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. 相似文献
Methane content in coal seam is an essential parameter for the assessment of coalbed gas reserves and is a threat to underground coal mining activities. Compared with the adsorption-isotherm-based indirect method, the direct method by sampling methane-bearing coal seams is apparently more accurate for predicting coalbed methane content. However, the traditional sampling method by using an opened sample tube or collecting drill cuttings with air drilling operation would lead to serious loss of coalbed methane in the sampling process. The pressurized sampling method by employing mechanical-valve-based pressure corer is expected to reduce the loss of coalbed methane, whereas it usually results in failure due to the wear of the mechanical valve. Sampling of methane-bearing coal seams by freezing was proposed in this study, and the coalbed gas desorption characteristics under freezing temperature were studied to verify the feasibility of this method. Results show that low temperature does not only improve the adsorption velocity of the coalbed gas, but also extend the adsorption process and increase the total adsorbed gas. The total adsorbed methane gas increased linearly with decreasing temperature, which was considered to be attributed to the decreased Gibbs free energy and molecular average free path of the coalbed gas molecular caused by low temperature. In contrast, the desorption velocity and total desorbed gas are significantly deceased under lower temperatures. The process of desorption can be divided into three phases. Desorption velocity decreases linearly at the first phase, and then, it shows a slow decreases at the second phase. Finally, the velocity of desorption levels off to a constant value at the third phase. The desorbed coalbed gas shows a parabolic relation to temperature at each phase, and it increases with increasing temperature at the first phase, and then, it poses a declining trend with increasing temperature at the rest phases. The experimental results show that decreasing the system temperature can restrain desorption of coalbed methane effectively, and it is proven to be a feasible way of sampling methane-bearing coal seams.
历史名人的行为轨迹反映了当时的历史文化背景,通过历史名人行为轨迹的空间化和可视化,可以对历史社会状态进行探索和分析。对历史名人的社交关系网络进行可视化研究,有利于剖析当时的政治背景与人物关系。目前,基于GIS的空间人文社会科学深层次分析方法和工具还很少,根据地理位置对历史人物的社交网络进行分时段的研究也很少。本文以玄奘和欧阳修为例,探索了基于WebGIS的历史人物轨迹空间可视化分析方法,基于核密度估计与标准差椭圆的空间分析方法,分析历史名人轨迹点的空间分布特征,统计迁徙指数、首都距、家乡距、成长地距以分析基于距离的轨迹点移动特点;分时段构建了历史名人的空间社交网络,并结合历史背景、名人事迹、名人作品和空间化结果进行了综合分析。分析结果表明： ① 历史名人的迁移轨迹与当时的历史人口迁移趋势基本是一致的,受社会变动影响较大;② 历史名人在事业上升期有更大的社交网络圈,而在人生没落阶段社交网络圈减小。本文对历史名人轨迹的空间可视化与分析方法进行了探索,可以为空间人文社会科学相关领域的分析研究提供参考。 相似文献
