Rasterization Computing-Based Parallel Vector Polygon Overlay Analysis Algorithms Using OpenMP and MPI

A GPU-based line-by-line method for thermal radiation transfer of H2O, CO2, and H2O/CO2 mixture

The machine learning method, now widely used for predicting well performance from unconventional reservoirs in the industry, generally needs large data sets for model development and training. The large data sets, however, are not always available, especially for newly developed unconventional plays. The objective of this work is to develop an innovative machine learning method for predicting well performance in unconventional reservoirs with a relatively small data set. For a small training data set, the corresponding machine learning model can significantly suffer from so-called overfitting meaning that the model can match the training data but has poor predictivity. To overcome this, our new method averages predictions from multiple models that are developed with the same model input, but different initial guesses of model parameters that are unknowns in a machine learning algorithm and determined in the model training. The averaged results are used for the final model prediction. Unlike traditional ensemble learning methods, each prediction in the new method uses all the input data rather than its subset. We mathematically prove that the averaged prediction provides less model uncertainty and under certain conditions the optimum prediction. It is also demonstrated that the method practically minimizes the overfitting and gives relatively unique prediction. The usefulness of the method is further confirmed by its successful application to the data set collected from less than 100 wells in an unconventional reservoir. Sensitivity results with the trained machine learning model show that the model results are consistent with the domain knowledge regarding the production from the reservoir.

Unlocking massively parallel spectral proper orthogonal decompositions in the PySPOD package

The application of parallel computers to reservoir simulation is studied with the aid of two recently introduced Multiple Instruction, Multiple Data (MIMD) computers. The problems posed by reservoir simulators are shown to be particularly well suited for such parallel computers* Fundamental parallel programming techniques using Fortran are presented for the Hetrogeneous Element Processor (HEP) produced by Denelcor and the iPSC Hypercube produced by Intel. These techniques are then applied to two of the most time consuming tasks in reservoir simulation: forming the matrix coefficients and producing a sparse matrix solution. The problem of parallel formation of sparse matrix coefficients is addressed for the single and multiphase cases using both black oil and compositional fluid models. Parallel sparse matrix solution is addressed by considering D4 ordered Gaussian elimination, and the multigrid, conjugate gradient, and Successive Over-Relaxation (SOR) methods. Three parallel algorithms based on SOR are developed to illustrate the possible diversity of parallel algorithms. The red-black and multicolored SOR methods which are often used on vector machines are shown to be easily adaptable to parallel MIMD machine. Matrix partitioning and a new method of iteration pipelining are also presented as parallel SOR methods. The various SOR algorithms, in parallel form, are mapped onto the HEP to illustrate the speedups currently possible on an MIMD machine. Projections are made on the future importance of these types of machines to reservoir simulation.

On parallel computers, the way the data elements are mapped to the processors may have a large effect on the timing performance of a given algorithm. In our previous paper, we have examined a few mapping strategies for the ordered radix-2 DIF (decimation-in-frequency) Fast Fourier Transform. In particular, we have shown how reduction of communication can be achieved by combining the order and computational phases through the use of i-cycles. A parallel methods was also presented for computing the trigonometric factors which requires neither trigonometric function evaluation nor interprocessor communication. This paper first reviews some of the experimental results on the Connection Machine to demonstrate the importance of reducing communication in a parallel algorithm. The emphasis of this paper, however, is on analyzing the numerical stability of the proposed method for generating the trigonometric factors and showing how the error can be improved. 16 refs., 12 tabs.

A parallel genetic algorithm to discover patterns in genetic markers that indicate predisposition to multifactorial disease

Deep neural network (DNN) has become the leading technology to realize Artificial Intelligence (AI). As DNN models become larger and more complex, so do datasets. Being able to efficiently train DNNs in parallel has become a crucial need. Data Parallelism (DP) is the widest-used solution today to accelerate DNN training but could be inefficient when processing DNNs with large-size parameters. Hybrid Parallelism (HP), which applies different parallel strategies on different parts of DNNs, is more efficient but requires advanced configurations. Not all AI researchers are experts in parallel computing, thus automating the configuration of HP strategies is very desirable for all AI frameworks. We propose a parallel semantics analysis method, which can analyze the trade-offs among different kinds of parallelisms and systematically choose the HP strategies with good training time performance. We demonstrated experimentally 260% speedup when applying our method compared to using a conventional DP approach. With our proposal, AI researchers would be able to focus more on AI algorithm research without being disturbed by parallel analysis and engineering concerns.

Machine learning for Big Data analytics in plants.

Overlay analysis is a common task in geographic computing that is widely used in geographic information systems, computer graphics, and computer science. With the breakthroughs in Earth observation technologies, particularly the emergence of high-resolution satellite remote-sensing technology, geographic data have demonstrated explosive growth. The overlay analysis of massive and complex geographic data has become a computationally intensive task. Distributed parallel processing in a cloud environment provides an efficient solution to this problem. The cloud computing paradigm represented by Spark has become the standard for massive data processing in the industry and academia due to its large-scale and low-latency characteristics. The cloud computing paradigm has attracted further attention for the purpose of solving the overlay analysis of massive data. These studies mainly focus on how to implement parallel overlay analysis in a cloud computing paradigm but pay less attention to the impact of spatial data graphics complexity on parallel computing efficiency, especially the data skew caused by the difference in the graphic complexity. Geographic polygons often have complex graphical structures, such as many vertices, composite structures including holes and islands. When the Spark paradigm is used to solve the overlay analysis of massive geographic polygons, its calculation efficiency is closely related to factors such as data organization and algorithm design. Considering the influence of the shape complexity of polygons on the performance of overlay analysis, we design and implement a parallel processing algorithm based on the Spark paradigm in this paper. Based on the analysis of the shape complexity of polygons, the overlay analysis speed is improved via reasonable data partition, distributed spatial index, a minimum boundary rectangular filter and other optimization processes, and the high speed and parallel efficiency are maintained.

Improving classification accuracy using data augmentation on small data sets

The accompanying CD-ROM contains a compilation of geophysical, geologic, and tectonic data for the Bering Shelf, the Chukchi Sea, the Arctic margin, and adjacent landmasses. These data sets focus on Alaska, the Russian Far East, and the continental shelves that link these two landmasses. For compatibility with other available geographic information system (GIS) products, our GIS compilation extends from 120°E to 115°W, and from 40°N to 80°N. This area encompasses the region from the modern Paciµc plate boundary of the Japan, Kurile, and Aleutian subduction zones, the Queen Charlotte transform fault, and the Cascadia subduction zone (in the south) to the continent-ocean transition from the Eurasian and North American continents to the Arctic Ocean (in the north); and from the diffuse Eurasian–North American plate boundary, including the probable Okhotsk plate (in the west) to the Alaskan-Canadian Cordilleran fold belt (in the east). The CD-ROM comprises thematic layers of spatial data sets for topography, gravity µeld, magnetic µeld, earthquakes, volcanoes, geology, tectonostratigraphic terranes, and cultural reference features, and also includes metadata (data about the data) for all these data sets. The spatial data sets can be viewed, analyzed, and plotted with commercial GIS software (ArcView and ARC/Info) or through a freeware program (ArcExplorer) that is included on this CD-ROM. This GIS compilation provides data for studies of the Mesozoic and Cenozoic collisional and accretionary tectonics that assembled this continental crust and of the neotectonics of active and passive plate margins in this region, and for constructing and interpreting geophysical, geologic, and tectonic models of the region.

Throughout the course of human history, significant pressures have been in place to develop the coastline. Whether these pressures are a result of commercial and industrial needs or recreational preference, humans have always tended to settle at or near the waterfront. While obvious benefits can be reaped from this anthropogenic pattern, it also results in significant portions of the world's population subject to flood hazards. With the potential for these hazards to increase due to climate change, coastal zone managers and planners require tools to mitigate the risk to this population. One such tool is Geographic Information Systems (GIS). With GIS technology being at the forefront of Federal Emergency Management Agency's (FEMA) Map Modernization efforts, it has proven to be an effective tool in coastal hazard mitigation. The GIS technology is extensively used to evaluate land use and population change in the coastal areas which ultimately drives the coastal hazard mapping. The identification of risk requires analyzing large GIS data sets including topographic data (e.g. TIN {triangular irregular network} and GRID), orthophotography, land use data, and census population data. The GIS technology can effectively analyze these large data sets using the calculation algorithms built in to the GIS software such ArcGIS and present information that will guide decision makers. The 3D rendering capability of GIS software can be used to visualize various scenarios which aid in effective hazard identification and mitigation. The 3D GIS mapping and animation tools can also be applied to simulate various disaster scenarios that can ultimately aid planners in prevention of human and property loss during coastal disasters. In FEMA Region I, CDM, a FEMA study contractor and its partner OCC have been involved in numerous coastal hazard studies across New England. Together CDM and OCC have scoped total of six counties and have been involved in coastal engineering and mapping activities in more than 10 counties across New England. This paper will present methodologies to analyze various data sources and identify coastal flooding risks.

Since spatial datasets are subject to sampling errors, a smoothing interpolation method should be employed to remove noise during DEM construction. Although least squares support vector machines (LSSVM) have been widely accepted as a classifier, their effect on smoothing noisy data is almost unknown. In this article, the smoothness of LSSVM was explored, and its effect on smoothing noisy data in DEM construction was tested. In order to improve the ability to deal with large datasets, a local method of LSSVM has been developed, where only the neighboring sampling points around the one to be estimated are used for computation. A numerical test indicated that LSSVM is more accurate than the classical smoothing methods including TPS and kriging, and its error surfaces are more evenly distributed. The real‐world example of smoothing noise inherent in lidar‐derived DEMs also showed that LSSVM has a positive smoothing effect, which is approximately as accurate as TPS. In short, LSSVM with a high efficiency can be considered as an alternative smoothing method for smoothing noisy data in DEM construction.

IntroductionWith advances in artificial intelligence, the use of computer-aided detection and diagnosis in clinical imaging is gaining traction. Typically, very large datasets are required to train machine-learning models, potentially limiting use of this technology when only small datasets are available. This study investigated whether pretraining of fracture detection models on large, existing datasets could improve the performance of the model when locating and classifying wrist fractures in a small X-ray image dataset. This concept is termed “transfer learning”.MethodFirstly, three detection models, namely, the faster region-based convolutional neural network (faster R-CNN), you only look once version eight (YOLOv8), and RetinaNet, were pretrained using the large, freely available dataset, common objects in context (COCO) (330000 images). Secondly, these models were pretrained using an open-source wrist X-ray dataset called “Graz Paediatric Wrist Digital X-rays” (GRAZPEDWRI-DX) on a (1) fracture detection dataset (20327 images) and (2) fracture location and classification dataset (14390 images). An orthopaedic surgeon classified the small available dataset of 776 distal radius X-rays (Arbeidsgmeischaft für Osteosynthesefragen Foundation / Orthopaedic Trauma Association; AO/OTA), on which the models were tested.ResultDetection models without pre-training on the large datasets were the least precise when tested on the small distal radius dataset. The model with the best accuracy to detect and classify wrist fractures was the YOLOv8 model pretrained on the GRAZPEDWRI-DX fracture detection dataset (mean average precision at intersection over union of 50=59.7%). This model showed up to 33.6% improved detection precision compared to the same models with no pre-training.ConclusionOptimisation of machine-learning models can be challenging when only relatively small datasets are available. The findings of this study support the potential of transfer learning from large datasets to improve model performance in smaller datasets. This is encouraging for wider application of machine-learning technology in medical imaging evaluation, including less common orthopaedic pathologies.

In the high-voltage power grid of coal mine, the existing adaptive topology analysis algorithm based on the correlation matrix for coal mine high-voltage power grid can effectively identify the network topology and construct a network topology analysis model to provide a basis for subsequent short-circuit current calculation and setting calculation. However, the topology analysis algorithm based on the correlation matrix takes much time. As the scale of the high-voltage power grid network of the coal mine increases, the time consumption will increase sharply as well. However, the prior-to-first-served parallel topology analysis algorithm has been used to reduce the topology analysis time consumption to some extent. On this basis, this article proposes a parallel topology analysis algorithm based on the simulated annealing and optimizes the scheduling strategy of parallel topology analysis. The simulation results show that compared with the prior-to-first-served parallel topology analysis algorithm, the parallel annealing topology analysis algorithm can further reduce the time consumption and improve the computational efficiency. And when the number of available threads increases, its computational efficiency will improve more significantly.