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.     

Generalized strain probing of constitutive models

Advanced material constitutive models are used to describe complex soil behaviour. These models are often used in the solution of boundary value problems under general loading conditions. Users and developers of constitutive models need to methodically investigate the represented soil response under a wide range of loading conditions. This paper presents a systematic procedure for probing constitutive models. A general incremental strain probe, 6D hyperspherical strain probe (HSP), is introduced to examine rate‐independent model response under all possible strain loading conditions. Two special cases of HSP, the true triaxial strain probe (TTSP) and the plane‐strain strain probe (PSSP), are used to generate 3‐D objects that represent model stress response to probing. The TTSP, PSSP and general HSP procedures are demonstrated using elasto‐plastic models. The objects resulting from the probing procedure readily highlight important model characteristics including anisotropy, yielding, hardening, softening and failure. The PSSP procedure is applied to a Neural Network (NN) based constitutive model. It shows that this probing is especially useful in understanding NN constitutive models, which do not contain explicit functions for yield surface, hardening, or anisotropy. Copyright © 2004 John Wiley & Sons, Ltd.     

NORTHERN AND SOUTHERN HEMISPHERIC POLAR LOWS——A COMPARATIVE STUDY

Northern Hemispheric“polar lows”,i.e.maritime,small,but fairly intense cyclonic systems have been studied formore than two decades.Recently researchers have turned their attention towards the occurrence of polar lows and othersmall-scale cyclonic disturbances on the Southern Hemisphere.In this paper a short account of our present knowledgeof Northern Hemispheric polar lows will be given,followed by some preliminary results of an investigation of the natureand occurrence of polar lows on the Southern Hemisphere.     

K-H Instability of Dust Plasma

Areview about K-H instability related to comet tail is presented.The effect of dust graing on KHI is theoretically analysed and relations of critical shear to excit KHI with the properties of dust grains both for sheared ion flow and for sheared dust flow re put out.Their possible application in comet tail is suggested.     

PROPAGATION OF 30—60 DAY LOW FREQUENCY OSCILLATIONS AND THEIR INFLUENCE UPON THE SUBTROPICAL WESTERLIES JET STREAM DURING NORTHERN HEMISPHERE WINTER

Based on daily ECMWF gridpoint data of two winters during 1981—1983 including an ENSOyear,propagation of low frequency oscillations(LFO)during Northern Hemisphere winters andtheir influences upon 30—60 day oscillations of the subtropical jet stream are studied with the sta-tistical methods as complex empirical orthogonal function(CEOF)and so on.Results show that inthe winter of a normal year(1981—1982),30—60 day oscillations in the subtropical zone aremainly in the northern and southern flanks of exit region of jet stream.In the ENSO year(1982—1983),they are mainly in the vicinity of entrance and exit regions of jet stream.Intraseasonalchanges of subtropical jet stream manifested themselves as latitudinal fluctuation or longitudinalprogression or regression of about 40 day period.There are marked differences between propagat-ing passages of low frequency modes responsible for changes of subtropical jet stream in the normalyear(1981—1982)and in the ENSO year(1982—1983).Changes of oscillation amplitude showobvious phases.In general,the one in late winter is stronger than that in early winter,strongestone occurs in February.     

Energy flow in Branchiura sowerbyi (Oligochaeta: tubificidae) in a shallow macrophyte-dominated lake, Biandantang Lake

The energy flow ofBranchiura sowerbyi was studied for the first time in China in a shallow macrophytic lake, Biandantang Lake, Hubei Province. The energy flow was calculated from the measurement of flesh production (12.5241kJ/m²a), egestion (517.7302kJ/m²a), metabolism (38.3273 kJ/m²a), and excretion (4.3798kJ/m²a). The net growth efficiency of the species is about 22.7%, which accords well with the generally reported value for oligochaetes. In addition, the relationship between starvation respiration (R, mgO₂/ind·d), wet weight (Ww, mg) and temperature (T, °C) were also measured, with the regression function beingR=0.008Ww^0.736 e^0.050T. Project supported by NSFC (30270278, 3960019), the foundation of the government of Hubei Province (No. 2000J109), and the foundation of Ecological Station, CAS in the Institute of Hydrobiology.     

ENERGY FLOW OF BELIAMYA AERUGINOSA IN A SHALLOW ALGAL LAKE, HOUHU LAKE （WUHAN, CHINA）

The energy budget ofBellamya earuginosa in a shallow algal lake, Houhu Lake (Wuhan, China) was investigated by the measurement of flesh production (32.8 kJ/(m²·a)), egestion (337.7 kJ/(m²·a)), metabolism (246.7 kJ/(m²·a)), and estimation of excretion (21.4kJ/(m²·a)). The net growth efficiency of the species is about 10.9%, which accords with the generally reported value for gastropods. In addition, the relationships between starvation respiration (R, mgO₂/(Ind·d)), body weight (Wd, mg in dry wt) and temperature (T, °C) were also determined. The regression equationR=0.044Wd ^0.537 e ^0.061T was obtained by the least square method, The measured SDA of the species is 26.51% of its gross metabolism. Project No. 3960019 and 39430101 supported by NSFC and also a granted for the East Lake Ecological Experimental Station, CAS, Wuhan.     

On the microwave millisecond spike emission and its associated phenomena during the impulsive phase of large solar flare

We propose a model of two acceleration regions, which can explain, on the basis of microwave maser caused by a “hollow-beam” distribution of electrons, the presence of millisecond spikes in the event of 1981 May 16 and their absence in the event of 1981 October 12, and the enhanced continuous emission in the latter. We have also uncovered relations among the features, e.g. the Type III_G, Type IV_DCIM and hard x-ray bursts, that accompany the microwave millisecond spikes during the impulsive phase of a large flare.     

Study on numerical simulation to water lowering in an open coal mine digging

In view of the situation of excavation of open coal mine for the underground water disaster,we should carry out simulation studies for the numerical value of the water lowering project and improve the accuracy and the level of the water lowering project.On the basis of the hydrological geological conditions of certain open mine digging,a more reasonable seepage numerical model was built according to MODFLOW.It was simulated in advance that the process of the confined water level descending with the time,and combining with the actual observations to test the correctness of the model.The calculation showed that the results coincided well with the results of actual measurement.Based on this,different water lowering numerical simulations were built for the open coal mine digging.It could be simulated and forecast that the changes of the groundwater level in drainage process within and outside the mine pit,and it was quantitatively assessed that the possible water lowering result of the opencast water drainage process,which provide an important basis for the actual water lowering project and the possible project disposal.