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.     

RECENT TECTONIC DEFORMATION ANOMALY AND EARTHQUAKES IN GANSU-NINGXIA-QINGHAI AREA

1　IntroductionAccordingtogeologicalstructure ,theGansu Ningxia Qinghaiareabelongstothenortheastmar ginofQinghai Tibetblock .Thisareahasbeenpaidmuchattentionby geo specialistsinChinaandabroadbecauseofitssignificanttectonicmovement,itsintensiveseismicity ,anditsimportanceinearth quakehazardmitigation .IntheDevelopmentPro gramonNationalKeyBasicResearchesundertheProject“MechanismandPredictionofContinentalStrongEarthquakes”,themechanismsofcontinen talstrongearthquakesarestudied ,usinghypo…     

FEASIBLE STUDY ON THE INTEGRATION SYSTEM FOR THE SPACE MONITORING OF MAJOR EARTHQUAKES AND VOLCANOES IN TERRESTRIAL LAND

With the rapid development of space technology, earth observation technology and sky observatory technolo-gy, they have played a more and more important part in monitoring and predicting of earthquakes and volcanoes in the terres-trial land. In recent years, the related agencies have done the experiments and researches on monitoring and predicting ofearthquakes and volcanoes in the forewarning period by means of many approaches, such as satellite thermal infrared re-mote sensing (TIRS), Global Positioning System (GPS), differential interferometric synthesis aperture radar (D-INSAR),astronomical time-latitude residual anomaly, and Geographic Information Systems (GIS), etc. A quite large number of re-search foundation has been built in the fundamental theories and application methods. The experiments and researcheshave shown that these technology is efficient methods for high frequency crust movement. If the existed separate scientificforces and results are possibly assembled together to form a more complete integration monitoring system with the combina-tion of space, sky observation, ground, deep geology and macro anomaly, it will come into a new stage of monitoring andpredicting of earthquakes and volcanic eruptions.     

大同—阳高Ms6.1地震前兆综合加权信息熵的变化

利用“地震预报计算机专家系统”的思想对大同 -阳高 Ms6 .1地震前每一前兆异常事件进行综合评估 ,以每一异常的最可能发震时间来计算发震概率 ,利用地震前兆综合加权信息熵研究了系统熵值与地震的关系。对华北地区的地震前兆综合加权信息熵研究表明 ,在大同 -阳高 Ms6 .1地震前 ,信息熵出现了明显的减熵有序变化     

地壳对海洋潮汐的响应

应用三维动态有限元方法研究了中国北部地区的地壳对邻近的渤海与黄海海平面变化的响应。虽然此应力场过于微弱不足以引发地震，但发现应力集中的位置及应力场变化较大的位置恰好与某些现代地震的震中一致。这一结果表明研究地壳对广泛分布的载荷的响应对研究区域地震构造是有帮助的。     

Application of micropolar plasticity to post failure analysis in geomechanics

A micropolar elastoplastic model for soils is formulated and a series of finite element analyses are employed to demonstrate the use of a micropolar continuum in overcoming the numerical difficulties encountered in application of finite element method in standard Cauchy–Boltzmann continuum. Three examples of failure analysis involving a deep excavation, shallow foundation, and a retaining wall are presented. In all these cases, it is observed that the length scale introduced in the polar continuum regularizes the incremental boundary value problem and allows the numerical simulation to be continued until a clear collapse mechanism is achieved. The issue of grain size effect is also discussed. Copyright © 2004 John Wiley & Sons, Ltd.     

An explicit representation of steady state response of a beam on an elastic foundation to moving harmonic line loads

A closed‐form deflection response of a beam rest is presented in this paper using the integral transform method. The theory of linear partial differential equations is used to represent the deflection of beam subjected to a moving harmonic line load in integration form. The solution is finally carried out using the inverse Fourier transform. To evaluate the integration analytically, poles of the integrand are identified with the help of algebraic equation theory. Residue theorem is then utilized to represent the integration as a contour integral in the complex plane. Closed‐form deflections and numerical results are provided for different combinations of load frequency and velocity. Copyright © 2002 John Wiley & Sons, Ltd.     

昆仑山8.1级地震前中国大陆的构造应变背景

利用“网络工程”1998～2001年累积的1181个测站的GPS重复观测资料，采用双三次样条函数模型建立中国大陆水平运动模型速度场，用大地坐标在椭球面上计算各类应变场，详细分析了2001年昆仑山8．1级地震前中国大陆水平构造应变场空间分布特征。各类构造应变场的最高值都出现在喜马拉雅构造带与昆仑山地块内(地震断裂带南侧)，鲜水河—安宁河断裂带次之。分析表明，昆仑山8．1级地震正好发生在张性面膨胀应变率的高值区，第一、第二和最大剪应变率高值区边缘的突变区和最大、最小主应变率的高值区。     

地震前后垂直形变场动态演化的量化指标

提出一种表达垂直形变场动态演化过程的量化指标——区域应变率、应变集中度。在此基础上，对南北地震带各水准监测区近30年的垂直形变资料进行了实际计算．并结合具体震例进行了对比研究。结果表明：量化指标在一定程度上能够反映地震前后形变场的动态演化过程和地壳运动状态，对地震的中长期预报有一定的积极意义。