A machine learning approach for predicting computational intensity and domain decomposition in parallel geoprocessing

ABSTRACT

High performance computing is required for fast geoprocessing of geospatial big data. Using spatial domains to represent computational intensity (CIT) and domain decomposition for parallelism are prominent strategies when designing parallel geoprocessing applications. Traditional domain decomposition is limited in evaluating the computational intensity, which often results in load imbalance and poor parallel performance. From the data science perspective, machine learning from Artificial Intelligence (AI) shows promise for better CIT evaluation. This paper proposes a machine learning approach for predicting computational intensity, followed by an optimized domain decomposition, which divides the spatial domain into balanced subdivisions based on the predicted CIT to achieve better parallel performance. The approach provides a reference framework on how various machine learning methods including feature selection and model training can be used in predicting computational intensity and optimizing parallel geoprocessing against different cases. Some comparative experiments between the approach and traditional methods were performed using the two cases, DEM generation from point clouds and spatial intersection on vector data. The results not only demonstrate the advantage of the approach, but also provide hints on how traditional GIS computation can be improved by the AI machine learning.     

Modelling of deformation response and chemo‐mechanical coupling in chalk

A constitutive relation is derived for describing the mechanical response of chalk. The approach is based on a phenomenological framework which employs chemo‐plasticity. The properties of the material are assumed to be affected by the physico‐chemical processes that occur through the interaction between the skeleton and the pore fluid. The underlying mechanism is discussed by invoking a micromechanical analysis. The performance of the framework is illustrated by examining the evolution of mechanical characteristics in the presence of different pore fluids. Copyright © 2006 John Wiley & Sons, Ltd.     

成团天体系统成员概率计算的有效性问题

本文提出了一个普适的衡量成团天体系统成员概率计算结果的有效性指数：Ｅ＝１－ＮΣ｛Ｐ（ｉ）［１－Ｐ（ｉ）］｝／ΣＰ（ｉ）Σ［１－Ｐ（ｉ）］｝，并以此对以Ｓａｎｄｅｒｓ方法为基础的各种成员确定模型的有效性进行了分析．从总体上说，运动学资料（自行、视向速度等）是一类有效的观测判据．     

Application of MODFLOW and geographic information system to groundwater flow simulation in North China Plain, China

MODFLOW is a groundwater modeling program. It can be compiled and remedied according to the practical applications. Because of its structure and fixed data format, MODFLOW can be integrated with Geographic Information Systems (GIS) technology for water resource management. The North China Plain (NCP), which is the politic, economic and cultural center of China, is facing with water resources shortage and water pollution. Groundwater is the main water resource for industrial, agricultural and domestic usage. It is necessary to evaluate the groundwater resources of the NCP as an entire aquifer system. With the development of computer and internet information technology it is also necessary to integrate the groundwater model with the GIS technology. Because the geological and hydrogeological data in the NCP was mainly in MAPGIS format, the powerful function of GIS of disposing of and analyzing spatial data and computer languages such as Visual C and Visual Basic were used to define the relationship between the original data and model data. After analyzing the geological and hydrogeological conditions of the NCP, the groundwater flow numerical simulation modeling was constructed with MODFLOW. On the basis of GIS, a dynamic evaluation system for groundwater resources under the internet circumstance was completed. During the process of constructing the groundwater model, a water budget was analyzed, which showed a negative budget in the NCP. The simulation period was from 1 January 2002 to 31 December 2003. During this period, the total recharge of the groundwater system was 49,374 × 10⁶ m³ and the total discharge was 56,530 × 10⁶ m³ the budget deficit was −7,156 × 10⁶ m³. In this integrated system, the original data including graphs and attribution data could be stored in the database. When the process of evaluating and predicting groundwater flow was started, these data were transformed into files that the core program of MODFLOW could read. The calculated water level and drawdown could be displayed and reviewed online.     

大凌河流域河道生态需水量估算

本文从大凌河流域存在的环境问题出发,分别估算了各水资源分区的河道生态需水量（包括生态基流量、自净需水和输沙需水）及各区的河道生态需水贡献量。提出了使用偏枯年（75％保证率）水文数据估算河道生态需水量的方法．弥补了Tennant法、枯水季最小流量法等方法计算结果偏大或偏小的不足。结果表明,大凌河的4个河段（建凌站以上、建凌站至朝阳站、朝阳站至义县站、义县站至锦县段）的河道分区需水量分别为0．55×10^8m^3、5．21×10^8m^3、3．9×10^8m^3和1．34×10×10^8m^3。锦县水文站以上河道生态需水总量11×10^8m^3,占地表水总资源量的59．75％,因此．大凌河地表水资源取用率应在40％以下。     

中国东北依兰地区块状蓝片岩的地球化学特征

黑龙江省依兰地区分布的黑龙江杂岩被认为足具有缝合带性质的构造混杂岩,其中所含变基性火山熔岩中含有的变质矿物不具有定向组构特征,因此称之为块状蓝片岩.特征变质矿物--蓝闪石的大量出现表明其经历过高压低温变质作用.常量地球化学分析表明,其SiO2含量均在53%以下,TiO2含量超过2%,Na2O K2O的含量为2.77%～6.41%,Na2O的含量普遍大于K2O的含量,Na2O K2O-SiO2变异图显示其具有碱性玄武岩特征;痕量元素地球化学分析显示其轻重稀土元素分异明显,稀土元素配分曲线表现为轻稀土元素富集、重稀土元素亏损的右倾型.球粒陨石标准化图解中,不具有Eu的负异常;Nbx2-Zr/4-Y、Ti/100-Zr-Sr/2与Ba-Th-Nb-La图解的综合分析表明,块状蓝片岩具有碱性洋岛玄武岩(OIB)的特征,它们应是佳木斯地块与松嫩地块之间古洋壳俯冲过程的残留,这为探讨佳木斯陆块与松嫩陆块之间古老洋盆俯冲消减过程提供了重要依据.     

I(L)值完全诱导空间

作者引入 I(L)值完全下半连续映射 ,研究其性质。利用 I(L)值完全下半连续映射定义I(L)值完全诱导空间 ,给出 I(L)值完全诱导空间的拓扑基的表达形式 ,证得两个 I(L)值完全诱导空间的映射是连续映射的充分必要条件 ,并建立了乘积空间的 I(L)值完全诱导空间与 I(L)值完全诱导空间的乘积空间的联系     

聚类分析在海洋产业分析中的应用

聚类分析是多元统计分析中的一种方法,是将数学引入分类学形成的,比以前仅凭经验及专业知识进行分类更为科学、适用.用聚类分析方法分析研究沿海地区海洋产业状况是较高层次统计知识的运用,也是对数据挖掘技术的一种探索性研究.本文仅从产值的角度考虑地区海洋经济实力的状况,是利用聚类分析方法、借助海洋经济统计数据对地区海洋经济的实证研究.     

虾蟹类亲体生殖营养需求研究的进展

本文综述了虾蟹类生殖营养的最新研究成果,并讨论了各营养素对虾蟹类亲体成熟、生殖和幼体质量的影响．此外,就今后在虾蟹类亲体生殖营养方面开展深入研究提出了一些建议。     

葫芦岛地区道路网络的建立

本文针对城市交通道路网的特点,根据现实的交通网络,对葫芦岛地区道路网络进行建立及分析。利用ArcGIS中的Geodatabase构建路网数据库,利用节点一弧段数据结构构建空间网络拓扑关系,生成点一弧矢量数据模型表示的道路网络。并实现四种网络分析功能:最佳路径分析、空间查询、追踪分析和缓冲区分析。