.88,compared with traditional models(support vector machine(SVM)=0.85,Bayesian network(BN)=0.81,frequency ratio(FR)=0.75,weight of evidence(WOE)=0.76).The landslide frequency ratio and fre-quency density of the high susceptibility zones were 6.76/km2 and 0.88/km2,respectively,which were much higher than those of the low susceptibility zones.The top 20％interval of landslide occurrence probability contained 89％of the historical landslides but only accounted for 10.3％of the total area.Our results indicate that the distribution of high susceptibility zones was more focused without contain-ing more"stable"pixels.Therefore,the obtained susceptibility map is suitable for application to landslide risk management practices.     

我国中东部平原地区临界气温条件下降水相态判别分析

基于2001—2013年地面观测和探空资料,对地面气温位于0～2℃(以下称临界气温)我国降雪的时空分布及其与降雨的垂直热力特征进行了研究,引入了决策树判别方法对上述条件下雪和雨进行了判别分析,结果表明:临界气温下降雪出现频率总体高于降雨、雨夹雪出现频率,且在我国华北南部至江南北部的中东部地区分布较多,年均可达7.69～15.38站次;临界气温下,降水相态为雨或雪对应的平均温度廓线最大差异位于650 hPa附近,且地面气温较低时,平均温度差异更明显,平均湿度廓线差异则主要位于低层,且在地面气温较高时,平均湿度差异更明显;临界气温下,降水相态为雨时,地面上空存在暖层样本占比,较降水相态为雪时更高,且降雨时暖层主要位于中层,降雪时暖层则主要位于低层,降雨时其暖层强度显著大于降雪时暖层强度;在临界气温下雨雪判别分析中,地面气温能显著提升判别准确率,湿球温度能在一定程度上提升判别准确率,基于云顶温度、中层融化参数、低层湿球温度构建的决策树判别模型,判别准确率达到91.86%,能较好地解决临界气温下雨和雪的判别问题。     

气象因子对半冬性小麦灌浆速度的影响效应研究

选用黄淮海冬麦区4个半冬性小麦品种郯麦98、山农18、徐麦33、皖麦52为试验材料,通过分期播种试验,利用方差分析、相关分析、逐步回归和通径分析等方法,分析半冬性小麦籽粒灌浆速度变化趋势和气象因子对灌浆速度的影响。结果表明,正常播期冬小麦灌浆速度波动性最小、千粒重最大,迟播10 d冬小麦灌浆速度波动性最大、千粒重最小;华北区品种郯麦98灌浆速度表现最稳定、千粒重最高,而黄淮区品种皖麦52灌浆速度最大;半冬性小麦灌浆持续期为35～39 d;南北气候差异是影响各品种冬小麦灌浆速度不同的原因之一。半冬性小麦各播期灌浆速度的变化趋势一致,灌浆速度变化与相关显著气象因子的变化规律相符合;灌浆速度峰值期一般出现在开花后15～25 d,迟播冬小麦最大灌浆速度出现时间较对照处理提前,不利于提高粒重;气温条件对冬小麦灌浆速度影响显著,其中最高气温要素是影响不同播期品种灌浆速度的共有关键因子。通径分析表明,最高气温对灌浆速度的作用由自身的直接效应决定,而日照时数与最低气温对灌浆速度的作用与间接效应一致;最高气温平均值对灌浆速度的影响最重要,日照时数和最低气温平均值对灌浆速度的影响较弱;最高和最低气温平均值、日照时数均为灌浆速度的限制因子,其中最高气温平均值对灌浆速度变化的决策作用最大。